**E**ff**ect of UV Radiation and Salt Stress on the Accumulation of Economically Relevant Secondary Metabolites in Bell Pepper Plants**

**Jan Ellenberger \*, Nils Siefen, Priska Krefting, Jan-Bernd Schulze Lutum, Daniel Pfarr, Maja Remmel, Lukas Schröder and Simone Röhlen-Schmittgen**

INRES Horticultural Sciences, University of Bonn, Auf dem Huegel 6, 53121 Bonn, Germany; nils.siefen@googlemail.com (N.S.); priska-krefting@gmx.de (P.K.); jbsl@uni-bonn.de (J.-B.S.L.); daniel-pfarr@gmx.de (D.P.); s7maremm@uni-bonn.de (M.R.); lukas-schroeder@onlinehom.de (L.S.); s.schmittgen@uni-bonn.de (S.R.-S.)

**\*** Correspondence: ellenberger@uni-bonn.de

Received: 19 December 2019; Accepted: 14 January 2020; Published: 18 January 2020

**Abstract:** The green biomass of horticultural plants contains valuable secondary metabolites (SM), which can potentially be extracted and sold. When exposed to stress, plants accumulate higher amounts of these SMs, making the extraction and commercialization even more attractive. We evaluated the potential for accumulating the flavones cynaroside and graveobioside A in leaves of two bell pepper cultivars (Mavras and Stayer) when exposed to salt stress (100 mM NaCl), UVA/B excitation (UVA 4–5 W/m2; UVB 10–14 W/m2 for 3 h per day), or a combination of both stressors. Plant age during the trials was 32–48 days. HPLC analyses proved the enhanced accumulation of both metabolites under stress conditions. Cynaroside accumulation is effectively triggered by high-UV stress, whereas graveobioside A contents increase under salt stress. Highest contents of secondary metabolites were observed in plants exposed to combined stress. Effects of stress on overall plant performance differed significantly between treatments, with least negative impact on above ground biomass found for high-UV stressed plants. The usage of two non-destructive instruments (Dualex and Multiplex) allowed us to gain insights into the ontogenetical effects at the leaf level and temporal development of SM contents. Indices provided by those devices correlate fairly with amounts detected via HPLC (Cynaroside: r2 = 0.46–0.66; Graveobioside A: r<sup>2</sup> = 0.51–0.71). The concentrations of both metabolites tend to decrease at leaf level during the ontogenetical development even under stress conditions. High-UV stress should be considered as a tool for enriching plant leaves with valuable SM. Effects on the performance of plants throughout a complete production cycle should be evaluated in future trials. All data is available online.

**Keywords:** *Capsicum annuum*; flavonoids; fluorescence monitoring; bio-waste utilization

## **1. Introduction**

#### *1.1. Green Biomass as a Source of Valuable Chemicals*

Commercial vegetable production is accompanied by large quantities of so far under-utilized green biomass in all stages of production and especially after harvest [1]. While the use of biomass for the purpose of energy production is becoming a standard procedure in northern Europe in recent years [2], the extraction and the use of high-value secondary metabolites (SMs) from vegetable plant leaves are just being developed. Research strategies in Europe are heading toward a cascade use of agricultural byproducts and pave the way for extracting and using "valuable substances or molecules before ultimately discarding the left-overs" [3]. The pharmaceutical industry—as an example—is

highly dependent on plant SMs, since approximately 60% of anticancer compounds and 75% of drugs for infectious diseases are derived from plants [4]. In this frame, research on targeted enrichment of valuable substances in plant biomass is gaining importance [5].

#### *1.2. Plant Stress as a Measure to Increase Leaf Secondary Metabolite Content*

The biochemical background of enhanced accumulation of SMs in plant leaves as a measure to cope with stress is a well-described phenomenon [2,6,7]. In short, the cultivation of plants under suboptimal conditions leads to an increased amount of reactive oxygen species (ROS) in plant tissues. Accumulation of SMs is a plant strategy to avoid oxidative damage caused by reactive oxygen species [8]. In theory, both biotic and abiotic stressors could lead to higher amounts of valuable SMs in plants. While biotic stressors such as fungi and insects are hard to control and may cause major phytosanitary problems, abiotic stressors are easier to manage and applicable by practitioners. The results of several studies in recent years indicate that abiotic stressors are a useful tool for SM accumulation in leaves of horticultural plants. Secondary metabolites in *Centella asiatica* leaves increase under enhanced UV-B light, especially in the epidermis [9]. In bell pepper, increased flavonoid contents can be found in leaves exposed to elevated UV [10]. The promoting effect of UV-B radiation on flavonoid accumulation in plant leaves has recently been reviewed [11]. The effects of salt stress on the antioxidant machinery may be adverse and depend on the plant's tolerance [12] and salt concentration in the rootzone [13]. Another extensive study on leaf metabolism in bell pepper under different levels of salt stress revealed an increasing reduction in growth with increasing NaCl contents in the rootzone [14]. While tolerant plants increase leaf SM contents to cope with salt stress, sensitive plants do not have this mechanism and senesce, finally dying off if the stressor is persistent [12]. Studies directly comparing effects of salt and UV stress on leaf SMs are rare. One study shows both stressors to similarly affect leaf contents of the flavonoids quercetin and luteolin in *Ligustrum vulgare* [15]. Abiotic stressors such as drought and salt stress are easily applicable in commercial greenhouse production in soilless systems, which are the predominant systems in many parts of the world, including Europe [16].

#### *1.3. Non-Invasive Monitoring of Secondary Metabolites in Plant Leaves*

Quantification of secondary metabolites including flavonoids with portable optical devices is well established in plant sciences [17]. The use of non-invasive optical sensors to investigate plant leaf components has several advantages over laboratory analyses: data acquisition is faster and more cost effective than laboratory analyses [18]. Moreover, considerate handling of leaves allows for several measurements of the same leaf, enabling to gain insights in temporal developments. Several studies demonstrated the viability of optical devices to access secondary metabolites in plant leaves: a multiparametric fluorescence sensor was used to evaluate the influence of nutrient deficiency on the chemical properties of tomato leaves and to quantify the content of the flavonoids rutin and solanesol [19,20]. In bell pepper, a fluorescence sensor was used to evaluate the impact of priming plants with high light conditions on leaf flavonoid content [10].

#### *1.4. Cynaroside and Graveobioside A*

The vast diversity and chemical complexity of plant SMs often prohibit an economically feasible chemical synthesis. Therefore, extraction either from wild or cultivated plants often represents the best source of supply [1].

Cynaroside (Luteolin-7-glucoside) potentially has a range of medicinal applications: it has the capability to prevent ROS-induced apoptosis in heart cells [21]. Cynaroside furthermore diminishes kidney injury as a side effect of cancer treatments with the chemotherapeutic drug cisplatin. A potential medicinal use of graveobioside A (Luteolin-7-apiosyl-glucoside) is proven by a patent on its application in preparation of drugs for preventing hyperuricemia and gout [22]. Graveobioside A was shown to be contained in several plants, such as celery seeds, parsley, and bell pepper [23,24].

Several SMs in Solanaceae leaves have the potential to biologically control insects [25]. Graveobioside A is such a potential natural insecticide, since oviposition of the American serpentine leafminer fly (*Liriomyza trifolii*) was shown to drop in kidney bean leaves treated with a graveobioside A containing solution [24]. It is expected that the demand for natural insecticides will increase across the EU due to more rigid legislation [26].

We hypothesize that cynaroside and graveobioside A contents in bell pepper leaves can be enhanced by abiotic stressors that are potentially applicable by practitioners in the future. Another aim is to check whether non-invasive devices can be used for assessments of cynaroside and graveobioside A in bell pepper leaves. Furthermore, we attempt to get insights in interactions between different stressors and differences in stress response between two bell pepper cultivars.

#### **2. Material and Methods**

#### *2.1. Plant Material and Growth Conditions*

Seeds of sweet pepper plants (*Capsicum annuum*) cultivar 'Stayer' (Rijk Zwaan, De Lier, The Netherands) and 'Mavras' (Enza Zaden, Enkhuizen, The Netherlands) were sown in soil under greenhouse conditions. Fourteen-days old pepper plants were transplanted into small rockwool cubes (3 × 3 × 5 cm) and one further week later into larger cubes (10 × 10 × 7.5 cm) (Grotop Master, Grodan, The Netherlands). On day 32 after seeding, plants were transferred to a grow chamber to ensure a stable environment. From that day on, stress was applied for 16 days, resulting in a plant age of 48 days at the end of the trial. A longer trial was not feasible due to limitations of the chosen facility. All plants received all nutrients mandatory for optimal growth prepared from two stock solutions (17.2 mM nitrogen, 5.4 mM calcium, 4.7 mM potassium, 0.4 mM phosphorous, 5.4 mM sulfur, 2.4 mM magnesium, 0.01 mM iron and micronutrients; electrical conductivity 2.5 mS cm−1; pH 5.5). Plants were cultivated at the greenhouse facility in Bonn-Endenich (University of Bonn, Bonn, Germany) at day/night temperatures of 24.5 ◦C <sup>±</sup> 5.4 and supplemental light intensity of 203–540 <sup>μ</sup>m m−2s−<sup>1</sup> provided by sodium vapor lamps (Philips Lighting GmbH, Hamburg, Germany).

To apply salt stress, a salt concentration of 100 mM NaCl for a period of 16 days was added to the standard nutrient solution, since that concentration was shown to trigger a higher total phenolic content in leaves of bell pepper seedlings in a previous study [14]. To apply UV stress, plants were exposed to UV light (UVA 4–5 W m−2; UVB 10–14 W m−2) for 3 h per day (Philips Lighting GmbH, Hamburg, Germany) over a 16-day period. In addition, some plants were exposed to combined salt and UV stress. Plant age at stress onset was 32 days. A total of 5 plants per treatment (control, salt stress, UV stress, combined stress) were randomized in the growth chamber.

#### *2.2. Non-Destructive Recordings*

Non-destructive measurements were performed on all leaves per plant, from mature to young. Measurements were conducted using two well-established devices in stress physiology monitoring. First device is the multiparametric fluorescence excitation system Multiplex® (Multiplex®, Force-A, Orsay, France), described in previous studies [27]. All recordings with this device were done at a constant distance of 0.10 m to the leaf surface and a frontal cover plate with an aperture of 4 cm in diameter opening to assess the index of epidermal flavonols (FLAV index): log *FRF*\_*<sup>R</sup> FRF*\_*UV* .

Secondly, the transmittance-based fluorescence measurements were conducted with the Dualex sensor (Force-A, Orsay, France). The Dualex is a device with a leaf-clip; measurements were taken with virtually no distance to the leaf surface. The device is extensively described in the literature [28,29].

#### *2.3. Plant Harvest*

Plants were harvested 16 days after treatment inception (DATI) at a plant age of 48 days. The total fresh weight of shoots was determined immediately. Leaves were dried for 7 days at 50 ◦C (drying oven) to collect dry weights.

#### *2.4. Leaf Sample Preparation and Laboratory Analysis*

Samples were taken at the harvesting at 16 DATI, of the mature leaf 4 and the young leaves 10 and 12, to assess the impact of stress application on the amount of the two luteolins, graveobioside A and cynaroside. All leaf numbers are given as the number of true leaves, counted from the base of the plant. The samples were freeze-dried and then stored at −20 ◦C until further processing. Ground leaf samples were prepared for HPLC determination (Agilent 1260 Infinity HPLC System Agilent Technology Deutschland GmbH, Ratingen, Germany). An amount of 0.3 g was extracted with water-diluted methanol (60:40, *v*/*v*) for 10 min in an ultrasonic bath, centrifuged for 10 min at 4 ◦C with 13,000 rpm (Centrifuge 5415R, Eppendorf AG, Hamburg, Deutschland) repeated four times. The supernatants were collected and stored at −20 ◦C until HPLC analysis. The samples were filtrated through a membrane filter (Phenomenex, Aschaffenburg, Germany) prior to injection. The HPLC system consisted of an autosampler, a diode array UV–Vis detector and was coupled with a quaternary solvent delivery system. The column (Nocleodur C18, 3 × 150 mm, 3 μm, Macherey-Nagel, GmbH & Co. KG, Düren, Germany) was isocratically eluted with a binary mixture of water and methanol (60:40) adjusted to pH 2.8 with phosphoric acid. The flow rate was 0.3 mL min<sup>−</sup>1; 10 μL samples were injected onto the column equilibrated at 25 ◦C (detection at 355 nm). Graveobioside A peak was detected at 14.1 min, and cynaroside at 15.6 min. Both calibration curves were obtained from diluted series of standards provided by PhytoLab (Vestenbergsgreuth, Germany).

#### *2.5. Data Analysis and Statistics*

All data is available online [30]. Data analysis was performed with R (R Core Team, Vienna, Austria) [31] in RStudio (R Studio Team, Boston, USA) [32]. According to the data structure, e.g., balanced or imbalanced, type I or type III ANOVA were used to compare group means. Applied post-hoc test was Tukey's HSD. Figures were created in RStudio, with the package ggplot2 [33].

#### **3. Results**

#### *3.1. Stress-Related E*ff*ect Varies Among Secondary Metabolites and Cultivars*

A treatment effect was observed on contents of both cynaroside and graveobioside A, while no significant effect of the variable cultivar on either metabolite content was found. There was a strong tendency for higher graveobioside A in 'Mavras' as compared to Stayer (*p* = 0.055). No interactions between cultivar and treatment were observed (Table 1). Both combined-stressed plants and plants under UV-exposure accumulated significantly higher amounts of cynaroside in their leaves than control and salt-stressed plants (Figure 1, A + B). Plants of the cultivar 'Mavras' accumulated significantly higher graveobioside A amounts in salt-stressed and combined-stressed plants than in control and UV-stressed plants (Figure 1C). No significant treatment effect on graveobioside A content in plants of the cultivar Stayer was found (Figure 1D). Levels of SM in leaves of different ontogenetical stages are shown as an illustration of uneven distribution within the plants. SM contents decrease with leaf ontogenetical stage (Figure 1).

**Table 1.** Interaction and main effect for treatments (control, salt-stress, combined-stress, UV-stress) and cultivars (Mavras and Stayer), calculated with a type I two-way ANOVA. Grayed area indicates significant effect (*p* ≤ 0.001).


**Figure 1.** HPLC-determined leaf cynaroside (**A**,**B**) and graveobioside A (**C**,**D**) contents, for bell pepper cultivars 'Mavras' (**A**,**C**) and 'Stayer' (**B**,**D**) under different growth conditions, 15 days after treatment inception (*n* = 5). Transparent boxplots show pooled data from all leaves (*n* = 15). Colored boxplots represent leaf age—subgroups (Leaf 4, 10, and 12 as counted from the base, with darkest colors for youngest leaves). Letters (a,b) indicate differences within each cultivar × secondary metabolite—combination (Tukey HSD, *p* < 0.05).

Both fresh and dry weight of bell pepper plants differed significantly depending on the cultivar, with Stayer attaining higher weights than Mavras. Treatment had a significant effect on both fresh and dry weight. There was no interaction between the treatment and cultivar regarding plant's fresh or dry weight. Dry weight of plants of the cultivar Mavras was significantly higher in control plants than in any other treatment (Table 1). UV-stressed plants of both tested cultivars exhibited higher fresh and dry weights than plants under salt-stress and combined-stress conditions (Figure 2). Observed mean fresh weight decreased in salt-stress and combined-stress plants compared to control and UV stress, which were in the magnitude of 50% (Figure 2C,D). The mean dry weight tended to be higher for salt-stressed plants as compared to plants under combined stress, but lower than the dry weights of plants experiencing UV stress or control conditions (Figure 2A,B).

**Figure 2.** Aboveground biomass (dry weight: (**A**), (**B**); fresh weight: (**C**), (**D**)) of bell pepper cultivars "Mavras" (**A**), (**C**) and "Stayer" (**B**), (**D**) under different growth conditions, 15 days after treatment induction (*n* = 5). Letters (a,b) indicate differences within each cultivar × dry/fresh weight—combination (Tukey HSD, *p* < 0.05).

## *3.2. Non-Invasive Monitoring of Secondary Metabolites*

Figure 3 shows exponential regressions between three indices (Multiplex indices FLAV and NBI\_R; Figure 3A–D and Dualex index Flav; Figure 3E,F) and leaf contents of the SMs cynaroside (Figure 3A,C,E) and graveobioside A (Figure 3B,D,F), respectively. Predictions of graveobioside A contents based on the indices are better than predictions of cynaroside contents. Multiplex indices are more accurate predictors than the Dualex index, as outlined by correlation coefficients (r2). Index values level off at cynaroside contents above 1.5 mg g−1. The connection between graveobioside A and the indices is more linear, but still leveling off at graveobioside A contents above approximately 25 mg g<sup>−</sup>1.

**Figure 3.** Exponential regression between indices of non-invasive devices and leaf secondary metabolite concentrations in bell pepper leaves, determined via HPLC. Contents of cynaroside and graveobioside A correlated with FLAV (Mx) (**A**), (**B**), NBI\_R (Mx) (**C**), (**D**), and Flav (Dx) (**E**), (**F**). Color of points represents leaf age (Leaf 4, 10, and 12 as counted from the base, with darkest colors for youngest leaves). Lines indicate exponential regressions (*n* = 60). RSS, residual sum of squares.

#### *3.3. Spatial and Temporal Development of Secondary Metabolite Contents*

The only significant changes in FLAV values within cultivar × treatment groups were seen among the fourth leaves of combined-stressed Mavras plants at days 0 versus 9 and 0 versus 15, respectively (Figure 4C). A clear trend was observed for the fourth leaves of combined-stressed Stayer plants at days 0 versus 15 (TukeyHSD, *p* = 0.053) (Figure 4D). Generally, FLAV values for stressed plants tend to increase, while the values for control leaves tend to decrease. A comprehensive overview of associated main effects is given in Table 2.

**Figure 4.** Temporal development of secondary metabolites in leaves of bell pepper cultivars "Mavras" and "Stayer", expressed with the FLAV-index (Multiplex). (**C**), (**D**), *n* = 5; (**A**), (**B**), *n* = 5–50; DATI, day after treatment initiation.

**Table 2.** Interaction and main effect for treatments (control, salt-stress, combined-stress, UV-stress) and DATI (0, 2, 7, 9, 15). To account for the unbalanced design (e.g., unequal numbers of observations within each level of DATI), type III ANOVA was selected to compare differences between factor means for FLAV values of "All leaves". Grayed area indicates significant effect at *p* ≤ 0.05 (light), *p* ≤ 0.01 (medium), and *p* ≤ 0.001 (dark).


#### **4. Discussion**

We are among the first groups accessing the amount of graveobioside A in pepper leaves [4]. For cynaroside, the range of values detected corresponds to the results of other studies [34,35].

#### *4.1. Stress-Related E*ff*ect Varies According to Secondary Metabolites and Cultivars*

Since cynaroside contents under single UV-stress and combined UV- and salt-stress are not significantly different (Figure 1A,B), cynaroside accumulation appears to be triggered mainly by high radiation conditions. Interestingly, and in contrast to cynaroside, graveobioside A accumulation is triggered more effectively by salt stress than by UV-stress, especially in the cultivar Mavras (Figure 1C). This is a surprising result, since biosynthesis of flavonoids is said to be enhanced similarly by UV radiation and salinity [15,36]. On the other hand, some authors report that the regulation of SM production in response to salt stress differs between salt-sensitive (upregulation) and salt-tolerant (downregulation) plants [12]. However, differences in salt-stress tolerance between the cultivars used in this study are not supported by differing plant biomasses (Figure 2). The chemical group of

flavonoids is highly diverse, and metabolic pathways are not entirely understood to date. At this point, it remains unclear how exactly upregulation of cynaroside synthesis under UV stress and upregulation of graveobioside A synthesis under salt stress occurs.

Our results indicate—as expected—that salt-stressed plants acquire a significantly lower biomass than both control plants and UV-stressed plants. Stunted growth is a well-described symptom of severe salt stress in plants [12,37]. If the applied salt concentration would have been lower, negative effects could probably have been avoided to a certain extent, as recently discussed in a review on the potential of seawater use in soilless culture [13]. Reaction of plants to UV-B exposure varies from growth reduction to enhancement, depending on species, cultivar, and stress level [11,38]. Since the overall aim of the stress application is the accumulation of higher amounts of secondary metabolites in the plant's green biomass, it is necessary to consider not only the share of desired metabolite in the plant´s biomass, but also the biomass reduction caused by the treatment. Considering this background, we can state that stressors with minor negative effects on plant biomass accumulation, but major positive effects on contents of desired metabolites in the plant tissues, are necessary to achieve these aims. Finding the perfect trade-off between biomass and fruit yield loss, on the one hand, and SM increase, on the other hand, will be crucial to improve the production system. In our specific setup with two single stressors and one combined stress, with respective levels of stress described above, the single UV stress is most promising, whereas salt stress (100 mM NaCl), although promoting the accumulation of graveobioside A, is less promising as a tool to enhance whole plant SM amounts, due to the decrease in total biomass. Effects on plants grown over a whole season are a matter of ongoing research.

#### *4.2. Non-Invasive Monitoring*

The indices provided by both optical devices deliver better estimates for leaf graveobioside A contents than for leaf cynaroside contents. That is an expected result, since the amount of graveobioside A as determined via HPLC is up to ten-fold higher than the amount of cynaroside (0–4 versus 2–40 mg g<sup>−</sup>1) and both secondary metabolites share similar optical properties. Any estimate of concentrations based on non-invasive, optical devices will be best for the predominant fraction of a group of metabolites with similar optical properties. By the same token, signals of metabolites that occur in small quantities are more likely to be superimposed by other signals and therefore difficult to quantify. Additional factors known to influence non-invasive assessment of leaf compounds include the concentration of other pigments potentially influencing the measurement [39], leaf thickness [40], and the device used [41].

In our study, the FLAV-index of the Multiplex shows an almost linear response to changes in leaf graveobioside A content (Figure 2B). The same applies for the NBI\_R index, which correlates negatively with the actual graveobioside A content. Both indices use the far-red fluorescence of leaves excited with UV-light and normalize that signal for the red fluorescence emitted after excitation with red light [29]. As an enhanced graveobioside A content leads to a stronger absorption of UV light in the leaf epidermis, less radiation penetrates into the mesophyll, which in turn leads to a lower chlorophyll fluorescence. We have to highlight the broad distribution of fluorescence values, though, which prohibits a precise prediction of actual graveobioside A levels on the individual leaf level. The Flav-index of the Dualex is almost indifferent to changes at graveobioside A levels above 25 mg g<sup>−</sup>1.

None of the indices is strongly related to the leaf cynaroside contents quantified by HPLC. Neither the Dualex nor the Multiplex provide any indices that allow to quantify cynaroside contents higher than approximately 1 mg g−<sup>1</sup> dry weight. An exact evaluation of high levels of this specific SM in bell pepper leaves is therefore not possible with the tested devices. However, the correlations we have identified between the FLAV index and HPLC measurements still allow us to analyze the gradual changes in SM contents as they occur during the prolonged period of stress.

#### *4.3. Insights in Spatial and Temporal Accumulation of Secondary Metabolites*

The usage of non-invasive phenotyping tools such as the Multiplex and Dualex devices allows to analyze leaf constituents during ontogenesis. The observed drop of the flavonol content in leaves of unstressed plants during ontogenesis (Figure 4C,D) is in line with the theories that (a) the production of phenolics, such as flavonols, is mainly caused by photodamage [42] and (b) that ontogenetically young leaves are, in general, more prone to be affected by high light stress than older leaves, since their photosynthetic apparatus is not yet well developed [43] and the photoprotective cuticula is thinner compared with older leaves [44]. Therefore, young leaves show stress-related reactions in conditions that are neither stressful for older leaves nor for the entire plant. However, the described ontogenetic effects tend to be overcompensated by stress-related effects in all three stress treatments (Figure 4C,D). Thus, flavonol contents of the fourth leaf as measured with the FLAV (Mx) index slightly increased in plants experiencing single stresses, while plants exposed to combined stress showed major increases in leaf flavonol contents (Figure 4C,D).

#### *4.4. Implications and Future Challenges*

The present study proves that abiotic stresses, in particular, salt stress and UV stress, can enhance the amount of economically valuable SMs, namely cynaroside and graveobioside A, in bell pepper leaves. The main objective of growing bell pepper plants, however, is the production of fruits of adequate quantity and quality for human nutrition. Considering the decline in plant biomass in response to stress conditions, it is very likely that the stressors applied would also lead to a reduction in fruit production. Severe salt stress, in particular, is known to be an important factor limiting crop productivity [45]. We have shown that the type of stressor has magnificent effects on both plant biomass and leaf secondary metabolite content. Other studies have proven that this also applies for different levels of abiotic stress [14,46]. The search for the best stressors and stress levels for the accumulation of secondary metabolites in plant leaves with negligible effects on fruit yield is a major future challenge for research in stress physiology. Several authors reported neutral or positive responses of product quality to mild stress [46]. For salt stress, several studies in the model-crop tomato reveal positive impacts of mild stress on fruit quality (e.g., antioxidant capacity and nutritional value) [47,48]. Low UV radiation reduces the antioxidative capacity and, therefore, the fruit quality of bell pepper fruits [49]. Additional UV radiation may help to overcome this problem and, at the same time, induce the production of valuable SM in the leaves. Cultivation of plants under mild water stress conditions can also enhance water use efficiency. To avoid any competition with food production, post-harvest treatment of leaves could be an appropriate measure to achieve high contents of promising metabolites [50,51]. These effects should also be taken into account when evaluating the value of production systems that are based on commercialization of both fruits and SMs in leaves of horticultural plants.

To enhance precision of non-invasive estimation of SMs in pepper leaves, future studies should consider hyperspectral sensors as well as chlorophyll fluorescence-based sensors, ideally a combination of both. Sensors covering the UV range are just entering the market and appear as a promising tool to access SMs in plants, as they cover absorption bands of flavones and other phenolic leaf compounds [52].

#### **5. Conclusions**

Both additional UV light and salt stress can enhance concentrations of the two SMs graveobioside A and cynaroside in bell pepper leaves. Highest concentrations were reached by combining both treatments. Stressed bell pepper leaves contain up to 30 mg graveobioside A and about 2 mg cynaroside per gram dry weight. While salt stress (100 mM NaCl) has a major negative impact on plant vegetative growth, UV stress (UVA 4–5 W m<sup>−</sup>2; UVB 10–14 W m−2; 3 h per day) has no significant impact on the fresh mass of the plants. The tendency of decreasing SM contents in leaves during ontogenesis is outweighed by the stress treatments. Graveobioside A contents can be assessed with the multiparametric fluorescence sensor Multiplex. Reliable quantification of cynaroside is not possible with the non-invasive sensors used. If future experiments exclude major negative impacts on fruit quality, UV stress can be recommended as one tool to enhance valuable SMs in bell pepper leaves and potentially in vegetable leaves in general. A less-intense salt stress should also be considered in future experiments.

**Author Contributions:** Conceptualization, S.R.-S.; data analysis, J.E., N.S., P.K., J.-B.S.L., D.P., M.R., and L.S.; methodology, S.R.-S.; writing—original draft preparation, J.E.; writing—review and editing, J.E., S.R.-S., N.S., P.K., J.-B.S.L., D.P., M.R., and L.S. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the German Federal Ministry of Education and Research (grant number: 031B0361C).

**Acknowledgments:** The authors are grateful to Libeth Schwager for her support in laboratory analysis, and for plant cultivation by the staff members of the plant service team "Dienstleistungsplattform" of the University of Bonn. We thank Eduardo Fernandez for discussions and support in data visualization. We appreciate the support of Katharina Krah, Simone Klein, Miriam Brink, and Mark Schmutzler during the measurements. We are thankful for the great support by our project partners Manuel Lück, Anika Wiese-Klinkenberg, Julia J. Reimer, and Alexandra Wormit.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **Abbreviations**


## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## **Promising Composts as Growing Media for the Production of Baby Leaf Lettuce in a Floating System**

**Almudena Giménez 1, Juan A. Fernández 1,2,\*, José A. Pascual 3, Margarita Ros 3, José Saez-Tovar 4, Encarnación Martinez-Sabater 4, Nazim S. Gruda <sup>5</sup> and Catalina Egea-Gilabert 1,2**


Received: 9 September 2020; Accepted: 5 October 2020; Published: 10 October 2020

**Abstract:** The floating system is a successful strategy for producing baby leaf vegetables. Moreover, compost from agricultural and agri-food industry wastes is an alternative to peat that can be used as a component of growing media in this cultivation system. In this study, we experimented with three composts containing tomato (*Solanum lycopersicum* L.), leek (*Allium porrum* L.), grape (*Vitis vinifera* L.), and/or olive (*Olea europaea* L.) mill cake residues, which were used as the main component (75/25 volume/volume) of three growing media (GM1, GM2 and GM3) to evaluate their effect on the growth and quality of red baby leaf lettuce (*Lactuca sativa* L.). We used a commercial peat substrate as a control treatment (100% volume) and in mixtures (25% volume) with the composts. The plants were cultivated over two growing cycles, in spring and summer, and harvested twice in each cycle when the plants had four to five leaves. We found that the percentage of seed germination was significantly higher in plants grown in peat than in those grown in compost growing media. The yield was affected by the growing media in the summer cycle, and we obtained the highest value with GM1. Furthermore, the second cut was more productive than the first one for all the growing media in both cycles. The lettuce quality was also affected by the growing media. In general, the total phenolic content and antioxidant capacity in the leaves was higher in plants grown in the compost growing media, particularly in the second cut, but the nitrate content in the leaves was greater in some of the compost treatments compared with the peat treatment. In addition, an in vitro suppressive activity study demonstrated that the interaction between different fungi and bacteria observed through metagenomics analysis could contribute to the effectiveness of the compost in controlling *Pythium irregulare*. The use of compost as a component of the growing media in the production of baby leaf vegetables in a floating system does not only favor the crop yield and product quality, but also shows suppressive effects against *P. irregulare*.

**Keywords:** germination; nitrate content; phenolic content; antioxidant capacity; microbial community

## **1. Introduction**

Nowadays, there is a high demand among consumers for ready-to-eat vegetables due to a growing interest in healthy, fresh convenience foods. Demand for baby leaf vegetables has especially increased [1]. Baby leaf vegetables come in a wide variety of textures, colors and flavors, which makes them very attractive for consumption. Lettuce is considered to be a health-beneficial food due to the high concentrations of vitamins, minerals, dietary fiber, and antioxidant compounds different lettuces contain [2]. A wide range of varieties can be used for baby leaf production.

Among the hydroponic methods used to produce baby-leaf vegetables, the floating system is a successful strategy for producing baby leaf vegetables, which consists of trays floating on a waterbed or hydroponic nutrient solution, which can be operated as a closed system [1], resulting in a more environmentally friendly crop production strategy (Nicola et al., 2016) [3]. Among other reasons for their use, floating systems make it possible to obtain clean and safe products for the processing industry and to reduce crop cycle duration with respect to soil culture [4]. In addition, some baby leaf crops can be harvested more than once, if regrowth is allowed. With this latter approach, the time to harvest is shortened, resulting in a lesser environmental impact and a reduction in the economic cost [5].

Peat is the usual substrate used to fill the holes in the trays used for growing baby leaf vegetables in floating systems [6]. Nevertheless, peat increases susceptibility to some diseases, such as damping off, which is caused by fungi or oomycetes like *Pythium* spp., which can lead to significant production losses [7]. Moreover, peat comes from peatland ecosystems, and harvesting peat despoils ecologically important peat bog areas [8]; degraded peatlands negatively and disproportionally contribute to released stored carbon and an increase in global greenhouse gas emissions, affecting the environment and CO2 balance [9,10]. The search for organic materials that can be used as peat alternatives has become increasingly important.

Compost from agricultural and from agri-food industry waste can be an alternative to peat in soilless culture systems. Furthermore, compost can control different plant pathogens, like *Fusarium* sp. [11] and *Pythium irregulare*, and improve the yield and quality of the final product [12]. In addition, compost use is an environmentally friendly practice in light of the circular economy [13]. Depending on its composition, compost made with so-called green materials, such as pruning waste, can in exceptional cases be used directly as a standalone substrate, but it is usually used as a growing media constituent [8,14]. The main limitations to the use of composts in growing media are their physical properties, salinity, high pH, and rate of residual degradation over time [15].

Compost can also be considered an important resource for the biofertilization and bio-stimulation of crops. During composting, organic matter is decomposed and transformed by microorganisms after a polymerization process to form humic substances [16], which have a very important effect on improving soil fertility, because they are rich in mature organic matter. Besides humic substances, the hormone-like molecules secreted by microbes and nutritional elements are compost components that may play a crucial role in the bio-stimulation of plants [17]. The compost microbiome composition plays an important role in the complex relationships that occur in the rhizosphere [18]. High-throughput sequencing technologies have provided an important way to determine compost microbiome information [19], rather than the isolation and identification of microorganism species.

The objective of this study was to characterize three composts from agro-industrial wastes and evaluate their impact as a growing media component on the yield and quality of a red baby leaf lettuce crop growing in a floating system. Our hypothesis was that composts could provide a biostimulant and biofertilizing effect on baby leaf lettuce in addition to its suppressive activity against *P. irregulare*.

#### **2. Materials and Methods**

#### *2.1. Compost Characterisation*

Three types of compost produced at the University Miguel Hernandez composting site were used for the experiment. In the compost feedstocks, the following raw materials were used: vineyard pruning, tomato and leek processing by-products, and olive mill cake. Their proportion in the composts is described in Table 1. The composting process for the three composts lasted 210 days and consisted of a composting phase with a mesophilic and thermophilic phase of 166 days and a maturation phase

of 44 days. The temperatures reached were >60 ◦C. The physical properties of the compost and peat (bulk density, total pore space, and total water holding capacity) were measured as described by Bustamante et al. [20] pH and electrical conductivity (EC) were measured in a water-soluble extract 1:10 (*w*/*v*) using a conductivity/pH meter (Crison). The total organic carbon (TOC) and total nitrogen (N) were measured using a LECO TruSpec C/N Elemental Analyzer. P, K, Ca, Mg, B, Fe, Mn, Mo, and Zn were determined by inductively coupled plasma-mass spectrophotometry (ICP-MS PQExCell, VG-Thermo Elemental, Winsford, Cheshire, UK), after HNO3/HClO4 high pressure digestion. Organic N, nitrate, and ammonium N were determined following the McKenzie and Young [21] method. Available phosphorus was extracted with ammonium citrate pH 7 and it colorimetrically determined on the extracts according to Watanabe and Olsen [22]. Available K was extracted with ammonium acetate pH 7 and later filtered through whatman 0.22 mm2; it was determined by inductively coupled plasma-mass spectrophotometry, as the rest of the above-measured elements. All the analyses were performed in triplicate. Available humic acids were measured according to Sanchez-Monedero et al. [23]. For the biological characteristics, the bacterial and fungal colony forming units (CFUs) were counted after plating different tenfold serial dilutions of water extract from the composts/peat in Trypto-casein Soy Agar (TSA) plus cycloheximide (100 mg mL<sup>−</sup>1) and potato dextrose agar (PDA) plus streptomycin (50 mg mL<sup>−</sup>1), respectively. The Petri plates were incubated at 28 ◦C, and a standard plate count (SPC) was performed to determine the number of colonies of bacteria and fungi grown on the respective media after 7 and 5 days, respectively. The CFUs were counted and the values were multiplied by the dilution factor and expressed in log CFU g−<sup>1</sup> of dry compost. Finally, dehydrogenase activity (DHA) was determined according to García et al. [24]

**Table 1.** Composition of composts in percentage of dry matter.


C1, C2, and C3 represent the composts used.

The main physical, chemical, and biological characteristics of the composts are shown in Table 2.




**Table 2.** *Cont.*

Values are the mean ± SD (*n* = 3). Asterisk indicates significances at \*\*\* *p* < 0.001; n.s: non-significant. Different letters indicate significant differences. C1, C2, and C3 represent the compost used. BD: Bulk density; TPS: total pore space; AC: air capacity; WHC: water holding capacity; EC: electrical conductivity; TOC: total organic carbon; HA: humic acids; CFUs: colony formed units; DHA: dehydrogenase activity; INTF: p-iodonitrotetrazolium formazan.

#### *2.2. Compost Microbial Community*

Total DNA was extracted from 500-mg compost samples using the DNeasy PowerSoil Kit (Qiagen, Hilden, Germany), following the modification described by Taskin et al. [25] For bacteria, the V4 region of bacterial 16S rDNA was amplified using the barcoded primers 515F and 806R [26]. For fungi, the ITS2 region was amplified with the primer pair gITS7/ITS4 [27]. Each sample was amplified in triplicate as described previously by Žifˇcáková et al. [28] Amplicons were purified using the QIAquick PCR Purification Kit (Qiagen, Hilden, Germany), and the DNA concentration was measured by Qubit (Thermo Fisher, Waltham, MA, USA). A TruSeq PCR-Free kit was used for library preparation. Sequencing of the bacterial and fungal communities was performed on Illumina MiSeq, and the sequences were generated with the MiSeq Reagent Kit v2 on a paired-end mode with sizes of 251 base pairs (Institute of Microbiology of the CAS, Prague, Czech Republic).

#### *2.3. In Vitro Suppressiveness against Pythium Irregulare*

*Pytium irregulare*isolate, originally recovered from over-used floating trays where baby-leaf lettuces were grown, was selected from the pathogen culture collection of CEBAS-CSIC. The pathogenicity of the isolate was tested every three months, by passing it through baby-leaf plants and re-isolating it again to assure that pathogenicity was not lost due over culture in petri dishes. The mycelial growth of *P. irregulare* was estimated on potato dextrose agar plates (PDA). An 8-mm agar disk of *P. irregulare* was placed on the edge of one side of the plate, and 0.5 mL of dilution 10−<sup>4</sup> of each compost and peat water extract was spread over the PDA surface on the other side of the plate. As a control, 0.5 mL of sterile water was spread on a PDA surface. Three replicates were performed per treatment. The plates were incubated in the dark at 28 ◦C, and the radial growth of the pathogen was measured every 24 h for 7 days. Growth inhibition was expressed by Mycelia Growth Inhibition, MGI (%) [29].

MGI% = ((RGcontrol − RGcompost)/RGcontrol) × 100; RGcontrol = radial growth of pathogen in control plates; RGcompost = radial growth of pathogen in plates with compost.

#### *2.4. Experimental Conditions*

The experiments were conducted at the 'Tomás Ferro' Experimental Agro Food Station of the Technical University of Cartagena (UPCT; lat. 37\_410 N; long. 0\_570 W). A cultivar of red baby leaf lettuce (*Lactuca sativa* L., cv. 'Ligier') from Rijk Zwaan, De Lier, the Netherlands, was cultivated in a floating system in an unheated greenhouse covered with thermal polyethylene. In the greenhouse, the light conditions during the experiments were an average daily light integral (DLI) of 14.07 mol m−<sup>2</sup> d<sup>−</sup>1; the minimum, maximum, and average air temperatures, in the spring cycle were 8.10 ◦C, 39.10 ◦C, and 19.41 ◦C, respectively; in the summer cycle, the average DLI was 15.48 mol m−<sup>2</sup> d−<sup>1</sup> and the minimum, maximum, and average air temperatures were 14.02 ◦C, 44.15 °C and 28.69 ◦C, respectively. Two crop cycles were carried out with sowings on 29 March 2019 (spring) and 14 June 2019 (summer).

Seeds were sown in 60 × 40-cm styrofloat trays [30] filled with the three compost-growing media (GM1, GM2, GM3) composed using each compost (C1, C2, C3) mixed with commercial peat (75:25, *v*/*v*). A commercial peat 315 (Blond/black 60/40 Turbas y Coco Mar Menor S.L.) was used as a control. The main chemical characteristics of the peat were as follows: pH 5.6; EC 1 dS m<sup>−</sup>1; total C 466.8 g kg<sup>−</sup>1; total N 9.4 g kg−1; total P 0.3 g kg−1; and total K 0.9 g kg−1. After sowing, the trays were placed in a climatic chamber at 18 ◦C and 90% relative humidity and left in the dark for 48 h to improve germination. After seedling emergence, the trays were transferred to flotation beds (1.35 × 1.25 × 0.2 m). Each level of treatment (peat, GM1, GM2, and GM3) was carried out in beds randomly located at three places inside the greenhouse described above, in both growing seasons; each bed had three floating trays of 60 cm <sup>×</sup> 41 cm. The trays were floating on tap water with an EC of 1.1 dS m−<sup>1</sup> and pH 7.8. Aeration was provided using a blow pump connected to a perforated pipe trellis positioned at the bottom of each flotation bed.

A week after sowing, the lettuce plants were thinned, and 10 plants were left per cell (2000 plants m−2). At the same time, the tap water in the beds was replaced with the nutrient solution [31]. The nutrient solution was adjusted to EC 2.5 dS m−<sup>1</sup> and pH 5.8. The EC and temperature of the nutrient solution were monitored throughout the growing cycles using Campbell CS547 sensors (Campbell Scientific In. Logan, UT) with an average of 2.76 dS m−<sup>1</sup> and 19.53 ◦C in spring and 2.76 dS m−<sup>1</sup> and 29.66 ◦C in summer, respectively. The oxygen concentrations were monitored using Campbell CS512 sensors located in each flotation bed with an average of 7.09 mg L−<sup>1</sup> and 6.97 mg L−<sup>1</sup> in the spring and summer cycles, respectively.

Harvesting was carried out twice per cycle at the same phenological stage for both cycles, when the plants had four to five leaves. The plants were harvested on April 25 (1st cut) and May 6 (2nd cut) in the spring cycle and on July 5 (1st cut) and July 12 (2nd cut) in the summer cycle. For each growing media, 90 plants from three cells fissure were randomly chosen from each tray for harvest; they were then stored at −80 ◦C for analysis.

#### *2.5. Germination*

To calculate the germination percentage, we used nine trays, i.e., three trays per growing media. Twenty baby leaf lettuce seeds were sown in each fissure on the tray, with 154 fissures per replication. After two days in a germination chamber at 18 ◦C and 90% relative humidity, the trays were transferred to flotation beds randomly placed on three stainless steel beds located in the greenhouse described above with tap water for five days (7 days after sowing (das)), with temperature conditions of 10.17 ◦C, 31.39 ◦C, and 17.63 ◦C as minimum, maximum, and average air temperature, respectively. Then, the percentage of seed germination with respect to the total seeds sown was calculated.

#### *2.6. Plant Analysis at Harvesting*

At harvesting time, the following parameters were analyzed in both cycles: biomass production (yield), calculated as g of fresh mass plant<sup>−</sup>1; nitrate content in leaves and in the nutrient solution; and the total phenolic content and antioxidant capacity in the leaves.

The nitrate content was determined by ion chromatography following Lara et al. [32]

The total phenolic content (TPC) was determined by the Folin-Ciocalteu colorimetric method, as previously described by Singleton and Rossi [33], with modifications previously reported by Martínez-Hernández et al. [34]

The total antioxidant capacity (TAC)was determined as described by Klug et al. [35], using three different approaches: via the free radical scavenging capacity with 2,2–diphenyl–1–picrylhydrazil (DPPH) [36]; the ferric-reducing antioxidant power (FRAP) [37]; and 2 2 –azino–bis (3–ethylbenzothiazoline– 6–sulfonic acid) (ABTS) [38]. The DPPH method was conducted by measuring the decrease in absorbance at 515 nm for 30 min. The TAC extract (21 μL) was mixed with a volume (194 μL) of DPPH solution (≈ 0.8 mM and adjusted to Abs515 = 1.1 ± 0.02) and allowed to react for 30 min. The ABTS method was conducted by measuring the absorbance increase at 734 nm for 30 min. A volume (200 μL) of ABTS solution (14 mM ABTS<sup>+</sup> and 4.9 mM K2S2O8 by 1:1 (*v*/*v*)) was added to each extract sample (11 μL) and allowed to react for 30 min. The FRAP method was conducted by measuring the increase in absorbance at 593 nm for 60 min. The freshly made FRAP solution (prepared at a ratio of 10:1:1 (*v*/*v*/*v*) using sodium acetate buffer, pH 3.6; 10 mM TPTZ solution in 40 mM HCl; and 20 mM FeCl3, respectively, and preincubated at 37 ◦C for 2 h) was added (198 μL) to the extract (6 μL) and allowed to react for 60 min. All TAC reactions were conducted at room temperature in darkness, and absorbance was measured using the same microplate reader that was used for TPC. TAC was expressed as mg of Trolox equivalent per 100 g DW of lettuce leaves, as the mean of three replicates per each treatment and cut.

### *2.7. Statistical Analysis*

Data were analyzed using Statgraphics Plus. To determine the compost characteristics, we performed an analysis of variance of measured parameters (one-way ANOVA). For the greenhouse experiment, we performed an analysis of variance of the measured parameters (two-way ANOVA), in which the growing media (peat, GM1, GM2, GM3) and time of cutting (1st cut and 2nd cut) were included for each crop cycle. When the interaction between factors was significant, ANOVA was carried out for each factor independently.

The amplicon sequencing data were processed using the SEED 2 program [39,40]. Pair-end reads were merged using fastq-join [41], and whole amplicons were processed. Chimeric sequences were detected using Usearch 7.0.1090 [42] and removed. Non-chimeric sequences were clustered to 97% similarity using UPARSE implemented within Usearch [43]. Consensus sequences were constructed for each cluster, and the closest hits at the genus level were identified using BLASTn against the GenBank databases for both bacteria and fungi [44]. Sequences identified as non-bacterial or non-fungal were excluded from subsequent analyses.

#### **3. Results**

### *3.1. Compost Characterisation*

The three composts showed a similar BD (ca. 0.20 g cm<sup>−</sup>3), but it was significantly lower than in peat (Table 2). There were no significant differences between treatments for TPS. Composts C2 and C3 showed a significantly higher AC (more than 30 vol %) than C1 and peat (ca. 20 vol %). However, the same two composts and peat showed a significantly lower WHC than C1 678 mL L<sup>−</sup>1. The three composts showed a basic pH higher than 8.0, significantly higher to peat pH (5.6). The EC values of the three composts were significantly higher than peat, with compost C1 showing the highest EC values, followed by C3 and C2 (Table 2). Composts C2 and C3 showed significantly higher TOC than compost C1 and peat (Table 2). HA was significantly higher in peat with respect to the composts, C2 and C3 also being significantly higher than C1. Composts showed significantly higher values in total and available N, P, and K. In general, peat had lower values with respect to composts for Ca, Mg, Cu, Mn, and Zn. Finally, there were no significant differences between treatments for Mo, Fe, and B.

The fungal and bacterial content (CFUs) of composts C2 and C3 was significantly higher than in C1 and peat (Table 2). The total bacteria content (CFUs) and DHA activity of composts was significantly higher than in peat, CFUs values in C2 and C3 being significantly higher than in C1. Nevertheless, no significant differences were observed between the three composts in terms of DHA activity (Table 2).

#### *3.2. Compost Microbial Community*

The Shannon and Simpson diversity index for bacteria did not show significant differences among composts, while for fungi, the compost C2 showed the highest diversity indexes followed by C3 and C1 (Table 3). The coverage value was estimated to be >99% and did not significantly differ between composts.


**Table 3.** The Shannon and Simpson diversity index for compost bacteria and fungi.

Values are the mean ± SD (*n* = 3). Asterisk indicates significances at \*\*\* *p* < 0.001; n.s: non-significant. Different letters indicate significant differences.

The dominant bacteria and fungi genera are shown in Figure 1. We identified different bacterial genera belonging to phyla Proteobateria (*Pseudomonas*, *Pseudoxanthomonas*, *Pseudofulvimonas*, *Luteimonas* or *Acinetobacter*); Bacteroidete (*Sphingobacterium*, *Prevotella* or *Chryseolina*); Firmicutes (*Weisella*, *Lactobacillus*, *Megasphaera*, *Clostridium* or *Brevibacillus*); and Thermus (*Truepera*). As for fungi, we recognized different genera belonging to Ascomycota (*Aspergillus*, *Thermomyces*, *Myceliophthora*, *Mycothermus*, *Madurella* and *Scedeosporium*); Basicomycota (*Coprinellus*, *Coprinopsis* and *Coprinus*); and Mucoromycota (*Mortierella*).

**Figure 1.** Relative abundance of different genera of bacteria (**A**) and fungi (**B**) in the three composts.

#### *3.3. Suppressiveness: Composts with Added Value*

In vitro, the three composts used in this study showed a higher percentage of mycelium growth inhibition (MGI) of *P. irregular* in comparison with peat, where no inhibition was observed. Compost C1 showed the highest MGI (100%), followed by C2 (73%), C3 (65%), and peat (49%) (*F* = 22.44, *p* = 0.001).

#### *3.4. Compost as a Component of Growing Media in Floating Systems*

#### 3.4.1. Percentage of Germination and Yield

The percentage of seed germination was significantly higher (*F* = 10.37, *p* = 0.0001) in plants grown in peat, at 94%, compared with those grown in compost growing media (GM) (82–84%).

The yield, on the other hand, was only affected by the growing media in the summer cycle (Table 4). The highest yield was observed in GM1, reaching more than 3 g/plant of the total yield (adding the two cuts), which means an increase of about 23% with respect to that obtained in peat (Table 4). Comparing the different cuts, the second cut was more productive than the first in both cycles, independent of the growing media.


**Table 4.** Yield and nitrate content in baby leaf red lettuce grown on different growing media (peat, GM1, GM2, GM3) in a floating system.

Asterisks indicate significances at \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; n.s.: non-significant. Different letters indicate significant differences. FW: fresh weight.

#### 3.4.2. Nitrate Content in Leaves and in the Nutrient Solution

Regarding the nitrate content in leaves, the two-way ANOVA indicated a significant interaction between the growing media and cut in both cycles (Table 4). In the spring cycle, the highest nitrate content values were obtained in the 2nd cut for GM2 and GM3. In the summer, the nitrate content was greater in the 2nd cut than in the 1st cut in all growing media, and the highest values were obtained with GM1 and GM2 in the 2nd cut.

During the spring cycle, in the 1st cut, the nitrate content measured in the nutrient solution was higher in plants grown in compost growing media than in plants grown in peat (Figure 2). In the 2nd cut, the nitrate concentrations decreased slightly for every compost growing media but not for peat, which remained constant, thus equalizing the values of all treatments by the end of the cycle. In the summer cycle, the lowest concentrations of nitrate in the nutrient solution in the 1st cut were found for GM1. In the 2nd cut, the nitrate concentration increased for every growing media, while GM1 maintained the lowest value.

**Figure 2.** Nitrate content in the nutrient solution for both cuts in the spring (**A**) and summer (**B**) cycles using different growing media (peat, GM1, GM2, GM3). Values are the mean ± SD (*n* = 9).

3.4.3. Total Phenolic and Antioxidant Capacity in Red Baby Leaf Lettuce Leaves and Roots

Regarding the total phenolic content, there was an interaction between growing media and cuts in both cycles (Table 5). The plants showed a similar phenolic content pattern in leaves when they were cultivated in the different growing media, although those with compost showed higher values, particularly in the 2nd cut. In general, the lowest values were found in the 1st cut in plants grown with peat. GM3 stood out for the high phenolic content values found in leaves in the 2nd cut in both cycles. The total phenolic content was always higher in summer than in spring.


**Table 5.** Total phenolic content in the leaves of baby leaf red lettuce grown on different growing media (peat, GM1, GM2, GM3) and harvested twice (1st and 2nd cut), cultivated in spring and summer cycles in a floating system.

Asterisks indicate significances at \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; n.s.: non-significant. Different letters indicate significant differences. GA: gallic acid. DW: dry weight.

The antioxidant capacity, measured by the ABTS and FRAP methods, highlighted a similar pattern to that found in the total phenolic content for both cycles, with higher values in the compost growing media, particularly in the 2nd cut and in the leaves of plants grown with GM3 (Table 6). However, there were some exceptions: we did not find significant differences between the growing media in terms of antioxidant capacity measured in leaves using either the ABTS method in the spring cycle or the FRAP method in the summer.

**Table 6.** Total antioxidant capacity (FRAP and ABTS methods) in leaves of baby leaf red lettuce grown on different substrates (peat, GM1, GM2, GM3) and harvested twice (1st and 2nd cut), cultivated in spring and summer cycles in a floating system.


Asterisks indicate significances at \* *p* < 0.05; \*\* *p* < 0.01; \*\*\* *p* < 0.001; n.s.: non-significant. Different letters indicate significant differences. DW: dry weight.

#### **4. Discussion**

To date, there have been few studies that have investigated the ideal growing media for a floating system. Among them, Cros et al. [4] demonstrated that a peat-based floating cultivation system can be considered the most suitable growing medium to grow purslane, because of its ideal physical and chemical characteristics. Nicola et al. [43] also recommended a peat-based horticultural medium for baby leaf vegetables grown in floating system. However, in recent years, there have been increasing environmental and ecological concerns about the use of peat as a growing medium because its harvest is jeopardizing endangered wetland ecosystems worldwide [45]. Furthermore, increasing demand and rising costs for peat as growing media in horticulture have led to a search for high-quality and low-cost substrates as an alternative. Compost may have physical, chemical, and biological properties that can contribute to partial peat reduction in growing media formulations [46]. In the case of a floating system, the three assayed composts showed physical properties similar to peat [47], although C1 showed a significantly higher WHC than peat, which could bring higher moisture and cause some negative aspects on plant growth [40]. Yet this issue was easily overcome given the type of trays used in this study, which contain a low volume of substrate per hole, and the system of cultivation (floating trays), where substrates obtain the water that they need and the roots mostly grow into the nutrient solution [30]. This makes the compost physical properties in this cultivation system a non-limiting factor for plant growth.

The C/N ratio of the three composts was less than 25, which indicates that the composts can be considered matured [48]. By using unmatured composts and/or unstable organic materials with a high C/N ratio, such as wood fibers, as the plant material degrades, N-immobilization occurs. This is accompanied by a decrease in soil volume (shrinkage), pore space, and air content [49–51]. Both pH and EC have an essential influence on seedling quality and plant growth. The three composts used in this study showed a basic pH ranging from 8.4 to 8.8; these values are higher than those recommended for growing media [52]. By mixing the composts with 25% peat, however, the pH was reduced for all growing media used (to 7.7, 7.8 and 7.9, for GM1, GM2 and GM3, respectively). As a result, we obtained a good germination rate and good seedling growth. These results are in accordance with those of Morales et al. [53]. With respect to EC, an EC <sup>≤</sup> 3.5 dS m−<sup>1</sup> is considered to be the limit for seedling growth in a growing medium [54]. Moreover, an EC > 4 dS m−<sup>1</sup> has been reported to inhibit seed germination [55]. In our study, the growing media presented a percentage of seed germination: ca. >82%, a level of germination that can be considered standard within the range normally found for this species.

Even if C1 and C3 exceeded the abovementioned EC limit, the peat used in the mixtures served as a thinner and reduced the salt concentration of the growing media to EC 3.3, 1.8, and 2.6 dS m−<sup>1</sup> for GM1, GM2, and GM3, respectively. Moreover, the high water holding capacity of C1 may also have positively influenced seed germination in GM1, because water retention is a decisive factor to this process [56]. Nevertheless, other factors could influence seed germination, due to the complexity of the mechanisms involved in it.

Furthermore, the growing media did not have any adverse effects on plant growth. In fact, the compost growing media promoted plant growth to a greater extent than peat, reaching higher yields. This beneficial effect of compost on yield could be due to the availability of nutrients and the production of auxin-like components from humic substances (Table 2). According to Trevisan et al. [57], compost acts as a reservoir for nutrients, ensuring their slow release to plant roots [58]. Moreover, some microorganisms found in our compost have been described as plant growth promoters (PGP). According to Castellano-Hinojosa et al. [59], strains of *Pseudoxanthomonas* promote plant growth via the production of ACC deaminase and siderophore and the solubilization of phosphate. In addition, Kuan et al. [60] found that inoculating maize with N2-fixing PGP strains belonging to genera *Acinetobacter* significantly increased the total N content and dry biomass of plants.

The time that the plants needed to reach the adequate phenological stage for the first cut was longer (27 and 21 days in spring and summer cycle, respectively) than the time from the first to the second cut (11 and 7 days in spring and summer cycle, respectively). As Jasper et al. [61] demonstrated recently in rocket plants, the growth rate prior to the first cut is slower than the subsequent regrowth rate due to the initial plant establishment. In addition, the second cut was more productive than the first in every growing media. Awan and Ahmad [62] and Suzuki et al. [63] also found that spinach foliage weight at the second harvest was greater than the weight at the first harvest, suggesting that new roots, which developed vigorously during the regrowth period, had a positive effect on the absorption of water and nutrients by the plants [63].

The three composts used in our study all showed suppressive activity against *P. irregulare*, which was not observed in the peat [64,65]. Hoitink et al. [66] pointed out a direct relationship between compost microbial activity and the suppression of *Pythium* and *Phytophthora* root rots. In our study, compost C1 showed the highest suppressiveness, which may be related to the fact that the primed plants displayed a faster and stronger activation of the various cellular defenses [64], it could also be due to microbial antagonism, nutrient competition, parasitism, and antibiosis [65]. Dehydrogenase activity, a potential indicator of general microbial activity [11], cannot be considered as an indicator for determining compost suppressiveness in this study, since no significant differences for this parameter were observed among the composts tested. Several studies have shown that compost microbial

composition primarily depends on the microbial competition for nutrients in different types of feedstock, which deeply influence compost recolonization during curing time [67,68]. The composts investigated here were recolonized by different microbial communities. Among them, we found microorganisms belonging to specific beneficial groups, such as *Aspergillus*, *Pseudomonas*, and *Morteriella* sp. These microorganisms are effective against different pathogens [69,70] and can induce systemic resistance in plants [71]. *Brevibacillus*, which has also been found to produce bioactive compounds against pathogens [72], showed higher relative abundance than other beneficial microorganisms in C1. Interaction between different beneficial microorganisms in the composts studied could contribute to their effectiveness in controlling *P. irregulare.*

According to Nicola et al. [1], baby leaf vegetables are a significant source of nitrates, so the nitrate content is an important quality characteristic to consider. Our data reveal that the nitrate concentrations did not exceed the maximum level allowed by the EU for this type of lettuce and cultivation system, although the use of compost did increase the amount of nitrates in comparison with peat. In both cycles, the nitrate content in the 2nd cut was higher than in the first. This fact could be linked to changes in the nutrient solution, due to both a gradual release of nitrate from the compost growing media and the evaporation of water from the floating beds, particularly in summer. Moreover, the nitrogen mineralization rate from organic substrates is higher in summer than in spring, due to the higher temperatures [73]. The influence of enriched nitrates in the nutrient solution could overcome the effect produced by the higher LDI in the summer cycle on nitrate reductase activity, which would increase the conversion rate of nitrate to amino acids, reducing nitrate levels in the leaves [74].

Plants grown in compost growing media showed a higher total phenolic content levels than plants grown in peat. The compost feedstocks used (tomato, leek, vineyard, and olive mill cake residues) are rich in compounds with the capacity to activate an oxidative process in plants [75]. As suggested by Santos et al. [76], these kinds of compounds induce the stimulatory effects of secondary metabolites on different parts of lettuce grown in different agro-industrial composts. The season also influenced the accumulation of phenolic compounds in the lettuce leaves: the total phenolics were higher in summer. This agrees with the results of Marín et al. [77], who found a positive correlation between the total phenolic content and temperatures. Besides, the root zone temperature has been found to influence the production of plant metabolites in several plants [78–80]. Temperature increases in the root zone in leafy vegetables can lead to alterations in the production of some secondary metabolites in greenhouse cultivation, phenolic compounds being the most pronounced secondary metabolites [81,82]. Furthermore, the higher phenolic compound levels after the 2nd cut could be linked to the increase in some phenolic metabolism enzymes due to the signals that spread from the injured tissue to the adjacent non-injured tissue after wounding, as observed by Salveit [83], who reported a 6 to 12-fold increase in PAL activity within 24 h after cutting in Batavia lettuce.

In our study, antioxidant capacity, measured by ABTS or FRAP methods, had a positive correlation with the total phenolic content in the leaves. The correlation coefficients were *r* = 0.67 and *r* = 0.86 for the first and second cuts in spring and *r* = 0.91 and *r* = 0.85 in summer, respectively. Santos et al. [76] found similar results in lettuce using agro-industrial composts as substrates. Moreover, we found a higher antioxidant capacity after the second cut. This agrees with the results of Kang and Saltveit [84], who demonstrated that the antioxidant capacity of lettuce leaf tissue increases after wounding. The baby leaf lettuce in our study showed the highest antioxidant capacity and total phenolic content in GM2 and GM3 growing media. This is most probably due to the original presence of olive mill cake, given that Chrysargyris et al. [85] reported an increase in antioxidant enzyme metabolism in marigold and petunia grown in soilless media, using up to 30% olive mill cake in place of peat. In general, these findings suggest that compost amendments can help add value to lettuce by increasing its antioxidant activity to a greater extent than other organic resources such as peat [76].

## **5. Conclusions**

Composts from different raw materials like vineyard wastes, tomato wastes, leek wastes, and olive mill cake can be an alternative to peat as a central component of the growing media in the production of baby leaf vegetables in a floating system. They not only increase crop yields due to their biofertilizer activity, but also boost the final quality of the product with a higher total phenolic content and antioxidant capacity. Moreover, composts were able to control the effect of *P. irregulare* due to their suppressive effect of bacterial–fungal interactions. However, the percentage of seed germination was higher in plants grown in peat than in those grown in compost growing media and the nitrate content in the leaves was greater in some of the compost treatments than in the peat treatment. Further studies are needed on the standardization of feedstocks origin, composting and stabilization processes in order to obtain standard growing media for this cultivation system.

**Author Contributions:** Conceptualization, M.R., J.A.P., J.A.F., C.E.G.; data curation, A.G., J.S.-T., E.M.-S.; formal analysis, A.G., M.R.; funding acquisition, J.A.F., J.A.P., C.E.-G., M.R.; investigation, A.G., J.S.-T., E.M.-S.; project administration J.A.F., J.A.P.; supervision, M.R.; J.A.F., C.E.G.; visualization, C.E.-G.; writing—original draft J.A.F., C.E.G., M.R.; writing—review and editing, J.A.F., C.E.-G., J.A.P., N.S.G. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Spanish Ministry of Economy and Competitiveness: Reference project: AGL2017-84085-C3-1-R, AGL2017-84085-C3-2-R, AGL2017-84085-C3-3-R.

**Acknowledgments:** This work was supported by projects AGL2017-84085-C3-1-R, AGL2017-84085-C3-2-R, AGL2017-84085-C3-3-R from the Ministry of Economy and Competitiveness of Spain.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## **Characterization of Physicochemical and Hydraulic Properties of Organic and Mineral Soilless Culture Substrates and Mixtures**

## **Mohammad R. Gohardoust 1, Asher Bar-Tal 2, Mohaddese E**ff**ati <sup>1</sup> and Markus Tuller 1,\***


Received: 20 August 2020; Accepted: 12 September 2020; Published: 16 September 2020

**Abstract:** Many arid and semiarid regions of the world face serious water shortages that are projected to have significant adverse impacts on irrigated agriculture and create unprecedented challenges for providing food and water security for the rapidly growing human population in a changing global climate. Consequently, there is a momentous incentive to shift to more resource-efficient soilless greenhouse production systems. Though there is considerable empirical and theoretical research devoted to specific issues related to control and management of soilless culture systems, a comprehensive approach that quantitatively considers relevant physicochemical processes within containerized soilless growth modules is missing. An important first step towards development of advanced soilless culture management strategies is a comprehensive characterization of hydraulic and physicochemical substrate properties. In this study we applied state-of-the-art measurement techniques to characterize six soilless substrates and substrate mixtures [i.e., coconut coir, perlite, volcanic tuff, perlite/coconut coir (50/50 vol.-%), tuff/coconut coir (70/30 vol.-%), and Growstone®/coconut coir (50/50 vol.-%)] that are used in commercial production in Israel and the United States. The measured substrate properties include water retention characteristics, saturated hydraulic conductivity, packing and particle densities, as well as phosphorus and ammonium adsorption isotherms. In addition, integral water availability and integral energy parameters were calculated to compare investigated substrates and provide valuable information for irrigation and fertigation management.

**Keywords:** soilless culture; organic and mineral substrates and mixtures; laboratory characterization; hydraulic properties; physicochemical properties

## **1. Introduction**

The projected growth of the world population to around 9.7 billion by 2050 [1] poses unprecedented challenges for providing and sustaining food and water security and mitigating associated economic inequalities and social tensions that threaten global security [2,3]. This is further exacerbated by climate change via alterations of precipitation patterns, more likely occurrence of climate extremes (e.g., prolonged droughts), and modification of diurnal and seasonal temperature regimes [4] and soil degradation that leads to an alarming reduction of arable land. Because of these imminent challenges as well as a strong demand for high-quality, out-of-season vegetables, fruits, and ornamentals in many industrial countries and the ban of methyl bromide fumigation of horticultural field soils, there is an increasing incentive to shift from soil to more resource-efficient soilless culture [5].

Substrates used in soilless culture systems exhibit major advantages over soils. Besides the alleviated risk for spreading soil-borne pathogens, physicochemical properties of growth substrates can be controlled within narrow margins, which commonly leads to healthier plants and higher yields when compared to soil-based production [6,7].

Organic substrates that are extensively used in soilless culture include peat moss, compost, coconut coir, bark and other wood-based materials, and biochar, all of which are commonly mixed with inorganic substrates such as perlite, volcanic tuff, expanded clay granules, pumice, zeolite, and sand, in order to improve their physicochemical properties [8,9].

Though the same physical principles apply to both soilless substrates and soils, their physical and hydraulic properties are vastly different, which is of significant importance for management and control of soilless growth systems. In addition, there are fundamental differences with regard to dynamic water, air, and nutrient distribution processes, and root growth and development between spatially confined containerized production systems and unconfined field soils. While water flow and nutrient transport in growth containers is restricted by an impermeable container bottom with drainage holes, water drains and redistributes to much deeper layers in agricultural soils unless natural impediments exist. This leads to vastly different infiltration and redistribution dynamics requiring more intensive management of soilless systems. The smaller the root zone the more intensive the production system needs to be managed to provide a stress-free rhizosphere environment for optimum plant growth [5].

While soils are well-researched, and many discovered soil physical principles are readily available for application to soilless substrates, their adaptation and translation to substrates appear to lag behind. The physical properties of essence for the design and management of containerized soilless production systems include bulk density (*BD*), particle density (i.e., specific gravity), the water characteristic (*WC*), and hydraulic conductivity (*K*) [10,11]. The substrate *WC* [12] that relates the water content to the matric potential (i.e., capillary and adsorptive surface forces that hold water under subatmospheric pressure within the substrate matrix) and *K* are the most important physical properties that govern water flow and distribution processes [13] and aeration [14] in containerized soilless systems. The matric potential (*h*) [15] determines the "ease" for plant roots to extract water from the substrate and is commonly expressed as a negative (subatmospheric) pressure. In general, all irrigation practices that explicitly attempt to avoid water stress in soilless production are confined to a matric potential range from 0 to −8 kPa. In some substrates, such as rockwool, the range is even narrower, with the onset of water stress occurring if *h* is allowed to attain values < −5 kPa. In contrast, matric potentials encountered in field soils may well go as low as −75 kPa; in such systems matric potentials rarely rise above −10 kPa, except during or immediately after irrigation [16].

Several concepts related to the *WC* and *K* have been introduced to determine plant water availability. These include plant available water capacity [17], easily available water and water buffering capacity [18], container capacity [19–21], limiting and least limiting water range [22,23], and the integral water capacity [24], the later accounting for aeration and root penetration. The more recently introduced integral energy concept calculates the energy required to extract water from the growth medium [14,25]. The water flux in soilless substrates that may significantly vary due to large changes of hydraulic conductivity within a narrow *h* range is another important parameter for irrigation management to avoid plant water stress [26,27]. Accurate measurements of hydraulic substrate properties (i.e., *WC* and *K*) are also essential for the parameterization of numerical computer codes for the simulation of water and nutrient dynamics in containerized soilless systems. Such simulations aid with the optimization of substrate mixtures for specific plants as well as with the design and management of soilless systems (i.e., container geometry and irrigation amount and frequency), which may reduce costly and time consuming trial and error greenhouse experiments [14].

Nutrient supply in conjunction with irrigation (i.e., fertigation) is another important aspect of soilless culture management that requires insights about the adsorption of nutrients on substrate surfaces. For example, phosphorus and nitrogen need to be continuously supplied due to limited container volumes and associated restricted nutrient buffering capacities [28–30]. Rapid depletion of phosphorus after fertigation is a well-documented phenomenon caused by electrostatic adsorption onto substrate surfaces and slow formation of new solid metal-phosphorus compounds [31,32]. To increase phosphorus uptake by plant roots, high frequency fertigation is commonly applied to induce nonequilibrium conditions [33–35]. Ammonium promotes optimum plant development and growth when the NH4-N/total-N ratio does not exceed plant specific thresholds that depend on species, rooting medium, root zone temperature, and pH [36–41]. For proper nutrient management, adsorption isotherms need to be determined to not only assure optimal growth conditions, but also to minimize nutrient loss in open-loop soilless culture systems.

The presented collaborative project that involves research teams from the U.S. and Israel was motivated by the rapidly growing demand for soilless growth media due to an ongoing momentous shift to more resource-efficient containerized soilless greenhouse production systems. It should be noted that the choice of soilless substrates and the selection of measured substrate properties was guided by ongoing production-scale greenhouse trials and the goal to utilize the obtained properties to parameterize a three-dimensional numerical code for simulation of water and nutrient dynamics in containerized growth modules to aid with their design and management. In the following we first discuss the selected substrates, then present a solid procedure for preparation of substrate mixtures, which is followed by an introduction of the applied state-of-the-art characterization techniques for the *WC*, saturated hydraulic conductivity (*Ksat*), and particle density as well as for the measurement of phosphorus and ammonium adsorption isotherms. We conclude the paper with a thorough discussion of obtained results.

#### **2. Materials and Methods**

#### *2.1. Investigated Soilless Substrates*

Six soilless substrates and substrate mixtures, including perlite (Figure 1a), volcanic tuff (Figure 1b), coconut coir (Figure 1c), a 50/50 vol.-% perlite/coconut coir mixture (Figure 1d), a 70/30 vol.-% volcanic tuff/coconut coir mixture (Figure 1e), and a 50/50 vol.-% foamed glass aggregate (i.e., Growstone®)/coconut coir mixture (Figure 1f), were investigated.

Horticultural perlite (Figure 1a) is a naturally occurring amorphous volcanic glass with high water holding capacity, typically formed through hydration of obsidian [42,43]. Perlite is usually sieved and then heated to 1000 ◦C. At high temperature water evaporates, and when rehydrated perlite expands to 4 to 20 times of its original volume [9], which yields a lightweight substrate with high porosity. Perlite aggregates are chemically inert and pathogen free [44], two desired attributes when plants remain in the same substrate for prolonged time periods [9,45]. However, if perlite is applied in high amounts, a negative impact on plant growth due to nutrient leaching may occur [46,47].

Tuff is a common name for pyroclastic volcanic material, exhibiting high porosity and surface area (Figure 1b). The physicochemical properties of tuff are mainly dependent on mineral composition and the weathering stage [48,49]. In addition, grinding and sieving processes may alter these properties. Tuff commonly exhibits a *BD* between 0.8 and 1.5 g cm−<sup>3</sup> and a total porosity between 60 and 80%. Tuff possesses a high buffering capacity and may adsorb or release nutrients, especially phosphorus, throughout the plant growth period [49,50].

Coconut coir (Figure 1c) is the mesocarp of Cocos nucifera L., containing short and medium length fibers left from industrial applications. Depending on origin and industrial source, there is a difference in physical and chemical characteristics [51,52]. The coconut coir dust is commonly sieved to desired sizes and washed to leach excess salts. Coconut coir exhibits remarkable physical and chemical properties such as high water holding capacity, good drainage and aeration properties, and high cation exchange capacity. It is also commonly used as a surrogate for peat and mixed with mineral substrates [53–55].

Foamed glass aggregates (Growstone®, Growstone, LLC, Santa Fe, NM, USA) are made of recycled glass bottles and windows. The production process starts with crushing and grinding glass into a fine powder of vitreous soda lime glass, which is mixed with calcium carbonate (2% on weight basis) that acts as a foaming agent. When the mixture is heated it expands, thereby creating a network of fine pores [9,56]. After the cooling process, the solid block of foamed glass is crushed, tumbled, and sieved to various aggregate sizes. The aggregates are commonly mixed with organic substrates.

**Figure 1.** Investigated soilless substrates and substrate mixtures. (**a**) perlite, (**b**) tuff, (**c**) coconut coir, (**d**) 50/50 vol.-% perlite/coconut coir mixture, (**e**) 70/30 vol.-% volcanic tuff/coconut coir mixture, and (**f**) 50/50 vol.-% foamed glass aggregate/coconut coir mixture.

#### *2.2. Sample Preparation*

To obtain uniform and reproducible substrate samples for hydraulic characterization we first performed comprehensive compaction trials to determine the lowest and highest achievable dry bulk densities for the considered soilless substrates. The average dry bulk densities were then used as initial target bulk densities for preparation of samples for substrate *WC* and *Ksat* measurements. Because of particle segregation during transport, the 50/50 vol.-% Growstone®/coconut coir mixture (Figure 1f) supplied by Growstone, LLC was separated, remixed, and homogenized. All tests were performed in sextuplicate for each substrate and substrate mixture. We used air-dry samples as this is the most realistic scenario for large-scale greenhouse applications and also to avoid potential problems with hydrophobicity of coconut coir that may be induced during oven drying.

Subsamples of perlite, tuff, and coconut coir were first oven-dried to determine the air-dry gravimetric water content. Then, the thoroughly homogenized air-dried substrates were compacted into cylinders with known volume (*VC*) in multiple layers to achieve a uniform packing density. To achieve the lowest potential packing density, the substrates were poured into and carefully manually distributed within the cylinders without imposing a significant compaction force. Only on the very top the substrate particles were gently pushed inside the cylinder to obtain a smooth surface. To achieve the highest potential packing density, the substrates were compacted layer by layer with a rubber stopper mounted on a push rod. At the end, the lowest and highest dry bulk densities were determined, and the average values were used as the target density for sample preparation for *WC* and *Ksat* measurements.

Compaction trials were also performed for the 50/50 vol.-% perlite/coconut coir mixture, the 70/30 vol.-% tuff/coconut coir mixture, and the 50/50 vol.-% Growstone®/coconut coir mixture. First, several subsamples of the individual substrates to be mixed were collected and oven-dried to determine their air-dried gravimetric water content. Once the gravimetric water content of the

individual mixture components was known, the substrates were poured into two separate cylinders of known volumes and compacted in the same fashion as described above for the lowest packing density. The air-dried mass of the substrates occupying a specific volume was then measured and the oven-dried masses per volume were calculated. The dry mass ratio (ϑ) may then be defined as:

$$\mathcal{S} = \frac{M\_{\text{OD}v1}}{M\_{\text{OD}v2}} \times R\_V \tag{1}$$

with *MODv* as the oven-dried mass of substrates 1 or 2 occupying a specific volume and *RV* the volumetric substrate mixing ratio (i.e., 50/50 vol.-% for perlite/coconut coir; 70/30 vol.-% for tuff/coconut coir; and 50/50 vol.-% for Growstone®/coconut coir).

For the compaction trials the air-dried substrate components were then mixed at the desired volumetric substrate mixing ratio and the resulting mixture was meticulously homogenized. The homogenized air-dried mixture was then compacted into cylinders in the same fashion as the individual substrates to obtain the lowest and highest achievable potential packing densities. After compaction, the mass of the air-dried mixture occupying the cylinder, *MADmix*, was determined and the oven-dried masses of the individual components composing the sample were calculated as:

$$M\_{\rm OD1} = \left[ M\_{\rm ADmix} - \frac{M\_{\rm ADmix}}{8\left(\frac{1+\partial\_{m1}}{1+\partial\_{m2}}\right)+1} \right] \cdot \frac{1}{1+\partial\_{m1}} \tag{2}$$

$$M\_{\rm OD2} = \frac{M\_{\rm ADmix}}{\mathfrak{sl}\left(\frac{1+\partial\_{m1}}{1+\partial\_{m2}}\right)+1} \cdot \frac{1}{1+\partial\_{m2}}\tag{3}$$

where *MOD*<sup>1</sup> and *MOD*<sup>2</sup> are the oven-dry masses of substrate 1 and 2, respectively, and θ*<sup>m</sup>* is the gravimetric water content. The dry bulk density of the mixture (ρ*b*-*mix*), which is used as target value for further measurements, was derived as:

$$
\rho\_{\text{b-mix}} = \frac{M\_{\text{OD1}} + M\_{\text{OD2}}}{V\_{\text{C}}} \tag{4}
$$

with *VC* as the cylinder volume. The mass of air-dry substrate required to fill a distinct volume (*V*) at target bulk density was calculated as:

$$M\_{AD\text{mix}} = \left[\frac{1+\theta\_{m1}}{1+\theta^{-1}} + \frac{1+\theta\_{m2}}{1+\theta}\right] \cdot \rho\_{b\text{-mix}} \cdot V\_{\mathbb{C}} \tag{5}$$

#### *2.3. Substrate Water Characteristic and Integral Energy and Water Storage*

Tempe cells (Soilmoisture Equipment Corp., Santa Barbara, CA, USA) were used to measure the substrate *WC* curve. The Tempe cells were connected to a pressure manifold (Figure 2) with a high-resolution pressure/vacuum regulator and initially saturated samples were sequentially desaturated by applying increasing pressures. Each pressure step was maintained until the sample was in equilibrium with the applied pressure and the outflow ceased. A detailed description of the pressure desaturation method is provided in [12]. All measurements were performed in quintuplicate and averaged values are reported.

**Figure 2.** Setup of the Tempe cell experiment.

The van Genuchten model [57] (Equation (6)) was fitted to *WC* measurements for substrates exhibiting unimodal behavior and the Durner model [58] (Equation (7)) was fitted to *WC* measurements exhibiting bimodal behavior.

$$\theta(h) = \begin{bmatrix} \theta\_r + (\theta\_s - \theta\_r) \end{bmatrix} \begin{bmatrix} 1 \\ 1 + |\alpha h|^n \end{bmatrix}^m \tag{6}$$

$$\theta(h) = \theta\_{\tau} + (\theta\_{\overline{s}} - \theta\_{\overline{t}}) \left[ (1 - w) \left( \frac{1}{1 + |a\_1 h|^{n\_1}} \right)^{m\_1} + w \left( \frac{1}{1 + |a\_2 h|^{n\_2}} \right)^{m\_2} \right] \tag{7}$$

where θ is the volumetric water content, θ*<sup>r</sup>* and θ*<sup>s</sup>* are the residual and saturated water contents respectively, *h* is the matric potential, and α, *n*, and *m* are shape parameters with *m* = 1 − 1/*n*. *w* is a weighting factor that varies between 0 and 1 and the indices 1 and 2 refer to the first and second substrate in the mixture, respectively.

To capture effects of the *WC* curve shape on plant water availability, rather than relying on two matric potential thresholds such as proposed in [17,22,23], we calculated the integral water (*WI*) and energy (*EI*) storage following [24,25] as:

$$\mathcal{W}\_{\rm I}[h\_{i\prime}h\_f] = \frac{1}{\left| l\_{\rm i\,} - h\_f \right|} \int\_{h\_f}^{h\_{\rm i}} \mathcal{O}(h) dh \tag{8}$$

where the indices *i* and *f* are the wet and dry matric potential cut-offs, respectively. *WI* has units of volumetric water content and represents the weighted average of water contents between *hi* and *hf*. The integral energy (i.e., the energy required to extract water from θ*<sup>i</sup>* to θ*f*) was calculated as:

$$E\_I[\theta\_{i\prime}\theta\_f] = \frac{1}{\theta\_i - \theta\_f} \int\_{\theta\_f}^{\theta\_i} h(h\theta)d\theta \tag{9}$$

The integrals in Equations (8) and (9) were numerically solved with the MATLAB®—Version R2019a software package (MathWorks, Natick, MA, USA). Based on obtained *WC* parameters and selection of plant specific cut-off values for the wet- and dry-end matric potentials and water contents, substrate water availability can be determined as:

$$\mathcal{R} = \frac{\mathcal{W}\_I}{E\_I} \tag{10}$$

*R* indicates the amount of water that can be extracted via exertion of a unit amount of energy by plant roots within the range of the wet- and dry-end thresholds. The higher *R*, the easier it is for plants to extract water.

### *2.4. Saturated Hydraulic Conductivity*

For the *Ksat* measurements we designed and fabricated an automated constant head device that was placed on a load cell attached to a laboratory jack and connected to a flow cell filled with substrate (Figure 3). The load cell was connected to a datalogger to record and monitor the weight change of the constant head container (i.e., Marriot tank) while water was flowing through the sample. In addition, the water temperature was continuously measured with a thermocouple and used to convert mass to volume change. The setup was initially thoroughly tested and adjusted to minimize flow resistance in the tubing and connectors. Each substrate was compacted into a flow cell at average bulk density (see Section 2.2). Before slowly saturating samples with water from the Marriot tank, they were flushed with CO2 for about 10 min at very low flow rate to enhance the saturation process. After sample saturation the constant head was adjusted with the lab jack and the experiment initiated. The experiment was terminated after several hours of steady state flow. Each substrate was measured in duplicate at 20, 15, 10, and 5 cm hydraulic heads. Darcy's law was applied to calculate *Ksat* from the measured water flux density and set hydraulic heads [59].

**Figure 3.** Automated constant head setup for *Ksat* measurements.

## *2.5. Particle Density*

While a standard water pycnometer [60] was used to measure the particle densities of tuff and coconut coir, nitrogen gas pycnometry was applied for the lighter perlite and Growstone® substrates. A Multipycnometer (Quantachrome Corp., Boynton Beach, FL, USA) with nitrogen as probing gas was used for the latter measurements. Gas pycnometry is based on Archimedes' fluid displacement principle and Boyle's gas expansion law. The volume of a solid or powder sample is determined via measuring the pressure drop that occurs when a known amount of pressurized gas initially contained

in a reference cell with known volume (*VR*) is allowed to expand into a cell of known volume (*VC*) that contains the sample. The sample volume vs. is calculated as:

$$V\_S = \left| V\_\mathbb{C} - V\_\mathbb{R} \right| \frac{P\_1}{P\_2} - 1 \Big| \tag{11}$$

where *P*<sup>1</sup> and *P*<sup>2</sup> are the pressures before and after gas expansion into the sample cell. All measurements were performed in quintuplicate. From the known oven-dry mass of the sample and its determined volume *VS*, the particle density can be calculated. The particle densities of the mixtures were calculated based on their mixing ratios.

#### *2.6. Phosphorus Adsorption Isotherms*

Phosphorus adsorption isotherms were measured in triplicate with adsorption batch experiments. The substrates were air dried and a 1-g subsample was added to a 50 mL equilibration tube. Then 20 mL of KH2PO4 solution with concentrations of 0, 1, 5, 10, 50, and 100 mg KH2PO4-P l<sup>−</sup><sup>1</sup> in the background of 0.01 M CaCl2 was added to the tubes to obtain a soil/solution ratio of 1:20. The pH of the solution was adjusted with 1M sodium hydroxide to fall between 6.5 and 7.0. The samples were left to equilibrate for 24 h in an end-over-end shaker. The supernatant was extracted after centrifuging for 15 min at 12,000 rpm and filtering with 0.2 μm paper filters.

The analysis of the filtrate soluble reactive phosphorus was carried out with the ascorbic acid colorimetric method for perlite, tuff, and tuff-coir substrates. Required reagents were prepared as follows: Molybdate Reagent: 12.0 g of ammonium molybdate was dissolved in 250 mL of deionized water and 0.1455 g of antimony potassium tartrate was also dissolved in 500 mL of 5N H2SO4. Then 125 mL of ammonium molybdate solution was thoroughly mixed with the 500 mL H2SO4-antimony potassium tartrate solution and diluted to one liter with deionized water using a volumetric flask. Color developing reagent was prepared as follows: in a 1L volumetric flask, 0.739 g of ascorbic acid was dissolved in deionized water and 70 mL of the molybdate reagent added and brought to volume. A series of standard PO4-P solutions with concentrations of 0, 1, 2, 3, and 4 ppm, were prepared for calibration of the spectrophotometer each time a measurement was made. 1 mL of sample solution was mixed with 9 mL of color developing reagent in a small tube and its P concentration was measured after about 1 h with a spectrophotometer at 880 nm wavelength.

Because colorimetric determination of the phosphorus concentration requires a clear solution, which was not the case for samples containing considerable amounts of coconut coir (i.e., coconut coir, perlite/coconut core mixture, and Growstone®/coconut coir mixture), the total phosphorus concentrations for these substrates were measured with Inductively Coupled Plasma Mass Spectrometry (ICP-MS) at the Arizona Laboratory for Emerging Contaminants (ALEC).

The linearized Langmuir adsorption equation was fitted to the measured data to obtain the substrate sorption parameters:

$$\frac{\mathcal{C}}{\mathcal{S}} = \frac{1}{k \mathcal{S}\_{\text{max}}} + \frac{\mathcal{C}}{\mathcal{S}\_{\text{max}}} \tag{12}$$

where *S* is the total amount of P retained (mg kg−1), *C* is concentration of P after 24 h equilibration (mg l<sup>−</sup>1), *Smax* is the maximum P sorption capacity (mg kg<sup>−</sup>1), and *k* is a constant related to the bonding energy, l (mg P)<sup>−</sup>1. Additional details are provided in [61].

#### *2.7. Ammonium Adsorption Isotherms*

Ammonium adsorption isotherms were calorimetrically determined in triplicate in batch experiments. Ammonium solutions were prepared in concentrations of 0, 1, 5, 10, 50, and 100 mg NH4Cl-N l<sup>−</sup>1. One gram of each substrate was agitated with 20 mL of the ammonium solutions in a centrifuge tube for 3 h after adjusting the pH with 1M sodium hydroxide to fall between 6.5 and 7.0. Samples were then centrifuged and filtered with 0.2 μm filter paper. The concentration of ammonium was measured with the salicylate method following [62] with the following reagents:


Four mL of extracted ammonium solution was mixed in a glass tube with 0.9 mL combined reagent (i.e., one part of reagent 1 mixed with two parts of reagent 2). Then within one minute 0.1 mL of reagent 3 was added to the tube, which was then placed in a dark room for 120 min to allow the establishment of the emerald blue color. The absorbance of the chromophore was measured with a spectrophotometer at 647 nm wavelength and the Langmuir adsorption model (Equation (12)) was fitted to the measured data.

## **3. Results and Discussion**

## *3.1. Bulk and Particle Densities*

The lowest and highest dry bulk densities achieved with the packing procedures described in Section 2.2 are listed in Table 1. The average values were used as the target bulk densities for the samples used for the *WC* and *Ksat* measurements. Perlite was the lightest of the investigated substrates with an average bulk density of 0.076 g cm<sup>−</sup>3, followed by the perlite/coconut coir mixture. Tuff exhibited the highest bulk density with an average value of 1.15 g cm<sup>−</sup>3. From transportation and handling point of view low bulk densities are desirable [63]. The determined dry mass ratios for the substrate mixtures are also displayed in Table 1.


**Table 1.** Dry bulk densities and oven-driedmass ratios of mixtures determined with compaction experiments.

The average particle densities and associated standard errors (SE) are listed in Table 2. As discussed in Section 2.5, a standard water pycnometer was employed for tuff and coconut coir and a gas pycnometer was used for the perlite and Growstone® substrates. Perlite exhibits the lowest particle density. The obtained value of 0.739 g cm−<sup>3</sup> falls within the reported range of 0.28–0.98 g cm−<sup>3</sup> [64,65]—the variations are attributable to differences in the production process. The highest particle density of 2.653 g cm−<sup>3</sup> was determined for tuff, which is due to the presence of significant amounts of metal oxides such as aluminum, iron, and magnesium [48].



#### *3.2. Substrate Water Characteristic and Saturated Hydraulic Conductivity*

The continuous parametric *WC* models of van Genuchten (VG) [57] and Durner [58] were fitted to the measured matric potential and volumetric water content pairs (Figure 4). For calculation of integral water storage (*WI*) and integral energy (*EI*) the threshold matric potential at the wet-end (*hi*) of the *WC* was set at −2 cm H2O below the substrate's air-entry potential (i.e., the potential at which the largest pore in the system starts draining and water is displaced by air—the transition from fully water-saturated to partially saturated) and the potential at the dry-end (*hf*) at −440 cm H2O. The latter was adapted from [66] for spring tomatoes, which is the major crop of our greenhouse trials (Figure 4). It should be noted that while the matric potential has a negative subatmospheric pressure (lower potential means larger negative number—−440 cm is lower than −2 cm), out of convenience it is commonly plotted on a positive scale with a minus sign in front of the units. It is also common to use units of lengths of H2O column (e.g., m) for the matric potential, which can be converted to pressure units (e.g., kPa) via multiplication with the density of water (kg m<sup>−</sup>3) and the acceleration due to gravity (m s<sup>−</sup>1). More details are provided in [12,15].

**Figure 4.** Measured *WC* data displayed with the fitted unimodal van Genuchten (Equation (6)) or bimodal Durner (Equation (7)) *WC* models for: (**a**) perlite, (**b**) tuff, (**c**) perlite/coconut coir, (**d**) tuff/coconut coir, (**e**) coconut coir, and (**f**) Growstone®/coconut coir. The error bars represent the standard deviation of the measured volumetric water contents. The wet- and dry-end matric potential thresholds, *hi* and *hf*, and their corresponding water contents, θ*<sup>i</sup>* and θ*f*, are marked with dash-dotted lines. The pore size distributions associated with the *WC* curves are plotted on the right side.

The hydraulic substrate properties consisting of the *WC* model parameters and the saturated hydraulic conductivity (*Ksat*) are listed in Table 3.


**Table 3.** Substrate WC parameters and measured *Ksat* (for parameter definitions see Equations (6) and (7) in Section 2.3).

\* Standard error of *Ksat* measurements.

The integral water storage and energy values calculated for each substrate are displayed in Table 4 together with their wet- and dry-end threshold water contents and *R* values (Equation (10)).

**Table 4.** Integral water storage and energy values and associated threshold water contents (for parameter definitions see Equations (8) to (10) in Section 2.3).


Because of its aggregated structure, perlite exhibits a bimodal pore size distribution (Figure 4a) with distinct contributions of inter- and intra-aggregate pores [67]. The bimodal pore structure and *WC* that was well approximated with the Durner model (Figure 4a) is consistent with observations by [68], who applied mercury intrusion porosimetry to measure the pore size distribution of uncrushed expanded perlite. It should be noted that a distinct bimodal pore structure of perlite was not reported in [69–71]. The saturated hydraulic conductivity of perlite that was slightly above that of tuff (Table 3) falls within the range provided in [70]. Based on the *WI*, *EI*, and *R* values listed in Table 4, it is obvious that the plant water availability (accessibility) of perlite is the highest of all investigated substrates. In other words, perlite yields the highest water amount between the respective threshold water contents (θ*<sup>i</sup>* and θ*f*) per unit energy exerted by plant roots.

In contrast to perlite, tuff provides the lowest water yield of the investigated substrates between θ*<sup>i</sup>* and θ*<sup>f</sup>* (Table 4). This in conjunction with its high *Ksat* (Table 3) indicates rapid drainage of the fertigation solution from the substrate, which provides valuable insights for irrigation and fertigation management to avoid problems with water and nutrient deficiencies. For example, an increase in irrigation frequency to keep the matric potential above −200 cm would double the plant water yield for the same applied energy. Wallach et al. [17] measured hydraulic characteristics of two red tuff varieties and reported *Ksat* values of 130 and 439 cm h−<sup>1</sup> and associated dry bulk densities of 1.227 and 1.091 g cm<sup>−</sup>3, respectively. They also evaluated the capability of the Mualem hydraulic conductivity model [72] to estimate unsaturated hydraulic conductivity from van Genuchten *WC* model parameters and found good agreement with data measured for tuff.

Coconut coir exhibits the lowest *Ksat* of the investigated substrates—about one-sixth of that of perlite and tuff (Table 3). Because of differences in industrial source and pretreatment of coconut coir, considerable variations in physicochemical and hydraulic properties can be expected [51]. The measured

*Ksat* of 56.2 cm h−<sup>1</sup> is about half of that measured by [73], who compacted the samples to a similar bulk density as used in this study. The *Ksat* value reported in [74] is more than one order of magnitude higher than our measurements, but due to the lack of information about the associated bulk density a direct comparison is not feasible. Their extremely high *Ksat* is most likely due to a much lower bulk density of the coconut coir in the narrow glass columns that were used in their experiments, which is also evident from the van Genuchten *WC* model α-parameter reported in [74]. In general, horticultural coconut coir does not contain a significant number of large pores. This is why it is commonly mixed with aggregated mineral substrates to enhance aeration properties [53]. In terms of plant water availability (i.e., *R*-value), pure coconut coir yields more water per unit of energy exerted by plant roots within the θ*i*–θ*<sup>f</sup>* range than the tuff/coconut coir and Growstone®/coconut coir mixtures (Table 4).

The perlite/coconut coir and tuff/coconut coir mixtures exhibit hydraulic properties that fall in between the properties of their constituents (Figure 4c,d; Tables 3 and 4). This includes their air-entry potentials, which enhances aeration relative to sole coconut coir. The addition of coconut coir to perlite and tuff also lowers the *Ksat* of the mixtures, slowing down drainage and increasing the water yield (availability) within the θ*i*–θ*<sup>f</sup>* range. For example, the 70/30 vol.-% tuff/coconut coir mixture has a 19% higher *R* value than the sole tuff substrate (Table 4). Such information may be utilized to optimize (engineer) substrate mixtures via varying mixing ratios to achieve optimum plant specific growth environments in terms of total porosity, air-filled porosity, available water, aeration, and bulk density [10,14,75] as well as to provide guidance for selection of container geometry and irrigation and fertigation management [76–78].

The Growstone®/coconut coir mixture has the highest (i.e., least negative) air-entry potential of the investigated substrates (Figure 4), which is also evident from the high α<sup>1</sup> Durner *WC* model parameter (Table 3). From Figure 4, it is evident that a −1 cm change in matric potential will cause an almost instantaneous drainage of water from about 25% of the entire pore space. As shown in [14], where both the *WC* and aeration properties of four soilless substrates were measured, caution is required when assessing aeration properties of mixtures containing large aggregates as water blockage and pore discontinuities might occur.

#### *3.3. Phosphorus and Ammonium Adsorption*

The Langmuir adsorption isotherm parameters for both phosphorus and ammonium are summarized in Table 5. The maximum amount of phosphorus adsorbed onto perlite, coconut coir, and the perlite/coconut coir mixture of about 20 mg per kilogram of solid is negligibly small. The low phosphorus absorptivity of perlite was previously reported by [63,79], who evaluated perlite as a potential filtration medium for urban runoff. Low phosphorus adsorption onto coconut coir was indicated in [80,81]. Tuff and the Growstone®/coconut coir mixture exhibited the highest phosphorus adsorption per unit substrate mass, about 15 and 12 times that of perlite and coconut coir (Table 5), respectively. It should be noted that while tuff and the Growstone®/coconut coir mixture show about the same capacity for phosphorus adsorption per unit substrate mass, the adsorption onto tuff within the same container volume will be more than six times higher than that onto the Growstone®/coconut coir mixture because of the significantly higher dry bulk density of tuff (Table 1).

Figure 5 depicts the measured equilibrium concentrations for both ammonium and phosphorus together with the fitted Langmuir isotherms. The low coefficients of determination (R2) for perlite, coconut coir, and Growstone®/coconut coir are attributable to low adsorption values (perlite and coconut coir) and the uncertainty inherent to the measurement procedure.


**Table 5.** Langmuir adsorption isotherm parameters for phosphorus and ammonium.

**Figure 5.** Measured *NH*<sup>4</sup> <sup>+</sup> and *H*2*PO*<sup>4</sup> − equilibrium concentrations displayed with the fitted Langmuir adsorption isotherms and the associated coefficients of determination.

The *k* coefficient in the Langmuir equation represents the affinity of the adsorbed species to the adsorbent (i.e., the higher *k*, the stronger the affinity). When the affinity is stronger, maximal adsorption is attained at lower adsorbate concentrations and there is a sharp increase in the adsorbed amount at low concentrations. The *k* values in Table 5 indicate that the order of affinities of phosphorus to the substrates is perlite > coconut coir > perlite/coconut coir mixture > tuff/coconut coir mixture > tuff > Growstone®/coconut coir mixture. Despite the high affinity of phosphorus to perlite, the importance of phosphorus adsorption is small due to the combination of low *Smax* and low bulk density. It was expected that the *k* values of mixtures of two components fall between the values of the pure components. However, note that the *k* of the perlite/coconut coir mixture is smaller than that of coconut coir, most likely due to chemical interactions between the coconut coir and perlite surfaces. The *k* value of the tuff/coconut coir mixture is much closer to that of tuff, which may be attributed to the much higher bulk density of tuff.

Because of its high cation exchange capacity (CEC) [51,82], which is the most important factor for ammonium adsorption [83], the maximum amount (*Smax*) of ammonium was adsorbed onto coconut coir. This translates to the mixtures containing coconut coir (Table 5). Perlite exhibited the lowest *Smax* value of the investigated substrates, which together with its low bulk density indicates that ammonium adsorption onto perlite is rather negligible. It should be noted that because the substrates were mixed on a volume basis, the dry mass ratio parameter (ϑ) in Equation (1) is crucial for estimation of adsorption properties of the substrate mixtures. For example, the 30 vol.-% coconut coir contained in the tuff/coconut coir mixture does not significantly increase ammonium adsorption. In contrast, the 50 vol.-% coconut coir contained in the perlite/coconut coir mixture significantly impacts ammonium adsorption due to an almost 30 times lower ϑ than that of the tuff/coconut coir mixture. The *k* values in Table 5 indicate that the order of affinities of ammonium to the substrates is perlite > tuff > tuff/coconut coir > perlite/coconut coir > coconut coir > Growstone®/coconut coir. Similar to phosphorus, the importance of ammonium adsorption to perlite is small due to the combination of low *Smax* and low bulk density. As expected, the *k* values for ammonium in the two component mixtures fall in between the values of the pure components.

#### **4. Conclusions**

A thorough physicochemical and hydraulic characterization of six soilless substrates and substrate mixtures that were selected based on ongoing greenhouse trials was presented. The investigated substrates included perlite, volcanic tuff, coconut coir, a 50/50 vol.-% perlite/coconut coir mixture, a 70/30 vol.-% volcanic tuff/coconut coir mixture, and a 50/50 vol.-% foamed glass aggregate (i.e., Growstones®)/coconut coir mixture. After developing a precise sample preparation procedure to assure high repeatability, the substrate *WC*, *Ksat*, particle densities, average bulk densities, as well as phosphorus and ammonium adsorption isotherms were measured with state-of-the-art techniques. The *WC* measurements were used to parameterize the unimodal van Genuchten [57] and bimodal Durner [58] *WC* models to derive integral water and energy storage parameters to estimate the amount of water that can be extracted from a specific volumetric water content range per unit energy exerted by plant roots. From integral energy calculations, it is evident that plant water availability (accessibility) of perlite is the highest of all investigated substrates, followed by the perlite/coconut coir mixture. Perlite also exhibits favorable nutrient adsorption characteristics. Despite the high affinity of phosphorus to perlite the importance of P adsorption is small due to a low maximum adsorption capacity and the low bulk density of perlite. In addition, ammonium adsorption to perlite is rather negligible. The obtained soilless substrate parameters can not only be applied for optimization (engineering) of soilless substrates via mixing of organic and inorganic constituents at different ratios to meet specific plant physiological demands, but also used for the parameterization of three-dimensional numerical computer codes for simulation of water and nutrient dynamics in containerized growth modules to aid with their design and management as well as to provide scientifically sound data for the design of greenhouse trials to avoid costly trial and error experiments, which motivated this study and is part of our ongoing research.

**Author Contributions:** Conceptualization, A.B.-T., M.R.G., and M.T.; methodology, M.R.G and M.T.; investigation, M.E. and M.R.G.; data analysis, M.E., M.R.G., and M.T.; writing—original draft preparation, M.E., M.R.G., and M.T.; writing—review and editing, A.B.-T., M.E., M.R.G., and M.T.; funding acquisition, A.B.-T. and M.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by The United States-Israel Binational Agricultural Research and Development Fund (BARD), grant number US-4764-14 R, and by the United States Department of Agriculture (USDA) National Institute of Food and Agriculture (NIFA), Hatch/Multi-State project number ARZT-1370600-R21-189.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Substrate Volumetric Water Content Controls Growth and Development of Containerized Culinary Herbs**

## **Christopher J. Currey \*, Nicholas J. Flax, Alexander G. Litvin and Vincent C. Metz**

Department of Horticulture, Iowa State University, 2206 Osborn Drive, Ames, IA 50011, USA; nickflax1@gmail.com (N.J.F.); aglitvin@ncsu.edu (A.G.L.); vmetz4@gmail.com (V.C.M.)

**\*** Correspondence: ccurrey@iastate.edu; Tel.: +1-515-294-1917; Fax: +1-515-294-0730

Received: 25 September 2019; Accepted: 21 October 2019; Published: 23 October 2019

**Abstract:** There are no chemical plant growth retardants that may be used on containerized culinary herbs intended for consumption. Our objective was to quantify the effect of substrate moisture content on the growth of four commonly produced culinary annual herbs grown in containers in the greenhouse. Seedlings of basil (*Ocimum basilicum* L.), dill (*Anethum graveolens* L.), parsley (*Petroselinum crispum* (Mill.) Fuss), and sage (*Salvia o*ffi*cinalis* L.) were transplanted into 11.4 cm diameter containers filled with commercial soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite and amended with 3.0 kg·m−<sup>3</sup> of controlled-release fertilizer. After the containers were thoroughly irrigated to container capacity, plants were placed into a sensor-controlled irrigation system, which maintained substrate volumetric water content (VWC) at 0.15, 0.28, 0.30, 0.38, or 0.45 m3·m−3. Chlorophyll fluorescence, photosynthesis, stomatal conductance, and transpiration were measured 27 d after initiating treatments, and the results showed that chlorophyll fluorescence of parsley and photosynthesis of basil increased as substrate VWC increased from 0.15 to 0.45 m3·m<sup>−</sup>3; the remaining parameters for basil, parsley, and sage were unaffected. Additionally, height, width, leaf area, and shoot dry mass of basil, dill, parsley, and sage increased as substrate volumetric water content increased from 0.15 to 0.45 m3·m<sup>−</sup>3. Our results show that growth of basil, dill, parsley, and sage can be promoted or inhibited by providing or withholding water, respectively, with no signs of stress or visual damage resulting from reduced substrate volumetric water content. Therefore, restricting irrigation and substrate volumetric water content is an effective nonchemical growth control method for containerized culinary herbs grown in peat-based substrate.

**Keywords:** restricted deficit irrigation; soil moisture sensors; nonchemical growth control; water use efficiency

## **1. Introduction**

One of the primary challenges associated with growing containerized herbaceous plants is controlling shoot growth to produce plants that are proportional and aesthetically balanced to the container height. Controlling shoot growth is important to produce plants that are sized proportionally to containers for aesthetic appearance as well as to increase container density in the greenhouse and during shipping [1]. Although chemical plant growth retardants (PGRs) are commonly used to control containerized ornamental crop growth, there are currently no PGRs that are labeled for use on containerized culinary herbs [2]. Therefore, nonchemical methods of controlling containerized herb growth must be used.

There are several nonchemical growth control techniques that may be used to control containerized herb growth [3–5]. Compact cultivars are available for some herb species, including basil and dill [3], and may be more appropriately sized for container production. The concentration of mineral nutrients provided to container-grown herbs, both total and specific nutrients, also affects growth. For example,

basil supplied with 200 mg·L−<sup>1</sup> N from a complete, balanced water-soluble fertilizer are 33% larger than plants supplied with 50 mg·L−<sup>1</sup> N from the same fertilizer [4]. Additionally, restricting P to 5 mg·L−<sup>1</sup> produced basil, dill, parsley, and sage shorter than plants provided with 40 mg·L−<sup>1</sup> [5]. While cultivar selection and nutrient management are useful forms of nonchemical growth control, it may be necessary to use multiple nonchemical methods of controlling growth to achieve the degree of control required in the absence of PGRs.

Reducing irrigation or substrate volumetric water content (VWC), commonly referred to as "deficit irrigation", is another effective method of controlling containerized plant growth [6–8]. The water available for plant uptake increases and growth is promoted as substrate VWC increases and, as such, restricting irrigation and reducing the substrate VWC can diminish turgor pressure and subsequent stem extension and growth [9]. For example, containerized angelonia (*Angelonia angustifolia* Benth.) and petunia (*Petunia* × *hybrid* Vilm.) bedding plant growth is promoted by substrate VWC and, by reducing VWC, compact plants of marketable quality can be produced [7,10]. Additionally, using regulated deficit irrigation can suppress stem elongation of flowering potted plants, such as poinsettia (*Euphorbia pulcherrima* Willd. ex Klotzsch), providing adequate height control during production [11]. While controlling the substrate VWC clearly has potential for use in containerized herb production, data specific to the effects of substrate VWC on containerized herb growth are lacking.

We have found some limited reports on the effects of substrate moisture on containerized perennial herb growth [8,9]. Zhen et al. [8] reported that limiting irrigation of rosemary (*Rosmarinus o*ffi*cinalis* L.) plants successfully controlled excessive growth. Additionally, Zhen and Burnett [9] showed that English lavender (*Lavandula angustifolia* Mill. 'Hidcote' and 'Munstead') growth diminished with decreasing substrate VWC. These data are promising for controlling containerized herb growth by limiting substrate VWC. However, we have found no other data on the use of drought stress to control excessive growth of more common containerized herb species grown with shorter production periods. Our objective was to quantify the effect of substrate VWC on the growth of four common culinary annual herbs grown in containers in the greenhouse. We hypothesized that the growth of parsley, sage, basil, and dill would be promoted by increasing substrate VWC and, as such, restricting irrigation would be an effective growth-control strategy for containerized culinary annual herb species with short growth cycles.

#### **2. Materials and Methods**

Seeds (Johnny's Selected Seed, Albion, ME, USA) of parsley (*Petroselenium crispum* (Mill.) Fuss 'Giant of Italy'), common sage (*Salvia o*ffi*cinalis* L.; Expt. 1), basil (*Ocimum basilicum* L. 'Italian Large Leaf'), and dill (*Anethum graveolens* L. 'Fernleaf'; Expt. 2) were individually sown in 288-cell propagation trays (PL-288-1.25; 7.1 cm<sup>3</sup> individual cell vol.; T.O. Plastics, Clearwater, MN, USA) filled with a soilless germination substrate comprising (by vol.) 65% fine sphagnum peat moss, 20% fine perlite, and 15% vermiculite (Propagation Mix; Sun Gro Horticulture, Agawam, MA, USA). Trays were initially hand-irrigated with clear, tempered tap water. Beginning at radicle emergence, seedlings were irrigated with tap water supplemented with a blend of water-soluble fertilizers (50 and 100 mg·L−<sup>1</sup> N provided from 21N–2.2P–16.6K and 15N−2.2P−12.5K, respectively; Everris NA, Inc., Marysville, OH, USA) to provide the following (in mg·L−1): 150 nitrogen, 8.6 phosphorous, 92.2 potassium, 33.3 calcium, 13.3 magnesium, 0.75 iron, 0.4 manganese and zinc, 0.2 copper and boron, and 0.5 molybdenum.

Seedlings were grown on expanded metal benches in a glass-glazed greenhouse at Iowa State University, Ames, IA (latitude 42◦ N) with fog cooling, radiant hot-water floor and perimeter heating, and retractable shade curtains controlled by an environmental computer (ARGUS Titan; ARGUS Control Systems, Surrey, BC, Canada). The day and night greenhouse air temperature set points were 23.0 ± 1 ◦C and 18.0 ± 1 ◦C, respectively. Aluminized shade cloth (XLS 15 Revolux; Ludvig Svensson, Kinna, Sweden) was drawn across the crop when outdoor light intensities exceeded 1000 <sup>μ</sup>mol·m−2·s−<sup>1</sup> to avoid leaf scorch. High-pressure sodium lamps delivered a supplemental photosynthetic photon flux (PPF) of ~190 <sup>μ</sup>mol·m−2·s−<sup>1</sup> at plant height (as measured with a quantum sensor (LI-190 SB; LI-COR Biosciences, Lincoln, NE, USA)) when ambient light intensity was below 100 <sup>μ</sup>mol·m−2·d−<sup>1</sup> between 0600 and 2200 hr to maintain a target daily light integral (DLI) of ~12 mol·m−2·d<sup>−</sup>1.

Four weeks after sowing, seedlings were planted into 11.4 cm diameter containers (655 mL vol.; HC Companies, Middlefield, OH, USA) filled with soilless greenhouse substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite (Sunshine® LB-2; Sun Gro Horticulture, Inc., Agawam, MA, USA) amended with 3.0 kg·m−<sup>3</sup> controlled-release fertilizer (Florikan Plus 16.0 N–2.2 P–9.1 K with a 90 d release period; Florikan ESA, Sarasota, FL, USA). For each experimental unit, 20 plant containers were placed into two 10-cell petroleum-plastic shuttle trays adjacent to each other with individual plants spaced on 12 cm centers (69.4 plants per m2). The inner six plant containers were measured for data gathered, while the surrounding plants were used as border plantings to simulate greenhouse practices.

An automated irrigation system controlled by soil moisture sensors was used to maintain VWC treatments similar to that described by Nemali and van Iersel [12]. Drip irrigation stakes attached to 1.9 L·h−<sup>1</sup> pressure-compensating emitters (Netafim USA, Fresno, CA, USA) were inserted into the substrate, and plants were irrigated overhead to container capacity with clear tempered water. After overhead irrigation, capacitance moisture sensors (EC-5; Decagon Devices Inc., Pullman, WA, USA) were inserted into the substrate of the two innermost plant containers within each experimental unit. Sensors connected to a multiplexer (AM16/32B; Campbell Scientific, Logan, UT, USA) cycling measurement readings to a data logger (CR1000; Campbell Scientific, Logan, UT, USA) calculated VWC using a manufacturer-provided calibration curve specific to soilless peat-based substrates. Substrate VWC thresholds were 0.15, 0.23, 0.30, 0.38, and 0.45 m3·m−3, and they were chosen to represent the range of VWC to be observed in commercial production. The VWC values were maintained by the data logger controlling a solenoid valve (Orbit Irrigation Products, Inc., Bountiful, UT, USA) connected to polyethylene tubing with drip emitters for each experimental unit. Irrigation events occurred as needed when the average measured VWC of the two moisture sensors within a given experimental unit fell below its respective threshold. The data logger program was executed every 10 min to determine need. Solenoid valves corresponding to each experimental unit were controlled by a relay driver (SDM-CD16AC controller; Campbell Scientific, Logan, UT, USA) connected to the data logger. Valves opened for 10 s during each irrigation event, providing 6.2 mL of clear water per plant per event. Substrate moisture content and total irrigation volumes are presented in Figures 1 and 2, respectively.

Plants were grown in the greenhouse as previously described. The air temperature was measured every 15 s by four temperature probes (41342; R.M. Young Company, Traverse City, MI, USA) in an aspirated radiation shield (43502; R.M. Young Company, Traverse City, MI, USA), while the PPF was measured every 15 s by eight quantum sensors (LI-190SL; LI-COR Biosciences, Lincoln, NE, USA) per greenhouse. Temperature probes and quantum sensors were connected to a data logger (CR1000 Measurement and Control System; Campbell Scientific, Logan, UT, USA) with means logged every 15 min. The mean day, night, and daily temperatures and DLI are reported in Table 1.

**Table 1.** Mean (± standard deviation) daily light integral (DLI), average daily air temperature (ADT), and average day (DT) and night (NT) air temperature for parsley and sage (Expt. 1) or basil and dill (Expt. 2) grown in 11.4 cm diameter containers filled with a soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite amended with 3.0 kg·m−<sup>3</sup> and maintained at 0.15, 0.23, 0.30, 0.38, or 0.45 m3·m−<sup>3</sup> substrate volumetric water content (VWC) for four weeks.


**Figure 1.** Substrate moisture for parsley and sage (Expt. 1) and basil and dill (Expt. 2) grown in 11.4 cm diameter containers filled with a soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite amended with 3.0 kg·m−<sup>3</sup> controlled-release fertilizer and maintained at 0.15, 0.23, 0.30, 0.38, or 0.45 m3·m−<sup>3</sup> substrate volumetric water content for four weeks.

**Figure 2.** Total irrigation volume and water use efficiency (WUE) for parsley and sage (Expt. 1) and basil and dill (Expt. 2) grown in 11.4 cm diameter containers filled with a soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite amended with 3.0 kg·m−<sup>3</sup> controlled-release fertilizer and maintained at 0.15, 0.23, 0.30, 0.38, or 0.45 m3·m−<sup>3</sup> substrate volumetric water content for four weeks. Regression lines are presented for significant correlations only with corresponding *R<sup>2</sup>* presented. \* and \*\*\* indicate significant at *p* ≤ 0.05 or 0.001, respectively.

Four weeks after transplanting seedlings, data were collected. Chlorophyll fluorescence of three plants per treatment per replication was measured on the adaxial epidermis of the most fully expanded leaf using a chlorophyll fluorescence meter (Handy Plant Efficiency Analyzer; Hansatech Instruments Ltd., Norfolk, U.K.). Using the manufacturer's clip, leaves were dark-acclimated for 15 min before measurements were taken. Fluorescence was measured by opening a shutter in the dark-acclimating clip and exposing the leaf to light with a peak wavelength of 650 nm provided by up to <sup>3000</sup> <sup>μ</sup>mol·m−2·s−<sup>1</sup> for 1 s to saturate photosystem II. Chlorophyll fluorescence was expressed as a ratio of the change in chlorophyll fluorescence from initial to maximum, to maximum fluorescence (*Fv*/*Fm*).

Gas exchange measurements were conducted with a portable photosynthesis system (LI-6400XT; LI-COR Biosciences, Lincoln, NE, USA) on two plants per treatment per replication. The second most recently matured leaf placed in a 6 cm2 leaf chamber with a light-emitting diode light source (6400-02B; red at 665 nm and blue at 470 nm) providing 400 <sup>μ</sup>mol·m−2·s−1. The reference CO2 concentration inside the leaf chamber was 500 <sup>μ</sup>mol·mol<sup>−</sup>1, and the flow of air into the chamber was set to maintain a constant mole fraction of 8.0 mmol·mol−<sup>1</sup> of water inside the chamber. Leaf temperature inside the leaf chamber was maintained at 23.0 ◦C.

Height was measured from the substrate surface to the tallest growing point. Width was determined by measuring the widest point and 90◦ perpendicular and averaging these two measurements. Branch length was determined by measuring a branch at a node approximately half the total height of the plant. The number of nodes was counted. Leaf area was determined by scanning all leaves of each plant with a leaf area meter (LI-3000; LI-COR Biosciences, Lincoln, NE, USA). Shoots were severed at the substrate surface, placed in a paper bag, and dried in a forced-air oven at 67 ◦C for 3 d, after which shoots were weighed and the dry mass recorded. Water use efficiency (WUE) was calculated by dividing the shoot dry mass by the total irrigation volume applied per plant. Internode length was determined by dividing the height by the node number.

The experiment employed a randomized complete block design for each species. There were three blocks (replications) for each VWC for each species, with six individual plants per block. Data were analyzed using regression analyses (Sigma Plot 21.0; Systat Software, San Jose, CA, USA), with VWC concentration as the independent variable.

#### **3. Results**

#### *3.1. Parsley*

Target substrate VWC for 0.15, 0.23, 0.30, 0.38, and 0.45 were achieved 13, 8, 6, 5, and 3 d later, respectively (Figure 1). Total irrigation volume increased linearly from 587 to 1825 mL as VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> (Figure 2). The photosynthesis (*Pn*), conductance (*gs*), and transpiration (*E*) of parsley was unaffected by VWC, while *Fv*/*Fm* increased from 0.82 to 0.84 as VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> (Figure 3). Height and width of parsley increased quadratically in response to VWC (Figure 4). For example, height increased by 14.8 cm as VWC increased from 0.15 to 0.38 m3·m<sup>−</sup>3, while plants grown at 0.45 m3·m−<sup>3</sup> were 1.6 cm shorter compared to those grown at 0.38 m3·m−<sup>3</sup> (Figure 4); width followed a similar trend. Increasing VWC promoted node appearance, as plants grown at 0.38 and 0.45 m3·m−<sup>3</sup> had approximately one additional node compared to those grown at 0.15 m3·m−<sup>3</sup> (Figure 4). Leaf area increased quadratically by 57.0 or 57.5 cm2 for plants grown at 0.38 or 0.45 m3·m<sup>−</sup>3, respectively, compared to plants grown at 0.15 m3·m−<sup>3</sup> (39.2 cm2; Figure 4). The shoot dry mass also increased quadratically from 4.5 to 14.9 g as substrate VWC increased from 0.15 to 0.45 m3·m<sup>−</sup>3, respectively. There was no significant relationship between substrate VWC and WUE of parsley (Figure 2).

**Figure 3.** Photosynthesis (*Pn*), conductance (*gs*), transpiration (*E*), and chlorophyll fluorescence (*Fv*/*Fm*) of parsley and sage (Expt. 1) and basil (Expt. 2) grown in 11.4 cm diameter containers filled with a soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite amended with 3.0 kg·m−<sup>3</sup> controlled-release fertilizer and maintained at 0.15, 0.23, 0.30, 0.38, or 0.45 m3·m−<sup>3</sup> substrate volumetric water content for four weeks. Regression lines are presented for significant correlations only with corresponding *R2* presented. \*\* indicates nonsignificant or significant at *<sup>p</sup>* <sup>≤</sup> 0.01.

**Figure 4.** Height, width, node number, leaf area, and shoot dry mass of parsley and sage (Expt. 1) and basil and dill (Expt. 2) grown in 11.4 cm diameter containers filled with a soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite amended with 3.0 kg·m−<sup>3</sup> controlled-release fertilizer and maintained at 0.15, 0.23, 0.30, 0.38, or 0.45 m3·m−<sup>3</sup> substrate volumetric water content for four weeks. Regression lines are presented for significant correlations only with corresponding *R<sup>2</sup>* presented. \*, \*\*, or \*\*\* indicates significant at *p* ≤ 0.05, 0.01, or 0.001, respectively.

*3.2. Sage*

The time to reach target substrate conditions decreased with increasing substrate VWC, taking 10 d to reach 0.15 m3·m−<sup>3</sup> and 4 d to reach 0.45 m3·m−<sup>3</sup> (Figure 1). The total irrigation volume required to maintain substrate VWC increased from 612 to 1531 mL as VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> (Figure 2). Neither *Fv*/*Fm* nor gas exchange of sage were affected by VWC (Figure 3). The height, width, and internode length increased from 15.7 to 24.4 cm, 14.5 to 23.3 cm, and 2.0 to 3.0 cm as VWC increased from 0.15 to 0.30 m3·m<sup>−</sup>3, respectively, then decreased to 24.0 cm, 22.6 cm, and 3.0 cm, respectively, as VWC further increased up to 0.45 m3·m−<sup>3</sup> (Figures 4 and 5). Similarly, leaf area

increased from 12.2 to 28.5 cm<sup>2</sup> as VWC increased from 0.15 to 0.38 m3·m−<sup>3</sup> (Figure 4). While node number and branch length for sage grown at 0.15 m3·m−<sup>3</sup> was 7.5 and 2.9 cm, respectively, plants grown at 0.23 to 0.45 m3·m−<sup>3</sup> had 8.2 to 8.3 nodes and branches between 6.7 and 8.9 cm long (Figure 5). Shoot dry mass increased from 4.8 to 12.3 g as VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> (Figure 4). The WUE of sage was unaffected by substrate VWC (Figure 2).

**Figure 5.** Branch and internode length of sage (Expt. 1) and basil (Expt. 2) grown in 11.4 cm diameter containers filled with a soilless substrate comprising (by vol.) 75% sphagnum peat moss and 25% coarse perlite amended with 3.0 kg·m−<sup>3</sup> controlled-release fertilizer and maintained at 0.15, 0.23, 0.30, 0.38, or 0.45 m3·m−<sup>3</sup> substrate volumetric water content for four weeks. Regression lines are presented for significant correlations only with corresponding *<sup>R</sup><sup>2</sup>* presented. \*\* or \*\*\* indicates significant at *<sup>p</sup>* <sup>≤</sup> 0.01 or 0.001, respectively.

#### *3.3. Basil*

Increasing substrate VWC from 0.15 to 0.45 m3·m−<sup>3</sup> reduced the time from 12 to 4 d to reach VWC targets, respectively (Figure 1), whereas the amount of water required to maintain target substrate VWC increased linearly from 616 to 1674 mL (Figure 2). Although *Fv*/*Fm*, *gs*, and *E* were unaffected by substrate VWC, *Pn* increased linearly from 5.0 to 11.6 <sup>μ</sup>mol·m−2·d−<sup>1</sup> as VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> (Figure 3). Similarly, as substrate VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> the height, width, internode length, leaf area, branch length, and shoot dry mass increased by 4.6 cm, 4.3 cm, 0.7 cm, 17 cm2, 5.9 cm, and 9.1 g, respectively (Figures 4 and 5). The WUE of basil ranged from 1.41 to 1.51 g·mL−<sup>1</sup> across substrate VWC and were unaffected by treatments (Figure 2).

#### *3.4. Dill*

Substrate VWC for dill reached 0.15, 0.23, 0.30, 0.38, and 0.45 m3·m−<sup>3</sup> 13, 9, 7, 5, and 2 d after imposing treatments, respectively (Figure 1). The height and width of dill increased quadratically by 12.2 and 8.1 cm, respectively, as substrate VWC increased from 0.15 to 0.38 m3·m−<sup>3</sup> but then diminished as VWC was further increased to 0.45 m3·m−<sup>3</sup> (Figure 4). Leaf area increased linearly from 9.0 to 56.1 cm<sup>2</sup> as substrate VWC increased from 0.15 to 0.45 m3·m<sup>−</sup>3, respectively (Figure 4). Similarly, dill shoot dry mass increased linearly by 5.5 g as substrate VWC increased from 0.15 to 0.45 m3·m−<sup>3</sup> (Figure 4). There was no effect of substrate VWC on the number of nodes. The WUE of dill increased by 0.71 g·mL−<sup>1</sup> as substrate VWC increased from 0.15 to 0.38 m3·m−<sup>3</sup> but then decreased as substrate VWC increased to 0.45 m3·m−<sup>3</sup> (Figure 2).

#### **4. Discussion**

The growth and development of containerized basil, dill, parsley, and sage is promoted with increasing substrate VWC. While the effect of substrate moisture on growth is better understood for containerized ornamental flowering crops, our results on the effect of substrate VWC on controlling growth of culinary herbs align well with the limited literature on container-grown herbs, including rosemary and English lavender [8,9]. For example, Zhen et al. [8] reported that, as substrate VWC increased from 0.05 to 0.40 m3·m<sup>−</sup>3, the height, width, leaf number and area, and fresh and dry mass of rosemary increased linearly. Similarly, height, width, leaf number, and area of 'Munstead' and 'Hidcote' English lavenders increased as substrate VWC increased from 0.10 to 0.40 m3·m−<sup>3</sup> [9]. The effect of substrate VWC on WUE of containerized herbs was not consistent among species in the study, with parsley, sage, and basil not being affected by VWC, whereas WUE of dill increased as VWC increased up to 0.38 m3·m<sup>−</sup>3. This variation reflects what is seen in the literature, where WUE was found to increase with increasing substrate VWC for burkwood vibrurnum (*Viburnum* × *burkwoodii* Burkwood & Skipwith) and butterfly bush (*Buddleja davidii* Franch.); decrease with increasing substrate VWC for potato (*Solanum tuberosum* L.), salvia (*Salvia splendens* Sellow ex Roem. & Schult.), vinca (*Catharanthus roseus* (L.) G. Don), and wax begonia (*Begonia* × *semperflorens*-*cultorum* Hort.); or remain unaffected by substrate VWC for cheddar pink (*Dianthus gratianopolitanus* L.), columbine (*Aquilegia canadensis* L.), geranium (*Pelargonium* × *hortorum* Bailey), petunia, and rosemary [8,13–17].

The growth of basil, dill, parsley, and sage are promoted or inhibited by the provision or restriction of water to the root zone and, as such, restricting the substrate VWC to plants and growing them drier using restricted deficit irrigation is a viable nonchemical growth control method for container-grown culinary herbs. Although growing containerized herbs with restricted VWC reduces shoot mass, the harvestable or useable portion of most culinary herbs, it is important to distinguish between containerized and fresh-cut herb production. Containerized herb plants are sold as individual units (i.e., per container), not on the unit weight basis (i.e., gram) that fresh-cut culinary herbs are sold. For producers of fresh-cut herbs grown in substrate, using higher substrate VWC can promote shoot growth and yields, potentially enhancing productivity and profitability.

Although growth and development of herbs were greater at increasingly higher VWC, gas exchange was unaffected for parsley and sage (Figure 3). Under low water availability, gas exchange is reduced in most plants compared to higher availability [18]. For example, *Pn* and *gs* of Mediterranean herbs sea beet (*Beta maritima*) and wall-rocket (*Diplotaxis ibicensis*) decreased with increasing water deficit stress [19]. Similarly, gas exchange (*Pn*, *gs*, and *E*) of English lavender grown with sensor-based irrigation increased with VWC increasing from 0.10 to 0.40 m3·m−<sup>3</sup> [9]. According to Yan et al. [20], annual herbs do not vary greatly in gas exchange with changing water status, suggesting limited response regulation, although the method of imposed stress may affect this. Montesano et al. [21] reported that, when irrigation was completely withheld for basil, the *Pn*, *gs*, and *E* decreased after three days. However, the authors also reported that, when VWC was controlled using sensor-based irrigation and maintained 0.20, 0.30, or 0.40 m3·m<sup>−</sup>3, fresh mass increased with increasing VWC, whereas *Pn*, *gs*, and *E* were unaffected by increasing VWC. In contrast, *Pn* in our study increased for basil as VWC increased from 0.15 to 0.45 m3·m<sup>−</sup>3; however, within 0.20 to 0.40 m3·m<sup>−</sup>3, *Pn* was similar to reports by Montesano et al. [21]. Taken together, our results align well with the literature for suppressed growth and development at lower VWC and for gas exchange under sensor-based irrigation for herbs. Drought stress reduced *Fv*/*Fm* in plants compared to well-watered conditions, which is in agreement with chlorophyll content for nontolerant species [22,23]. Nemali and van Iersel [14] reported that, as VWC increased, the quantum yield efficiency of photosynthesis increased for petunia, salvia, impatiens, and vinca, similar to parsley in this study, although basil and sage were unaffected, similar to previous reports by [9].

Sensor-based precision irrigation effectively restricted irrigation of containerized herbs in this experiment. This is especially useful for edible crops with no chemical PGRs labeled for use on them during greenhouse forcing [8] and for using drought as a nonchemical growth control method [6]. To consistently produce containerized crops at a lower substrate, VWC can be a challenge using non-sensor-controlled systems as judging the appropriate time to irrigate becomes more difficult [24,25]; automated sensor-based systems are well suited for controlling substrate VWC at desired set points [26]. Sensor-based irrigation also precisely controls substrate moisture, with minimal variation in VWC within treatment groups after initial dry down (Figure 1). However, aside from implementing precision irrigation strategies for producing containerized crops, there are other benefits when using these systems in commercial applications. Automated sensor-based irrigation is not only used to restrict irrigation for controlling height [6] but also to improve water use [24], plant growth uniformity [27], biomass [28], flower number [29], plant stress symptoms, and disease pressure [30] and can increase profitability of commercial producers compared to visual inspection- or timer-based irrigation scheduling [31].

## **5. Conclusions**

The research presented here comprehensively quantifies the effect of substrate moisture on container-grown basil, dill, parsley, and sage regarding growth, development, and gas exchange. The growth and development of containerized culinary herbs, including height, width, node number, leaf area, and branching, were all controlled by substrate VWC, with growth and development restricted at lower VWC compared to those at higher VWC. However, while growth was suppressed when substrate VWC was lower, there were a few instances where *Pn*, *gs*, *E*, or *Fv*/*Fm* were negatively impacted. Taken together, reducing substrate VWC and implementing restricted deficit irrigation is an effective growth-controlling strategy for containerized culinary herb production. Sensor-based irrigation allows for precise substrate moisture control to implement restricted deficit irrigation for controlling crop growth, although other tangible benefits may be realized in commercial production facilities. The research presented herein was performed using a round plastic container with a peat and perlite substrate. However, the different substrates that are either currently used or will be used in the future as peat alternatives [32], as well as different container shapes and sizes [33], can affect the water-holding capacity of substrates; therefore, additional work on culinary herb growth and substrate moisture content grown with different substrates and containers would be useful. While the results we have presented support the use of restricting substrate moisture to control containerized herb growth, commercial producers should conduct their own trials to determine the effectiveness of this growth-controlling technique under their unique circumstances, including the specific species and cultivars produced under specific greenhouse environmental conditions and crop culture.

**Author Contributions:** Conceptualization, C.J.C.; methodology, C.J.C., N.J.F., and A.G.L.; formal analysis, A.G.L.; investigation, V.C.M. and N.J.F.; writing—original draft preparation, C.J.C.; writing—review and editing, N.J.F., A.G.L. and V.C.M.; supervision, C.J.C.; funding acquisition, C.J.C.

**Funding:** This research was funded by the Fred C. Gloeckner Foundation.

**Acknowledgments:** We gratefully acknowledge Peter Lawlor for greenhouse assistance. The use of trade names in this publication does not imply endorsement by Iowa State University of products named, nor criticism of similar ones not mentioned.

**Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **Microbe-Plant Growing Media Interactions Modulate the E**ff**ectiveness of Bacterial Amendments on Lettuce Performance Inside a Plant Factory with Artificial Lighting**

## **Thijs Van Gerrewey 1,2,3,4, Maarten Vandecruys 3, Nele Ameloot 4, Maaike Perneel 5, Marie-Christine Van Labeke 6, Nico Boon <sup>2</sup> and Danny Geelen 1,\***


Received: 22 August 2020; Accepted: 21 September 2020; Published: 23 September 2020

**Abstract:** There is a need for plant growing media that can support a beneficial microbial root environment to ensure that optimal plant growth properties can be achieved. We investigated the effect of five rhizosphere bacterial community inocula (BCI S1–5) that were collected at three open field organic farms and two soilless farms on the performance of lettuce (*Lactuca sativa* L.). The lettuce plants were grown in ten different plant growing media (M1–10) composed of 60% *v*/*v* peat (black peat or white peat), 20% *v*/*v* other organics (coir pith or wood fiber), 10% v/v composted materials (composted bark or green waste compost) and 10% *v*/*v* inorganic materials (perlite or sand), and one commercial plant growing medium inside a plant factory with artificial lighting. Fractional factorial design of experiments analysis revealed that the bacterial community inoculum, plant growing medium composition, and their interaction determine plant performance. The impact of bacterial amendments on the plant phenotype relied on the bacterial source. For example, S3 treatment significantly increased lettuce shoot fresh weight (+57%), lettuce head area (+29%), root fresh weight (+53%), and NO3-content (+53%), while S1 treatment significantly increased lettuce shoot dry weight (+15%), total phenolic content (+65%), and decreased NO3-content (−67%). However, the effectiveness of S3 and S1 treatment depended on plant growing medium composition. Principal component analysis revealed that shoot fresh weight, lettuce head area, root fresh weight, and shoot dry weight were the dominant parameters contributing to the variation in the interactions. The dominant treatments were S3-M8, S1-M7, S2-M4, the commercial plant growing medium, S1-M2, and S3-M10. Proper selection of plant growing medium composition is critical for the efficacy of bacterial amendments and achieving optimal plant performance inside a plant factory with artificial lighting.

**Keywords:** plant growth-promoting rhizobacteria (PGPR); growing media; rhizosphere; lettuce; plant factory; soilless culture; plant quality; plant yield; microbiome; beneficial bacteria

## **1. Introduction**

A growing world population in the course of climate change requires the food supply chain to be revised to secure future universal access to food in a sustainable way [1,2]. In controlled-environment agriculture (CEA), the recent development of state-of-the-art plant factories with artificial lighting (PFAL) allows maximizing plant growth in a resource use efficient way (water, CO2, fertilizer, energy, etc.) [3]. Plant factories with artificial lighting can tap into new markets that are inaccessible to open-field production and conventional greenhouses by locally producing leafy greens, herbs, medicinal plants, and transplants year-round for local consumption [4].

Plant factories with artificial lighting utilize soilless culture methods [5]. Soilless culture typically requires a plant growing medium that provides a proper physicochemical and biological environment for rooting and plant growth during the seedling stage [6]. Peat, partially degraded *Sphagnum* mosses that accumulated over thousands of years under waterlogged conditions within mires, has been widely used as a plant growing medium because of its low economic cost and good performance [7,8]. However, access to peat will be limited because of sustainability and environmental concerns involving the peat production process [9–11]. Sustainable alternatives are being investigated and a variety of these are on the market (e.g., coir pith, wood fiber, composted materials, biochar, etc.) [6,8,12–16]. Nevertheless, peat will remain an essential plant growing medium constituent, for dilution purposes at any rate as it allows the blending of alternative and circular raw materials [7]. At the same time, because of the expanding world population, the demand for plant growing media is expected to increase drastically [17]. Newly developed peat-reduced plant growing media have to perform equal to or even outperform peat, to ensure universal access to food.

When selecting new plant growing medium materials, environmental factors have become as important as performance and economic cost. However, little attention is given to the microbial properties of these products and their potential to support the amendment of plant growth-promoting rhizobacteria (PGPR). Contrary to plant growing media, soil bacterial communities are widely researched [18]. Soils contain an immense diversity in bacterial communities, enabling various soil ecosystem functions [19]. However, only a minority of bacterial taxa, including Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria, encompass the diversity present in soils [19,20]. Plants are in continuous contact with soil bacterial communities through their roots. Via rhizodeposition, plants recruit soil bacteria to the rhizosphere and endosphere that improve the capacity of the plant to adapt to the environment [20–25]. These PGPRs can stimulate germination, enhance growth, improve nutrient acquisition, promote stress resistance, and enable disease suppression [26–29].

Globally, agro-industries are starting to embrace PGPR technology but are confronted with strong variation in efficacy of PGPR application, with no benefits to considerable benefits being reported [30–33]. The underlying factors causing the differential activity are not well known. The development of bacterial amendments mainly focused on single strain PGPR products [34–37]. The complexity of bacterial communities and their interactions with environmental factors and crop specificity is suspected to play an important role in the success of the plant-microbe interaction [26,34,38,39].

Plant growing medium composition may be a determining factor in the successful amendment of microbes in a soilless environment. Rhizosphere bacteria show specific microbial substrate uptake traits that drive the assembly of the rhizosphere bacterial community [21]. In addition to plant root exudate chemistry, plant growing media could provide a source of microbial substrate allowing modulation of the rhizosphere microbiome for improved plant performance [40,41]. The role of plant growing media in beneficial plant-microbe interactions is not well studied [26]. There is evidence that plant growing media have distinct microbial features that can provide stability and resilience to crops in a diverse soilless environment. The complex biological and physicochemical interactions in organic plant growing media influence the rhizosphere microbial communities of the plant [42]. Organic plant growing media have a more diverse and sTable microbial community that decreases the susceptibility of the eggplant *Solanum melongena* to the hairy roots pathogen *Agrobacterium rhizogenes* [43]. Composts maintain a high microbial diversity that is critical to the suppression of soil-borne pathogens and improving plant performance [44–47]. Biochar amendment to peat growing media and soil may improve plant growth and disease suppressiveness [48–51]. These positive effects of biochar amendment are linked to the activity, diversity, and composition of the rhizosphere microbial community [52,53]. There is evidence that PGPR amendment can improve plant growth and decrease phytopathogen infections in soilless culture [37,54–57]. Though, the role of plant growing medium composition as a potential driver in the success of PGPR amendment is much less clear. Recent research has studied the use of plant growing medium constituents as a carrier material for bacterial inocula [33,40,41,58,59]. For example, Nadeem et al. [41] showed that the combined use of biochar, compost, and the PGPR *Pseudomonas fluorescens* alleviated the negative effect of water deficit on cucumber growth. More research has to be done on the mechanisms of action and the efficiency of using different plant growing medium constituents as a carrier for PGPR consortia.

At the start of our work, we hypothesized that plant growing medium composition plays a decisive role in the effectiveness of PGPR amendment inside a complex PFAL environment. Here we report results that show that specific microbe-plant growing medium interactions are the major determinants of performance for *Lactuca sativa* L. (lettuce). Seedlings of lettuce, a leafy green that is abundantly produced in PFALs, were grown in different plant growing media, inoculated with a few selected bacterial communities, and transferred to a PFAL. The different plant growing media were composed by varying five raw material groups: (a) peat (black peat and white peat), (b) other organics (coir pith and wood fiber), (c) composted materials (composted bark and green waste compost), (d) inorganic materials (perlite and sand), and (e) Arabic gum dosed at 1 kg·m−<sup>3</sup> or 5 kg·m<sup>−</sup>3. Lettuce root-associated bacterial community samples were collected from soil and soilless farms and used as an inoculum. Shoot fresh weight (FW), lettuce head area (LHA), root fresh weight (RW), shoot dry weight (DW), total phenolic content (TPC), NO3-content, and leaf pigments were quantified.

#### **2. Materials and Methods**

#### *2.1. Collection of Root-Associated Bacterial Communities*

Lettuce root-associated bacterial community samples (S1–5) were collected at five different locations in Flanders, Belgium during the growing season: three open field organic farms and two soilless farms. An overview of all sampling locations can be found in Table 1. Sampling and extraction were performed following the method described by Barillot et al. [60]. Briefly, 30 plant and root-associated soil samples (20 cm<sup>2</sup> by 30 cm deep) were collected at each location, transported in polyethylene bags, and stored at 4 ◦C. Bulk soil was removed by manually shaking the roots. The rhizosphere fraction was collected by manually washing the roots in a sterile 0.9% NaCl solution for 10 min. Roots were subsequently washed by hand in sterile 0.9% NaCl + 0.01% Tween 80 for 10 min to obtain the rhizoplane fraction. Both fractions were homogenized on an orbital shaker (125 rpm, 90 min, room temperature). The homogenized samples were centrifuged at low speed (150 g, 10 min, room temperature) to separate soil particles and other debris from the supernatant containing bacteria. The supernatants were centrifuged at high speed (9425 g, 10 min, room temperature) to collect the bacteria in the pellet. The bacterial pellets were resuspended in tryptic soy broth (TSB) + 15% glycerol and stored at −80 ◦C.

**Table 1.** Overview of Rhizosphere Sampling Locations.


The amount of live bacterial cells present in the rhizosphere and rhizoplane fractions was estimated using flow cytometric analysis to standardize bacterial inoculation in further analysis (see Section 2.3). The samples were diluted and stained with SYBR® Green I combined with propidium iodide (SGPI, 100 <sup>×</sup> concentrate SYBR® Green I, Invitrogen, and 50 <sup>×</sup> 20 mM propidium iodide, Invitrogen, in 0.22 μm-filtered dimethyl sulfoxide) for live-dead analysis. Staining was performed as described previously, with incubation for 13 min at 37 ◦C [61]. Samples were analyzed immediately after incubation on a C6+ flow cytometer (BD Biosciences, Belgium), which was equipped with four fluorescence detectors (530/30 nm, 585/40 nm, >670 nm, and 675/25 nm), two scatter detectors and a 20-mW 488-nm laser. The flow cytometer was operated with Milli-Q (Merck, Darmstadt, Germany) as sheath fluid.

#### *2.2. Plant Growing Media Composition*

Ten different experimental plant growing media were composed (M1–10; Table 2). The raw material collection took place at Agaris Belgium NV, Gent, Belgium. All plant growing media have following volumetric composition: 60% *v*/*v* peat, 20% *v*/*v* other organics, 10% *v*/*v* composted materials and 10% *v*/*v* inorganic materials. For eight plant growing media (M1, M3, M4, M5, M7, M8, M9, and M10), selection of the raw material and the Arabic gum dose was based on a 25−<sup>2</sup> III fractional factorial design (Tables S1 and S2). Based on a previous study [62], two more plant growing media were composed: M2 and M6, both showing high microbial activity potential. The peat and coir based commercial plant growing medium (75% peat and 25% coir fibers, Jiffy International AS, Kristiansand, Norway) was used as a control to evaluate the performance of the experimental plant growing media. The physicochemical properties of each plant growing medium were analyzed in triplicate following Verdonck and Gabriels [63], and Gabriels et al. [64]. The data obtained is shown in Table S3.

**Table 2.** Composition of Plant Growing Media. Each plant growing medium consists of 4 raw material groups at different volume per volume (% v/v): peat (black peat BP or white peat WP), other organics (coir pith CP or wood fiber WF), composted materials (composted bark CB or green waste compost GC) and inorganic materials (perlite P or sand S). Arabic gum was dosed at 1 kg·m−<sup>3</sup> or 5 kg·m<sup>−</sup>3.


## *2.3. Plant Growth and Inoculation*

Sterilized hydroponic mesh pots, with 6.5 cm height, 5 cm bottom diameter, and 7 cm top diameter, were fitted with hydroponic paper (Ellepot, Esbjerg, Denmark), filled with 200 mL of plant growing medium, and watered to saturation. Batavia lettuce seeds (Enza Zaden, Enkhuizen, The Netherlands) were sown in ten pots of each plant growing medium (Table 2). The seeds were wetted by spraying water. The pots were placed in a sterilized tray inside a growth chamber (Urban Crop Solutions, Beveren-Leie, Belgium) with a temperature of 22–23 ◦C, relative humidity of 60–70%, and 800 ppm CO2-fertilization. LED light fixtures (Urban Crop Solutions, Beveren-Leie, Belgium) provided an 18 h light regime at 220 <sup>μ</sup>mol.m−2·s<sup>−</sup>1. For the following two weeks, the pots were irrigated by hand with tap water when necessary. Two weeks after sowing six pots with uniform lettuce seedlings were selected per plant growing medium and placed in a sterilized tray fitted with an overflow drain for

automated irrigation. In each tray, the six selected plants from a single plant growing medium were positioned at a distance of 18.6 cm in length and 22.2 cm in width from each other.

At this point, the bacterial community inocula (BCI S1–5) were applied to all experimental plant growing media at the base of the plant. Based on the live bacterial cell counts, determined with flow cytometric analysis (see Section 2.1), equal volumes of the collected rhizosphere and rhizoplane fractions were mixed and diluted with TSB. Application of 1 mL of inoculum provided a dose of 3.2 <sup>×</sup> <sup>10</sup><sup>9</sup> CFU per L plant growing medium. As a positive control treatment (PGPR), *Bacillus* sp. with plant growth-promoting properties was added as an inoculum to each plant growing medium at a dose of 3.2 <sup>×</sup> <sup>10</sup><sup>9</sup> CFU per L plant growing medium. As a negative control treatment (C), 1 mL of sterile TSB solution was added to every plant growing medium. Unlike the experimental plant growing media, the commercial plant growing medium was only treated with 1mL of sterile TSB solution. After inoculation, the trays were placed inside a PFAL (Urban Crop Solutions, Beveren-Leie, Belgium) for three weeks under the growing conditions as mentioned above. During these three weeks, all plants were irrigated automatically four times a day with the following nutrient solution: 14 mM NO3 −, 2 mM PO4 <sup>3</sup>−, 7 mM K+, 4 mM Ca2+, 2 mM Mg2+, 635 μM SO4 <sup>2</sup>−, 72 μM Fe2+, 18 μM Mn2+, 2 μM Zn2+, 46 μM B, 0.8 μM Cu2+, 1 μM Mo2−, and 356 μM Si.

The experiment was split into five batches. Each batch consisted of all ten experimental plant growing media treated with one bacterial community inoculum, two randomly selected experimental plant growing media treated with the positive control, and two randomly selected experimental plant growing media treated with the negative control. The commercial plant growing medium was added to the last batch.

#### *2.4. Plant Sample Analysis*

#### 2.4.1. Plant Sample Processing

The plants were harvested three weeks after inoculation. During harvest, top view images were taken to determine the lettuce head area (LHA) by image processing in ImageJ [65]. The harvested plants were transported in polyethylene bags to avoid excessive transpiration and stored at 4 ◦C until further processing. Within 24 h, the plant samples were cut at the base to separate root and shoot. Shoot fresh weight (FW) was determined by weighing the lettuce head immediately after cutting. After weighing, a section (weighing approximately 10 g) of the whole lettuce head, containing young and mature leaves, was cut out. This subsample was ground using an IKA A11 liquid nitrogen mixer (IKA, Staufen, Germany) and stored at −80 ◦C until further analysis. To determine the shoot dry weight (DW), the remaining shoot was placed in a paper bag and dried at 70 ◦C for 72 h. The difference in weight before and after drying was used to calculate the shoot dry weight of the sample. Next, the dried subsample was ground with a coffee mill (Proficook PC-KSW 1021, Clatronic International GmbH, Kempen, Germany) and stored until further analysis. If the total shoot weight was too low for obtaining both fresh and dry subsamples, priority was given to the fresh subsample. This was the case for the following treatments: S1-M6, S2-M3, PGPR-M3, PGPR-M6, and PGPR-M9.

The roots and plant growing medium of the sample were used to isolate the root-associated bacterial community following the procedure described in Section 2.1. After the second washing step, plant roots were weighed to determine root fresh weight (RW).

#### 2.4.2. Total Phenolic Content

Total phenolic content (TPC) was ascertained following the Folin–Ciocalteu method [66]. Colorimetric TPC measurements of fresh subsample extracts in 80% methanol were carried out with a Tecan infinite plate reader (Tecan Group Ltd., Männedorf, Switzerland) at a wavelength of 765 nm. Total phenolic content was expressed as mg gallic acid equivalents (GAE) per 100 g FW.

#### 2.4.3. Nitrate Content

The nitrate (NO3) concentration was determined colorimetrically with salicylic acid as described by Cataldo et al. [67]. Oven-dried subsamples were used. Measurements were performed by a Tecan infinite plate reader at a wavelength of 410 nm.

#### 2.4.4. Chlorophylls and Carotenoids

Chlorophyll a (Chla) a, chlorophyll b (Chlb), and carotenoids were quantified by UV-VIS spectroscopy of a whole-pigment extract of the fresh subsamples in 80% acetone [68]. Absorption at 470 nm, 648.8 nm, and 663.2 nm wavelengths, and zero absorption at 750 nm were measured with a Tecan infinite plate reader. The amount of Chla, Chlb, and carotenoids (Cx+c) were calculated in μg.mL-1 with the following equations:

$$\rm Chl\_a = 12.25 \times A\_{663.2} - 2.79 \times A\_{646.8} \tag{1}$$

$$\text{Chl}\_{\text{b}} = 21.5 \times \text{A}\_{646.8} - 5.1 \times \text{A}\_{663.2} \tag{2}$$

$$\mathbf{C\_{x+c}} = (1000 \times \mathbf{A\_{470}} - 1.82 \times \mathbf{C\_{lh}} - 85.02 \times \mathbf{C\_{lh}})/198 \tag{3}$$

#### *2.5. Statistical Analysis*

Before subjecting the plant performance data to statistical analysis, any data points further than 1.5 times the interquartile range from the mean were considered as outliers, and were removed from the dataset. Analyses of differences between BCI means and principal component analysis (PCA) were carried out using R 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria). All statistical analyses were performed at the 95% confidence level. Differences between BCI means were analyzed per plant growing medium. Quantile-quantile plots were used to check for normality of the data (*stats* package). Levene's test was used to determine the homogeneity of variance across groups (*car* package). In case the assumptions of normality and homoscedasticity were met, one-way ANOVA was used to determine significant differences between BCI means (*stats* package). As a post hoc test, a linear model was created using the *stats* package. Following, the estimated marginal means were calculated, using the Tukey's honest significance test for separation of the means at the *P* < 0.05 level (*emmeans* package). Finally, a compact letter display was created using the *multcomp* package. In case the assumptions of normality and homoscedasticity were not met, the Kruskal–Wallis test was used to compare BCI means (*stats* package). Dunn's test with Bonferroni correction was used as a post hoc method to separate the means (*FSA* package). A compact letter display of the comparison of means was produced using the *rcompanion* package.

Principal component analysis was used to determine which BCIs and plant growing media contribute most to the variation in data, and to which key performance parameters they are associated to. The data was standardized by scaling to unit variance before analysis (*stats* package). A quality of representation (cos2) correlation circle, a contribution plot of the variables, and a contribution plot of the samples were generated using the *factoextra* package.

A 1/4 fractional factorial statistical design of experiments (DOE; 25−<sup>2</sup> III ) was used to simultaneously evaluate the effect of the plant growing medium raw material groups (five control factors having a high +1 and a low −1 factor level) and their interactions on the plant performance parameters. The fractional factorial design was established and analyzed in Minitab 17 (Minitab Inc., State College, Pennsylvania, United States) using main effects plots, ANOVA, and response optimization. The design was extended with an additional control factor to determine the effect of inoculation. The levels of the inoculation control factor were: negative control treatment (C) as low factor level and inoculum treatment (S1–5) as high factor level. An overview of all control factors and the final fractional factorial design can be found in Tables S1 and S2. Following decisions were made to deal with aliasing effects. (a) Three-factor and higher-order interactions are extremely rare and were omitted. (b) When aliasing occurred between the

main effect and two-factor interactions, the main effect was assumed significant. (c) Aliasing between two-factor interactions was resolved by following the heredity principle: an interaction effect is likely significant when the main effects involved are also significant [69].

### **3. Results**

## *3.1. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on Shoot Fresh Weight*

Both BCI (*P* < 0.001) and plant growing medium (*P* < 0.001) significantly altered FW. Lettuce FW varied from 6.03 g (S2-M3) to 74.85 g (S3-M8). Significant differences in FW were observed between BCIs in each plant growing medium (Figure 1). Bacterial community inoculum S3 significantly (*P* < 0.05) increased FW in multiple plant growing media (M5, M7, M8, M9, and M10) compared to C. For example, FW was 17.78 g in C-M7 compared to 50.83 g in S3-M7, which is more than a 2.5-fold increase. S3-M8 (74.85 g) and S3-M10 (64.63 g) were the only BCI and plant growing medium combinations that had significantly (*P* < 0.05) higher FW than the commercial plant growing medium (48.65 g). On average, inoculating the plant growing media with BCI S3 increased FW with 57% (*P* < 0.001; Figure 2f). Response optimization showed that, excluding BCI S3, the addition of a BCI was not vital to reaching maximal FW (Table S4). Moreover, BCI S3 treatment was the largest contributor to FW, compared to the plant growing medium raw material groups (Figure S1). The positive effects of S3 on FW do not occur in each plant growing medium, underlining the importance of plant growing medium composition on the effectiveness of BCI treatment.

**Figure 1.** Boxplot of shoot fresh weight (FW; g) grouped per plant growing medium. Letters show comparison of BCI means per plant growing medium at the 95% confidence level. S indicates the bacterial community inoculum, M indicates the plant growing medium, C indicates the negative control treatment without addition of inoculum, and PGPR indicates the positive control treatment with a *Bacillus* sp. inoculum. Number of plants ≥ 3.

**Figure 2.** Main effects of plant growing medium constituents on shoot fresh weight (FW; g) under different bacterial community inoculum treatments (S1–5 and positive control PGPR). (**a**) Peat (PT; −1 = black peat and 1 = white peat); (**b**) Other organics (OO; −1 = coir pith and 1 = wood fiber); (**c**) Composted materials (CM; −1 = composted bark and 1 = green waste compost); (**d**) Inorganic materials (IM; <sup>−</sup>1 = perlite and 1 = sand); (**e**) Arabic gum (AG; <sup>−</sup>1 = 1 kg.m-3 and 1 = 5 kg.m<sup>−</sup>3); (**f**) Bacterial inoculum (BCI; −1 = C and 1 = S1–5 or PGPR). Dashed lines indicate mean levels of FW for each bacterial treatment. Asterisks indicate level of significance: *P* < 0.05 (\*), *P* < 0.01 (\*\*) and *P* < 0.001 (\*\*\*).

Surprisingly, the positive control treatment PGPR significantly decreased FW in several plant growing media (M1, M4, M6, M8, M9, and M10) compared to negative control C (Figure 1). For example, treating M9 with PGPR decreased FW with 70% compared to C. On average, the positive control treatment significantly decreased FW with 41% (*P* < 0.05; Figure 2f).

A significant (*P* < 0.001) interaction between plant growing medium and BCI was observed. Indeed, DOE analysis revealed a significant (*P* < 0.05) interaction effect between BCI S3 and the type of other organics (Figure S2). In the absence of S3, FW of lettuce grown in plant growing media containing coir pith (42.15 g) was higher than plant growing media with wood fiber (25.37 g). When treated with S3, lettuce FW increased and the difference in FW between the OO raw materials was no longer visible (coir pith: 53.42 g and wood fiber: 52.72 g). Treatment with BCI S3 negates the advantage of using coir pith over wood fiber.

Design of experiments analysis showed significant differences in FW between plant growing media, following similar trends in each BCI treatment (Figure 2). Use of coir pith increased (*P* < 0.05 in S1, S2, S3, and S5) FW compared to the use of wood fiber (+37% averaged over S1–5). Plant growing media containing green waste compost showed significantly (*P* < 0.01 in S1–5) higher FW compared to plant growing media comprising composted bark (+47% averaged over S1–5). Application of perlite instead of sand as inorganic material showed a positive trend (*P* < 0.05 in S3 and S5) in FW (+20% averaged over S1–5). The type of peat (black peat or white peat) did not significantly affect FW. Increasing the dose of Arabic gum significantly (*P* < 0.05 in S1–5) lowered FW (−35% averaged over S1–5). For the majority of the BCI treatments (S1, S2, S3, and S5) the use of coir pith, green waste compost, and a low dose of Arabic gum in the plant growing medium was required to reach maximal FW (Table S4). Additionally, the use of perlite was needed in BCI treatments S3 and S5.

A significant (*P* < 0.05 in S3 and S5) interaction effect occurred between the type of other organics and the dose of Arabic gum (Figures S2 and S3). Under a low dose of Arabic gum (1 kg·m<sup>−</sup>3), the use of coir pith increased FW (59.64 g in S3) compared to wood fiber (44.22 g in S3). By increasing the amount of Arabic gum in the plant growing medium (5 kg.m<sup>−</sup>3) FW dropped and the difference in FW between coir pith (35.92 g in S3) and wood fiber (33.87 g in S3) was lost.

#### *3.2. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on Lettuce Head Area*

Lettuce head area varied significantly depending on BCI (*P* < 0.001), plant growing medium (*P* < 0.001), and BCI-plant growing medium interaction (*P* < 0.001). The BCI-plant growing medium combination S3-M8 (457.24 cm2) exhibited the highest LHA, while S2-M3 (86.91 cm2) showed the lowest LHA. Bacterial community inoculum treatment resulted in significant differences in LHA in each plant growing medium (Figure S4). Treatment with BCI S3 significantly (*P* < 0.05) increased LHA compared to C in the plant growing media M3, M5, M7, and M8. For example, the treatment of plant growing medium M3 with BCI S3 (305 cm2) resulted in a more than 1.5-fold increase in LHA compared to the M3 control treatment (170 cm2). The LHA of S3-M8 (457.24 cm2), the highest LHA of all treatments, did not differ significantly from the LHA of the commercial plant growing medium (429.35 cm2). The average increase in LHA under BCI S3 treatment was 29% (*P* < 0.01; Figure S5f). Response optimization showed that treatment with BCI S3 was necessary to obtain maximal LHA (Table S5). Also, S3 treatment was the largest contributor to LHA, compared to the plant growing medium raw material groups (Figure S6).

Bacterial community inoculum S2 treatment significantly (*P* < 0.05) decreased LHA compared to Fifurdecreased LHA with 51% compared to C (358 cm2). On average, BCI S2 significantly (*P* < 0.01) decreased LHA with 33% (Figure S5f). Response optimization towards maximal LHA is reached after the removal of BCI S2 (Table S5). Additionally, BCI S2 was the largest contributor to change in LHA (absolute), compared to the plant growing medium raw material groups (Figure S7).

As also noted in lettuce FW analysis, treatment with the positive control PGPR unexpectedly decreased (*P* < 0.05) LHA compared to negative control C in M1, M3, M4, M6, M8, and M10. For example, compared to C (335 cm2), LHA decreased by 43% when M10 was treated with PGPR (189 cm2). On average, the application of PGPR showed a strong downward trend in LHA (−26%; Figure S5f).

Plant growing medium composition significantly affected LHA (Figure S5). The use of green waste compost resulted in significantly (*P* < 0.01 in S1–5) higher LHA compared to composted bark (+35% averaged over S1–5). Application of coir pith over wood fiber showed a positive trend (+16% averaged over S1–5) but was only significant (*P* < 0.01) under BCI S1 treatment (+28% under S1). The type of peat and inorganic material did not significantly affect LHA, though utilization of perlite resulted in a positive shift in LHA compared to sand (+13% averaged over S1–5). A high dose of Arabic gum significantly (*P* < 0.05 in S1–5) lowered LHA (−22% averaged over S1–5). For all BCI treatments (S1–5) the use of green waste compost and a low dose of Arabic gum in the plant growing medium were required to reach maximal LHA (Table S5). Also, the use of coir pith was needed under BCI treatment S1.

The treatment with BCI S1 showed a significant (*P* < 0.05) interaction effect between the type of other organics and composted materials (Figure S8). When plant growing media contained green waste compost, the application of coir pith increased LHA (382.68 cm2) compared to wood fiber (265.09 cm2). Changing the type of compost in the plant growing media to composted bark resulted in a decline in LHA, and the difference in LHA between coir pith (243.11 cm2) and wood fiber (223.99 cm2) vanished.

### *3.3. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on Root Fresh Weight*

Treatment of the plant growing media with BCI S3 significantly (*P* < 0.05) increased RW (+53%), while both S2 and S4 significantly (*P* < 0.05) decreased RW (−53% and −18% respectively; Figure S9). Indeed, optimization of RW response towards maximum showed that the application of BCI S3 maximizes RW, while BCI S2 and S4 have to be removed to reach maximum RW (Table S6). Both BCI S2 and S3 treatment were the largest contributors to RW, compared to the plant growing medium raw

material groups (Figures S10 and S11). The application of the positive control PGPR biostimulant resulted in a strong downward trend in RW (−49%; Figure S9).

Contrary to FW and LHA, the type of peat significantly (*P* < 0.05 in S1 and S4) affected RW (Figure S9). Application of white peat increased lettuce RW compared to black peat (+41% averaged over S1–5). For the remaining plant growing medium raw material groups, similar effects on RW were observed compared to FW and LHA (see Sections 3.1 and 3.2). However, the discerned trends were only marginally significant under certain BCI treatments. The interaction between the type of other organics and composted materials significantly (*P* < 0.05) affected RW under BCI S1 and S4 treatment (Figures S12 and S13). This interaction effect was also detected in the LHA DOE analysis (see Section 3.2). Bacterial community inoculum treatment S4 showed significant (*P* < 0.001) interaction with several plant growing medium raw material groups (peat, other organics, and inorganic materials): when the plant growing media were inoculated with S4, the observed differences in RW between the raw material group levels vanished.

#### *3.4. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on Shoot Dry Weight*

Shoot dry weight was significantly affected by BCI (*P* < 0.001), plant growing medium (*P* < 0.01), and their interaction (*P* < 0.001). Shoot dry weight varied from 4.25% DW (PGPR-M10) to 7.39% DW (S1-M7). Figure 3 shows the effect of BCI treatment on lettuce DW in each plant growing medium. Compared to C, DW rose significantly (*P* < 0.05) after treatment with BCI S1 in several plant growing media (M1, M2, M4, M7, M8, and M10). For example, the treatment of plant growing media M1 and M7 with BCI S1 (S1-M1: 7.05% DW; S1-M7: 7.39% DW) resulted in a 1.3-fold increase in DW compared to C (C-M1: 5.48% DW; C-M7: 5.80% DW). Only S1-M7 (7.39% DW) and S1-M10 (7.23% DW) showed significantly (*P* < 0.05) higher DW compared to the commercial plant growing medium (6.35% DW). On average, treatment of the plant growing media with BCI S1 significantly (*P* < 0.05) increased DW (+15%; Figure S14) and BCI S1 treatment was required to optimize DW response towards maximum (Table S7). Moreover, BCI S1 treatment was the largest contributor to DW, compared to the plant growing medium raw material groups (Figure S15).

**Figure 3.** Boxplot of shoot dry weight (%DW) grouped per plant growing medium. Letters show comparison of BCI means per plant growing medium at the 95% confidence level. S indicates the bacterial community inoculum, M indicates the plant growing medium, C indicates the negative control treatment without addition of inoculum, and PGPR indicates the positive control treatment with a Bacillus sp. inoculum. Number of plants ≥ 3.

Bacterial community inoculum S2 significantly (*P* < 0.05) lowered DW compared to C in M1, M2, M5, M6, M8, and M10 (Figure 3). For instance, treating M2 with S2 (4.54% DW) decreased DW with 18.5% compared to C (5.57% DW). The average decrease in lettuce DW caused by BCI S2 treatment was 16% (*P* < 0.01) and S2 treatment was the largest contributor to DW compared to the plant growing medium raw material groups (Figure S16). Application of the positive control PGPR biostimulant resulted in a significant (*P* < 0.01) decline in DW (−5.3% on average; Figure S14). No significant effects of the plant growing medium raw material groups on lettuce DW were observed (Figure S14).

## *3.5. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on Total Phenolic Content*

Bacterial community inoculum treatment (*P* < 0.001), plant growing medium (*P* < 0.001), and their interaction (*P* < 0.001) impacted the TPC of lettuce. Total phenolic content levels were located between 91.50 mg GAE/100 g FW (S1-M2) and 12.73 mg GAE/100 g FW (S5-M1). Significant changes in TPC were detected between BCIs in each plant growing medium (Figure S17). Bacterial community inoculum S1 significantly (*P* < 0.05) increased TPC, compared to C, in several plant growing media (M1, M3, M5, and M10). For example, treating M1 with S1 (81.73 mg GAE/100 g FW) increased TPC with 210% compared to C-M1 (26.32 mg GAE/100 g FW). Compared to the commercial plant growing medium (43.13 mg GAE/100 g FW), TPC of lettuce grown in M1, M2, M5, and M7 was significantly (*P* < 0.05) higher when inoculated with S1. On average, BCI treatment S1 (P < 0.001) and S4 (*P* < 0.01) significantly increased TPC (+65% and +26% respectively), while BCI S5 significantly (*P* < 0.05) decreased TPC (−15%) (Figure S18). Response optimization of TPC towards maximum required the addition of S1 and S4 (Table S8). Furthermore, BCI S1 and S4 treatment were the largest contributors to TPC, compared to the plant growing medium raw material groups (Figures S19 and S20).

Design of experiments analysis (Figure S18) showed a significant effect (*P* < 0.05) of the OO raw material group on lettuce TPC under BCI S1, S3, and S5: the use of wood fiber increased TPC compared to coir pith (+26.5% averaged over S1–5). Remarkably, between BCI S1 and S5, an opposite effect of the CM raw material group on TPC was observed. Application of green waste compost over composted bark showed a negative trend in TPC (−35%) under S1 treatment, while TPC increased (+34.5%) under S5. These differences in TPC are caused by a significant (*P* < 0.01) interaction effect that occurred between the type of CM and inoculation with BCI S1 or S5 (Figures S21 and S22). In the C treatment, we observed no difference in TPC between lettuce grown in plant growing media containing either composted bark (33.88 mg GAE/100 g FW) or green waste compost (32.62 mg GAE/100 g FW). Treating BCI S1 to plant growing media containing composted bark resulted in a sharp increase in TPC (73.19 mg GAE/100 g FW), while lettuce TPC (36.65 mg GAE/100 g FW) in green waste compost plant growing media did not differ from C. Contrary to this, BCI S5 treatment of plant growing media containing composted bark resulted in a decrease in TPC (18.65 mg GAE/100 g FW), while TPC (38.04 mg GAE/100 g FW) in green waste compost plant growing media did not differ from C.

The interaction between BCI S4 and the OO raw materials group showed a significant effect (*P* < 0.05) on TPC (Figure S23). In the C treatment, TPC of lettuce grown in plant growing media containing wood fiber (39.16 mg GAE/100 g FW) was higher than in coir pith plant growing media (27.37 mg GAE/100 g FW). Inoculation with S4 increased lettuce TPC in coir pith plant growing media (43.19 mg GAE/100 g FW) but did not affect wood fiber plant growing media (40.47 mg GAE/100 g FW), whereby the difference in lettuce TPC between coir pith and wood fiber plant growing media was nullified.

## *3.6. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on NO3-Content*

NO3-content was significantly impacted by BCI source (*P* < 0.001), plant growing medium (*P* < 0.001), and BCI-plant growing medium interaction (*P* < 0.001). NO3-content of all samples was well below the EU regulation limit (4000 mg/kg FW; EU 1258/2011), varying from 213 mg/kg FW (S1-M3) to 1952 mg/kg FW (S3-M8). Significant differences in lettuce NO3-content between BCIs are shown in Figure S24. Bacterial community inoculum S3 treatment significantly (*P* < 0.05) increased

NO3-content in M4, M5, M8, and M10 compared to C. For instance, NO3-content of S3-M5 (1567 mg/kg FW) was close to 5-fold higher than C-M5 (322 mg/kg FW). On average, S3 significantly (*P* < 0.05) raised NO3-content (+53%; Figure S25). Compared to C, NO3-content significantly (*P* < 0.05) decreased in multiple plant growing media when treated with BCI S1 (M1, M3, M7, M8, and M10). For example, treating M1 with S1 (253 mg/kg FW) decreased NO3-content with 84% compared to C-M1 (1626 mg/kg FW). Treatment of the plant growing media with BCI S1 significantly (*P* < 0.01) lowered NO3-content with 67% on average (Figure S25). Both BCI S1 and S3 treatment were the largest contributors to NO3-content, compared to the plant growing medium raw material groups (Figures S26 and S27). We observed no BCI-plant growing medium combinations with a significantly lower NO3-content than the commercial plant growing medium (644 mg/kg FW). Contrary, multiple treatments (C-M1, C-M7, PGPR-M2, PGPR-M4, S2-M4, S2-M8, S3-M1, S3-M2, S3-M4, S3-M5, S3-M7, S3-M8, S3-M10, S4-M8, and S4-M10) showed significantly (*P* < 0.05) higher NO3-content than the commercial plant growing medium.

Design of experiments analysis revealed that the plant growing medium raw material groups had no significant effect on NO3-content (Figure S25). Treatment with BCI S5 showed a significant (*P* < 0.05) interaction effect between the type of other organics and the Arabic gum dose (Figure S28). NO3-content of plant growing media containing coir pith was higher than wood fiber plant growing media under a low dose of Arabic gum. Contrary, an opposite shift in NO3-content was observed under a high dose of Arabic gum. Application of BCI S1 was required to minimize NO3-content, while BCI S3 should not be applied when minimizing NO3-content (Table S9).

#### *3.7. E*ff*ect of Bacterial Community Inoculum and Plant Growing Medium on Leaf Pigments*

Chlorophyll a+b was significantly affected by BCI (*P* < 0.001), plant growing medium (*P* < 0.01), and their interaction (*P* < 0.001). Chlorophyll a + b varied from 11.65 mg/100 g FW (S5-M8) to 22.67 mg/100 g FW (S5-M7). Multiple treatments (C-M5, PGPR-M3, S1-M6, S2-M2, S2-M4, S2-M6, S2-M8, S2-M9, S2-M10, S5-M8, and S5-M10) showed significantly (P < 0.05) lower Chla<sup>+</sup><sup>b</sup> levels than the commercial plant growing medium (21.48 mg/100 g FW). The effect of BCI treatment on Chla<sup>+</sup><sup>b</sup> in each plant growing medium is shown in Figure S29. Overall, no clear trends in Chla<sup>+</sup><sup>b</sup> levels, caused by BCI treatment or plant growing medium composition, were observed. However, DOE analysis revealed that BCI S2 treatment significantly (*P* < 0.05) decreased (−12%) Chla<sup>+</sup><sup>b</sup> levels compared to C (Figure S30). Bacterial community inoculum S2 showed a significant (*P* < 0.05) interaction with the type of composted materials, where no difference in Chla<sup>+</sup><sup>b</sup> levels was observed between BCI S2 and C in the composted bark plant growing media. However, BCI S2 treatment strongly decreased Chla<sup>+</sup><sup>b</sup> levels in the green waste compost plant growing media compared to C (Figure S31). Contrary, BCI S4 treatment did not affect Chla<sup>+</sup><sup>b</sup> levels in the green waste compost plant growing media compared to C. Instead, BCI S4 treatment increased Chla<sup>+</sup><sup>b</sup> levels compared to C in the composted bark plant growing media (Figure S32). Indeed, response optimization towards maximal Chla<sup>+</sup><sup>b</sup> showed that the combination of green waste compost with no BCI S2 application and BCI S4 application in combination with composted bark was optimal (Table S10).

When comparing all treatments, we observed that BCI treatment (*P*<0.001), plant growing medium composition (*P* < 0.001), and their interaction (*P* < 0.001) significantly affected lettuce carotenoid content. Carotenoid levels were located between 3.12 mg/100 g FW (S2-M4) and 4.16 mg/100 g FW (S5-M7). Carotenoid content of lettuce grown in the commercial plant growing medium (4.11 mg/100 g FW) was significantly (*P* < 0.05) higher than of lettuce from S1-M8, S2-M2, S2-M4, S2-M6, S2-M8, S2-M10, S3-M8, and S3-M9. When examining the effect of BCI treatment grouped per plant growing medium (Figure S33), no clear shifts in carotenoid levels can be distinguished. Also, DOE analysis did not show any significant effects of the plant growing medium raw material groups on carotenoid content (Figure S34). However, it was revealed that BCI S2 significantly (*P* < 0.05) decreased carotenoid content compared to C (−6%). Additionally, BCI S2 treatment was the largest contributor to lettuce carotenoid content, compared to the plant growing medium raw material groups (Figure S35). A significant (*P* < 0.05) interaction effect between composted materials and inorganic materials was observed under BCI S3 treatment (Figure S36a). When using perlite as inorganic material, carotenoid content was not affected by the type of compost. However, carotenoid content decreased when using sand in combination with green waste compost compared to composted bark. Bacterial community inoculum S3 directly interacted (*P* < 0.05) with the type of inorganic material (Figure S36b). The carotenoid content of the plant growing media containing sand was not affected by BCI S3 treatment, while BCI S3 treatment decreased carotenoid content in the plant growing media containing perlite. When treating with BCI S4, carotenoid content of the composted bark plant growing media was higher than the green waste compost plant growing media. This difference in carotenoid content was not present under C treatment (Figure S37). The type of other organics did not affect carotenoid content under a low dose of Arabic gum nor under C treatment. However, a high dose of Arabic gum or inoculation with BCI S5 increased carotenoid content in the plant growing media containing wood fiber (Figure S38a,b). No difference in carotenoid content was observed between C and BCI S5 treatment in the composted bark plant growing media. But, BCI S5 increased carotenoid levels when using green waste compost (Figure S38c). Maximizing carotenoid content depended on the BCI treatment (Table S11). Bacterial community inocula S4 and S5 were required to reach maximal carotenoid levels, while maximal carotenoid levels cannot be reached when treated with BCI S2 or S3.

#### *3.8. Principal Component Analysis*

The first two components of the PCA analysis explained 65% of the variance in the lettuce dataset (PC 1: 37.8; PC 2: 27.2%) (Figure 4). Quality of representation (cos2) values of the plant performance parameters showed that FW (96%), LHA (93%), RW (88%), and DW (71%) are well represented in PC 1 and PC 2, while the representation of NO3 (55%), TPC (39%), Chla<sup>+</sup><sup>b</sup> (34%), and carotenoids (44%) is low (Figure 4b). Correlation analysis of the yield and quality parameters (Table 3) demonstrated that FW, LHA, RW, and NO3-content were significantly positively correlated to PC 1, while TPC, Chla+b, and carotenoids were significantly negatively correlated to PC 1. Correlation analysis on PC 2 revealed that LHA, RW, DW, TPC, Chla+b, and carotenoids correlated positively and NO3-content was negatively correlated. Hereby, we observed grouping of the yield parameters (FW, LHA, and RW) along the positive PC 1 and PC 2 axis, while the quality parameters DW, TPC, Chla+b, and carotenoids were clustered towards the negative PC 1 axis and the positive PC 2 axis. NO3-content was separated from the other yield and quality parameters along the positive PC 1 and negative PC 2 axis (Figure 4a). Shoot fresh weight (18.5%), LHA (17.8%), RW (16.8%), and DW (13.6%) were the dominant variables, contributing the most to PC 1:2 (Figure 4d). The PC 1:2 contribution values of NO3 (10.7%), TPC (7.5%), Chla<sup>+</sup><sup>b</sup> (6.5%), and carotenoids (8.5%) remained below the expected average contribution.


**Table 3.** Dimension Description of the Lettuce Yield and Quality Variables to PC 1 and PC 2 at the 95% Confidence Level.

**Figure 4.** Principal component analysis (PCA) of the lettuce yield and quality variables under different BCI-plant growing medium treatments. (**a**) PCA biplot of individual samples to PC 1 and PC 2. Symbols indicate the type of plant growing medium (M1–10 and control M, the commercial plant growing medium) and colors indicate BCI treatment (S1–5, negative control C, and positive control PGPR). Ellipses denote 95% confidence interval of C, S1, S2, and S3. The plant performance parameters are shoot fresh weight (FW), lettuce head area (LHA), root fresh weight (RW), shoot dry weight (DW), total phenolic content (TPC), Nitrate content, chlorophyll a+b (Chl), and carotenoids (Carot); (**b**) Quality of representation (cos2) correlation circle of variables to PC 1 and PC 2. The color gradient indicates the quality of representation of the variables; (**c**) Contribution plot of the top 25 samples to PC 1 and PC 2. Colors are the same as in a. The dashed line indicates the expected average contribution if the contribution of the samples were uniform; (**d**) Contribution plot of variables to PC 1 and PC 2. The dashed line indicates the expected average contribution if the contribution of the variables were uniform.

Principal component analysis showed grouping of the BCI-plant growing medium samples depending on BCI treatment (Figure 4a). Plant growing media treated with BCI S3 were separated from the C treatment towards FW, LHA, RW, and NO3. Similarly, separation of BCI treatment S1 was observed towards DW, TPC, Chla+b, and carotenoids. Bacterial community inoculum S2 treated plant growing media clustered in the opposite direction of the plant performance parameters. We did not observe any clear grouping of BCI-plant growing medium samples based on plant growing medium type. The dominant treatments were S3-M8, S1-M7, S2-M4, the commercial plant growing medium, S1-M2, and S3-M10, each contributing more than 3% to PC 1:2 (Figure 4c).

#### **4. Discussion**

Reported evidence shows that plant growing media properties can enhance the beneficial impact of specific microbes on plant performance and stress resistance [41,42,46]. However, the role of plant growing medium composition and its interaction with rhizosphere bacterial communities in successful PGPR amendment and plant performance in soilless cultivation systems is not well understood. The presented study shows that microbe-plant growing medium interactions are important during the young plant stage for plant growth-promoting responses.

#### *4.1. Plant Growing Medium Constituents Have Di*ff*ering E*ff*ects on Lettuce Performance*

The five plant growing media raw material groups peat, other organics, composted materials, inorganic materials, and Arabic gum had varied effects on the tested plant performance parameters. First, changing black peat to white peat significantly increased RW (Figure S9). Mathers et al. [70] reviewed that proper plant growing medium aeration is a vital physical characteristic influencing root growth. We observed that the air volume of the white peat growing media varied from 19.33% *v*/*v* (M5) to 26.33% *v*/*v* (M2), and for the black peat growing media from 13% *v*/*v* (M3) to 17% *v*/*v* (M7) (Table S3). Brückner [71] also reported higher air volume in white peat (24% *v*/*v*) compared to black peat (17% *v*/*v*). Thus, the positive impact of white peat on air volume improved the rooting of lettuce. Although white peat improved root weight compared to black peat, this advantage did not result in increased FW. This may be caused by the fact that after transplantation, for both white and black peat plant growing media, the roots grew out of the plant growing medium into the nutrient solution, having direct access to abundant nutrients. Since PFALs require high energy input, the production of non-salable plant parts must be minimized to reduce energy consumption [72]. The black peat growing media reduced RW of lettuce without affecting shoot FW, compared to the white peat growing media. So, the use of black peat blended with alternative materials as a plant growing medium can help minimize PFAL energy consumption through reduced lettuce root mass production. Alternatively, the use of white peat combined with alternative materials may be more advantageous for crops where the root system is the prime salable plant part.

Second, the use of perlite increased FW compared to sand, with LHA and RW showing similar trends (Figure 2, Figures S5 and S9). Similar to what we observed in the peat raw material group, the increase in plant growth likely resulted from a higher air volume and water capacity of plant growing media amended with perlite, compared to sand. Perlite is commonly amended to plant growing media to increase the air-filled pore space and water-holding capacity [73]. Contrary, sand has a small water buffer and pore volume [74]. Brückner [71] observed air volumes in sand-peat growing media ranging from 14–18% and 24–27% in perlite-peat growing media. In a previous study, we reported a higher air volume in perlite mixtures (20.5% *v*/*v*) compared to sand mixtures (17.8% *v*/*v*). Moreover, the water-holding capacity of perlite mixtures (615 g.(100 g dry matter)<sup>−</sup>1) was double of that from sand mixtures (269 g.(100 g dry matter)<sup>−</sup>1) [62]. The current physical analysis also showed a higher air volume and water-holding capacity of the plant growing media amended with perlite (20.8% *v*/*v* and 604 g.(100 g dry matter)−<sup>1</sup> respectively) compared to sand amendment (17.4% *v*/*v* and 287 g.(100 g dry matter)−<sup>1</sup> respectively) (Table S3).

Third, the application of green waste compost significantly increased lettuce growth (FW, LHA, RW) compared to composted bark (Figure 2,Figures S5 and S9). Spiers and Fietje [75] reported that green waste compost EC (3.43 dS·m−1) was higher than bark compost EC (0.10 dS·m−1). The high amount of available K<sup>+</sup> in green waste compost was mainly responsible for the high EC, with amounts reported up to 916 ppm for green waste compost compared to 19 ppm for composted bark. Previously, we also observed that plant growing media amended with green waste compost have higher EC (149 <sup>μ</sup>S·cm−1) than plant growing media containing bark compost (60 <sup>μ</sup>S·cm−1), with K<sup>+</sup> levels of 228 mg·L−<sup>1</sup> and 70 mg·L−<sup>1</sup> respectively [62]. In the current study, we also observed higher EC values for green waste compost growing media, varying from 130 <sup>μ</sup>S·cm−<sup>1</sup> (M4) to 275 <sup>μ</sup>S·cm−<sup>1</sup> (M8), compared to composted bark growing media, varying from 51 <sup>μ</sup>S·cm−<sup>1</sup> (M5) to 207 <sup>μ</sup>S·cm−<sup>1</sup> (M3) (Table S3). These differences in EC values between green waste compost and bark compost growing media were related to the K+-content, respectively varying from 255.8 mg·L−<sup>1</sup> (M8) to 335.5 mg·L−<sup>1</sup> (M9), and from 84.7 mg·L−<sup>1</sup> (M5) to 122.6 mg·L−<sup>1</sup> (M6). The increased availability of salts, and especially K+, in the plant growing media amended with green waste compost proved to be advantageous for lettuce growth.

Fourth, using wood fiber over coir pith, in the other organics raw material group, decreased all plant growth parameters tested (FW, LHA, and RW) (Figure 2, Figures S5 and S9). This reduction in growth may be caused by N-immobilization, which is a known problem in wood fiber growing media [76]. To avoid N-immobilization it is necessary to apply fertilizer from the start of plant cultivation [77]. Contrary to the commercial plant growing medium, we did not apply starter fertilizer to the experimental plant growing media. Only after transplantation to the PFAL (2 weeks after sowing), plants were irrigated regularly with nutrient solution.

These examples highlight the strong variety in which different plant growing medium constituents affect the physicochemical properties of the plant growing medium and thus plant performance. Proper selection of plant growing medium raw materials is required to achieve the desired enhancement of specific plant performance parameters.

### *4.2. Microbe-Plant Growing Medium Interactions and the Bacterial Source Determine Plant Performance*

Plant growth-promoting rhizobacteria technology is becoming increasingly popular. However, there is still much doubt about the effectiveness of microbial amendment [31]. Design of experiments analysis revealed that bacterial amendment was the main driver affecting plant performance. However, the effectiveness of bacterial amendment and the plant performance parameters affected depended on microbe-plant growing medium interactions and the bacterial source.

Statistical analysis showed a significant interaction between the BCI and plant growing medium-class variables for all the tested plant performance parameters. Both BCI S1 and S3 positively affected plant performance. But, the observed effects did not occur in each plant growing medium (Figures 1 and 3), suggesting the potential influence of plant growing medium composition on the effectiveness of BCI treatment. Vandecasteele et al. [78] also reported that successful microbial inoculation depended on the type of plant growing medium. Biocontrol fungi showed better colonization in defibrated pure miscanthus, reed straw and flax shives compared to peat since peat did not provide the necessary compounds for fungal growth. Also, DOE analysis revealed several interaction effects between BCI treatment and plant growing medium constituents. For example, lettuce TPC was not affected by the type of compost under control treatment. However, inoculating the plant growing media with BCI S1 raised the TPC of lettuce grown in composted bark growing media while the TPC levels observed in the green waste compost growing media were unchanged (Figure S21). The higher organic matter content of the bark compost growing media, compared to the green waste compost growing media, may have provided a specific source of nutrients for the bacterial community present in S1 [62]. Overall, plant growing media without the BCI amendment did not perform as well as the commercial peat-coir based growing medium. We did not add starter fertilizer to the experimental plant growing media, while NPK levels of the commercialized plant growing medium were much higher, which may have caused retardation in growth (Table S3). However, we did observe that the BCI S3 amendment improved plant growth and even outperformed the commercial plant growing medium when amending BCI S3 to M8 and M10. This proves that specific microbe-plant growing medium combinations can create a synergistic effect that can outperform commercialized plant growing media.

Our results suggest that specific microbe-plant growing medium interactions determine plant performance. Moreover, bacterial amendment resulted in different effects on plant performance depending on the bacterial source. The BCIs were collected at separate locations. Bacterial community inoculum S1, S2, and S5 were collected at three different open field organic farms, while S3 and S4 were collected at different greenhouse soilless farms. Differences in cultivation method, fertilizer management, soil type, and crop species among others may have affected the composition of the collected root-associated bacterial communities. For instance, organic systems show greater microbial community diversity and higher microbial activity than conventional systems [79]. Roesti [80] concluded that the bacterial community structure varied between high and low fertilization strategies. Pii et al. [81] detected different microbial communities in two bulk soils. The recruitment of microbes from the soil to the rhizosphere is host-specific [82]. Rhizobia—legume interactions are well-studied, and their symbiosis is so specific that certain rhizobial species only interact with a selection of legumes [83].

All these parameters shape the bacterial community of the collected BCI samples, resulting in different effects on plant performance when amended to lettuce. For example, PCA analysis showed a grouping of the S3-plant growing medium combinations towards increased plant growth (FW, LHA, and RW) and NO3-content (Figure 4a). Design of experiments analysis confirmed this, showing a significant increase in lettuce FW, LHA, RW, and NO3-content under BCI S3 treatment (Figure 2, Figures S5, S9 and S25). Plant growth-promoting rhizobacteria are known to improve plant growth by enhancing nitrate uptake [84]. Because BCI S3 treatment increased lettuce NO3-content, we suspect that BCI S3 includes certain PGPRs that improve plant growth through better nutrient acquisition.

Contrary to BCI S3, we observed a separation of BCI treatment S1 towards DW, TPC, Chla+b, and carotenoids, and away from NO3 in the PCA analysis (Figure 4a). Indeed, DOE analysis showed a significant increase in DW and TPC, and a significant decrease in NO3-content (Figures S14, S18 and S25). Plants are known to produce more phenolics under N-deficient conditions [85]. Also, there is evidence that PGPR treatment can induce systemic resistance against plant pathogens, and an elevated content of phenolics is suggested to play a role [86,87]. Bacterial community inoculum S1 may contain PGPRs that induce systemic resistance as suggested by the elevation in TPC.

Both BCI S1 and S3 positively affected plant performance. Meanwhile, BCI S2 treatment resulted in negative plant performance (LHA, RW, DW, Chla+b, and carotenoids) (Figure 4a), which may indicate that BCI S2 contains plant pathogenic bacteria. Surprisingly, the PGPR biostimulant (*Bacillus* sp.), which we applied as a positive control, also reduced plant performance. Design of experiments analysis even indicated a negative effect on FW and DW (Figure 2 and Figure S14). Research suggests that the amendment of several PGPRs could be more effective than individual species due to different mechanisms being used [38,88]. Moreover, PGPR application efficacy can depend on local environmental conditions and crop specificity [26]. Our results show that a complex bacterial community is a driver for successful bacterial amendment.

#### **5. Conclusions**

In summary, the reported results display the potential of bacterial enhancement of plant growing media to modulate plant performance in horticultural systems. Plant growing medium composition determines plant performance, and successful bacterial amendment can result in improved plant performance. We revealed that bacterial amendment was a key driver affecting plant performance. Not only does the effectiveness of bacterial amendment on plant performance depend on the bacterial source, but it also depends on the interaction with the plant growing medium. Further research will focus on determining how the rhizosphere bacterial community structure is associated with the observed microbe-plant growing medium interactions, and identifying the modes of action of the PGPRs affecting plant performance.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/10/1456/s1, Figure S1: Pareto chart of the standardized effect (absolute) of the significant terms on shoot fresh weight under BCI S3 treatment, Figure S2: Interaction effects between substratum raw material groups on shoot fresh weight under BCI S3 treatment, Figure S3: Interaction effects between substratum raw material groups on shoot fresh weight under BCI S5 treatment, Figure S4: Boxplot of lettuce head area grouped per substratum, Figure S5: Main effects of substratum constituents on lettuce head area under different bacterial treatments, Figure S6: Pareto chart of the standardized effect (absolute) of the significant terms on lettuce head area under BCI S3 treatment, Figure S7: Pareto chart of the standardized effect (absolute) of the significant terms on lettuce head area under BCI S2 treatment, Figure S8: Interaction effect between other organics and composted materials on lettuce head area under BCI S1 treatment, Figure S9: Main effects of substratum constituents on root fresh weight under different bacterial treatments, Figure S10: Pareto chart of the standardized effect (absolute) of the significant terms on root fresh weight under BCI S2 treatment, Figure S11: Pareto chart of the standardized effect (absolute) of the significant terms on root fresh weight under BCI S3 treatment, Figure S12: Interaction effect between other organics and composted materials on root fresh weight under BCI S1 treatment, Figure S13: Interaction effects between substratum raw material groups on root fresh weight under BCI S4 treatment, Figure S14: Main effects of substratum constituents on shoot dry weight under different bacterial treatments, Figure S15: Pareto chart of the standardized effect (absolute) of the significant terms on shoot dry weight under BCI S1 treatment, Figure S16: Pareto chart of the standardized effect (absolute) of the significant terms on shoot dry weight under BCI S2 treatment, Figure S17: Boxplot of total phenolic content grouped per substratum, Figure S18: Main effects of

substratum constituents on total phenolic content under different bacterial treatments, Figure S19: Pareto chart of the standardized effect (absolute) of the significant terms on total phenolic content under BCI S1 treatment, Figure S20: Pareto chart of the standardized effect (absolute) of the significant terms on total phenolic content (TPC) under BCI S4 treatment, Figure S21: Interaction effects between substratum raw material groups on total phenolic content (TPC; mg GAE/100 g FW) under BCI S1 treatment, Figure S22: Interaction effects between substratum raw material groups on total phenolic content under BCI S5 treatment, Figure S23: Interaction effect between other organics and BCI on total phenolic content under BCI S4 treatment, Figure S24: Boxplot of nitrate content grouped per substratum, Figure S25: Main effects of substratum constituents on nitrate content under different bacterial treatments, Figure S26: Pareto chart of the standardized effect (absolute) of the significant terms on NO3-content under BCI S1 treatment, Figure S27: Pareto chart of the standardized effect (absolute) of the significant terms on NO3-content under BCI S3 treatment, Figure S28: Interaction effect between other organics and Arabic gum on NO3-content under BCI S5 treatment, Figure S29: Boxplot of chlorophyll a+b grouped per substratum, Figure S30: Main effects of substratum constituents on chlorophyll a+b content under different bacterial treatments, Figure S31: Interaction effect between composted materials and BCI on chlorophyll a+b content under BCI S2 treatment, Figure S32: Interaction effect between composted materials and BCI on chlorophyll a+b content under BCI S4 treatment, Figure S33: Boxplot of carotenoid content grouped per substratum, Figure S34: Main effects of substratum constituents on carotenoid content (mg/100 g FW) under different bacterial treatments, Figure S35: Pareto chart of the standardized effect (absolute) of the significant terms on carotenoid content under BCI S2 treatment, Figure S36: Interaction effects between substratum raw material groups on carotenoid content under BCI S3 treatment, Figure S37: Interaction effect between composted materials and BCI on carotenoid content under BCI S4 treatment, Figure S38: Interaction effects between substratum raw material groups on carotenoid content under BCI S5 treatment, Table S1: Control factors and level settings for substratum optimization, Table S2: The 25−<sup>2</sup> III fractional factorial design, Table S3: Physicochemical properties of the experimental substrata and the commercial substratum, Table S4: Shoot fresh weight response optimization under each BCI treatment, Table S5: Lettuce head area response optimization under each BCI treatment, Table S6: Root fresh weight response optimization under each BCI treatment, Table S7: Shoot dry weight response optimization under each BCI treatment, Table S8: Total phenolic content response optimization under each BCI treatment, Table S9: NO3-content response optimization under each BCI treatment, Table S10: Chlorophyll a+b content response optimization under each BCI treatment, Table S11: Carotenoid content response optimization under each BCI treatment.

**Author Contributions:** T.V.G.: investigation, formal analysis, data curation, visualization, writing—original draft; M.V.: supervision; N.A.: supervision; M.P.: conceptualization, project administration; M.-C.V.L.: resources, writing—review and editing; N.B.: conceptualization, funding acquisition, writing—review and editing; D.G.: conceptualization, funding acquisition, writing—review and editing, supervision. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the project grant VLAIO Baekeland mandate HBC.2017.0209.

**Acknowledgments:** The authors acknowledge Oscar Navarrete for supervision of the research and Hanne Denaeghel for assisting with the collection of the plant growing media raw materials. Seppe Top and Elsa Vancoppenolle for helping with the plant sample analysis.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **Abbreviations**



## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **Plant Nutrient Availability and pH of Biochars and Their Fractions, with the Possible Use as a Component in a Growing Media**

**Munoo Prasad 1,2,\*, Antonios Chrysargyris 2, Nicola McDaniel 3, Anna Kavanagh 3, Nazim S. Gruda <sup>4</sup> and Nikolaos Tzortzakis 2,\***


Received: 20 November 2019; Accepted: 17 December 2019; Published: 19 December 2019

**Abstract:** Biochar has the potential to be used as a growing media component, and therefore plays a role in reducing peat usage. It has unique properties apart from the ability to sequester carbon. Here we investigated the nutrient contents of four commercial biochars and their fractions. The biochars' feedstock was wood waste, except for one with paper fibres and husk. The fine or finer fractions in wood waste biochars contained higher levels of nutrients that were available to plants. The coarse fraction of the biochar derived from husk and paper fibre feedstock had a higher level of total N, P and K in contrast to the other three biochars. The pH of the finer fraction (pH of 9.08) was also higher compared with coarse fraction (pH of 8.71). It is important that when biochar a is used as a component of a peat based growing media, particle size information should be provided, as fractions from the same biochar can have different levels of total extractable nutrients and pH levels. If biochar is used to replace or reduce lime application rates of a peat-biochar mixtures, one must take into account the levels of total and extractable Ca and Mg levels, as these can vary. The variation of these elements was not only between biochars' feedstocks, even at similar pH-values, but within different fractions in the same biochar. We concluded that biochars should be characterized from the feedstock as well as from the particle size aspect, as it could have a profound effect on nutrient availability of Ca and Mg. This could lead to nutrient imbalances in cultivating plants on substrate mixtures. In addition to nutrient ratios, the suitable pH-level for a given grown species should be adjusted.

**Keywords:** peat replacement; particle size; calcium; magnesium; extractable nutrients

#### **1. Introduction**

Biochar is an organic carbon-rich solid by-product, which is gaining great interest in research for its utilization under the environmental and agricultural management [1–3]. Therefore, in addition to the common use as soil amendment material, biochar is being explored in terms of use for soil remediation [4–7], water filtration [8] and soilless substrates [9,10]. Recently there has been great interest in the use of biochar as a bioresource and growing media material [11–14]. Biochar and hydrothermal carbonization might play more important roles as constituents of growing media [15].

Biochar can be produced from several organic sources using pyrolysis under minimal oxygen supply, but also, for the use as growing media, from chunky timber waste; e.g., wood chips are suitable.

This has to be milled to particles sized generally less than 10 mm as a result of crushing, in order to be considered as a growing media component. A number of commercial producers supply biochars in different sizes. There are also limitations for the feedstocks used, as soft woody materials (e.g., from greenhouse crop waste), including stems, are not suitable due to excessive salinity levels [16]. In this case a hydrothermal carbonization process that requires only moderate temperatures and pressures is usually used [15].

A raft of publications has described biochar as a partial peat replacement over last three years [17–21], with several different organic materials playing an important role in decreasing the C footprint of the horticultural industry [15]. Peat is the principal material for container growing media in Europe, and peat production in Europe is more than 40 million m3. Peat currently represents 77%–80% of the growing media annually used in the horticultural industry in Europe [22–24]. However, peat comes from peatland ecosystems, which are very important for carbon sequestration. Peatlands are most important carbon sinks and one of the most important effective eco-systems in the terrestrial biosphere. The carbon storage in peatlands in Europe is estimated to be approximately 43,000 million tons [24]. Peat has a low pH, around pH 4, and is generally almost devoid of plant nutrients. It has the advantage that due to its low pH and nutrient levels, both pH and other nutrients can be brought to pre-determined levels for crop growth [25]. Hence, when peat is used as a substrate, stored carbon is released, negatively affecting the environment and CO2 balance [26]. For instance, Vaughn et al. [27] reported that biochar can substitute peat at levels lower than 15% (v/v), whereas higher rates derived unsatisfactory results, possible due to the high salinity and pH values, imbalance of nutrients and high C:N ratio.

Biochar has unique chemical properties. It can reduce leaching of nutrients, including nitrate [28] and P; act as a bio stimulant, especially affecting the roots and suppressing root disease [10,29,30]; reduce greenhouse gases (GHG) from growing media when ebb and flood irrigation is practiced, as anoxic conditions occur for short periods; and sequester C at the end of life of growing media, when it ultimately ends in soil [17]. At present, the costs of biochars are prohibitive. However, if during the pyrolysis process all products are used, e.g., for heat to generate electricity, bio oil for heating and using biochar as a minor component of a growing medium, and there is a gate fee for waste wood, biochar would be particularly viable. Biochar has generally a very high pH and contains nutrient elements, e.g., K, at quite high levels. These elements could be used for plant nutrition. However, only in few plant growth trials the nutrients presented in biochar were taken into account when used as a growing medium [30,31]. On the other hand, biochar's high pH was considered and the rates of lime were reduced [32,33]; or lime rate can be eliminated [34], when biochar is added. Bedussi et al. [32] used biochar that was produced by gasification instead of pyrolysis. Gasification takes place at a much higher temperature (1100 to 1200 ◦C) than pyrolysis (400 to 600 ◦C). Higher temperature has a major effect on biochar properties such as pH, total Ca and Mg levels [1,35]. Thus, products from these two above processes may not be directly comparable.

Numerous papers have evaluated biochar's positive effects on plant growth [10,11,36], but very few, such as the studies from Bedussi et al. [32], Kaudal et al. [20] and Mendez et al. [37], have characterized the biochar used in the growing media. Recently, some studies have been conducted to look at the effect of biochar particle size on physical properties of growing media [27,38]. Zaccheo et al. [33], have also shown that biochar having fine granulometry can be more effective in increasing the pH, and therefore, eliminating the need for liming of peat [32]. The nutrient properties of the particle size of biochar have been studied but only in terms of soil application [39] or for environmental use [40,41].

We are not aware of relevant studies on pH, total and extractable macronutrients and micronutrient contents of different biochar particle size fractions. Generally, quite high levels of fine fractions < 1 mm and < 2 mm have been administered, when biochar has been used in growing media [20,29,30,32,37,38]. In ad hoc trials in a few commercial biochar products with different fractions, we were surprised to find differences in pH and nutrient contents in the different fractions from the same biochar. This is in contrast to peat, as no difference in pH was reported between fine, medium and coarse peat [42]. In the

present study, we studied the pH, total and available macronutrients and micronutrient contents of biochar fractions from a number of commercial biochars that have the potential for use as growing media. The nutrient availability test that we used has been found to be very strongly correlated to plant nutrient uptake in a series of publications [43], and on this basis has been accepted as a European test [44]. We particularly looked at both total and extractable (available) Ca and Mg in relation to pH in the biochar fractions.

#### **2. Materials and Methods**

For our investigations we selected four commercial grade biochars: three from Europe and one from China. One of them derived from having cereal husk and paper fibre as a feedstock (Biochar A—DU). Three were derived from wood-based materials; namely, bamboo (Biochar B—No commercial name), wood screenings (Biochar C—Verora) and forest wood (Biochar D—Carbon Terra). Following pyrolysis by different manufactures, the biochars were chunky and had been milled to approximately less than 10 mm. The four biochars were characterized for pH [45] and electrical conductivity [46] in water extract at a 1:5 (v:v) ratio. The materials were characterised by extractable NH4-N and NO3-N; total N, K and P [44]; bulk density; and specific surface areas of the different biochars, including pore volume and diameter that were measured with a Quantachrome Autosorb-1 surface area Analyser, using N2 (Brunauer, Emmett and Teller—BET) sorption methods [47].

Each material was sieved to three fractions as < 1 mm, 1–2 mm and 2–4 mm (see Figure A1). We took care to ensure each entire fraction that passed through the screens for the various fractions was obtained. For each particle size fraction, 20 to 30 g of biochar was obtained, dried and mixed well prior to physic-chemical analyses. They were then analysed for total macronutrients and micronutrients, as described by Chrysargyris et al. [48]. Nitrogen was analysed using Kjeldahl method (Buchi Digest Automat K 439 and Distillation Kjelflex K 366, Switzerland). Other macro and micronutrients were ashed and the ash was digested with hydrochloric acid (2N HCl). Ash extracts were analysed using atomic absorption spectroscopy (PG Instruments AA 500 FG, Leicestershire, UK). The determinations of K and Na were made using a flame photometer (Lasany Model 1832, Lasany International, Haryana, India). Results are expressed in g kg−<sup>1</sup> and mg kg−<sup>1</sup> for macronutrients and micronutrients, respectively. Biochars' fractions were extracted for available nutrients using BaCl2/DTPA extraction on volume basis according to EN 13651 [44]. BaCl2 was used at the same concertation as the standard CaCl2/DTPA, EN 13651 [44]. The extracts were analysed for macro and micronutrients. The four biochar materials by three particle sizes were assessed for pH [45] in water extracts, at 1:5 (v:v) ratios, and BaCl2/DTPA extractable NH4-N and NO3-N [44,49]. Determination of BaCl2/DTPA extractable nutrients P, K, Ca, Mg, Fe, Na, Zn, Cu, Mn and B were determined in the filtered extract by inductively coupled plasma atomic emission spectrometry (Perkin Elmer ICP-OES, Waltham, MA, USA). Sulphate was measured in the extract by separating the anions by their affinity for an anion exchange resin packed in the anion separator column. The concentration of the sulphate anion was determined by measuring the conductance as it passed through conductivity cell.

#### *Statistical Methods*

Data were statistically analysed with the IBM SPSS version 22 (IBM Corp., Armonk, NY, USA) and results are expressed as means (*n* = 3) ± standard errors (SEs). Differences between treatment means were compared at *p* = 0.05 with ANOVA, followed by Duncan's multiple range test (DMRT).

#### **3. Results**

The characterization of the examined commercial, and an unscreened biochar, showed high pH, relatively high electrical conductivity, high extractable K and very low extractable N in all four biochars (Table 1). Extractable P was significantly (*p* < 0.05) higher in Biochar B compared with the other Biochars. The surface area tended to be highest with Biochar D, and lowest with Biochar B. The bulk

density was highest for Biochar D. The pore volume and pore size followed the same trend as the surface area (Table 1).

The total macronutrients and micronutrients of different fractions of four biochars are given in Figures 1 and 2, respectively. The pH values and extractable contents of elements of the three fractions are presented in Figure 2. The extractable macronutrients and extractable micronutrients of three fractions of four biochars are given in Figures 3 and 4, respectively, while selected correlation analysis among some elements are presented in Figure 5.

#### *3.1. Biochar's Total Elemental Contents and Correlations*

#### 3.1.1. Total Macronutrients N, K, Ca, Mg, P and Na

The fine fraction had significantly (*p* < 0.05) higher levels of N (25.9%, 19.0% and 18.9%) for the Biochars B, C and D, respectively, compared with the coarse fraction (Figure 1), while the opposite was found in case of the Biochar type A. The coarse fraction of 2–4 mm of Biochar A showed 26.41 mg kg−<sup>1</sup> N, which was 77% higher than the fine fraction, and that high N level was present only in Biochar A, which was the opposite to the other three biochars (Figure 1). The total K levels followed the same pattern as N, except that Biochar C that showed no significant effects of particle size (Figure 1). The biochar derived by husks and paper fibres, named Biochar A, revealed 17.50 mg kg−<sup>1</sup> of K, which was the highest content at 2–4 mm fraction. Similar to N and K, the higher levels of total P were found in Biochar B, Biochar C and Biochar D in the finer fraction. An exception was Biochar A, which had the opposite trend with higher levels in the coarser fraction (Figure 1).

The Ca levels were significantly higher in the fine (<1 mm) and medium (1–2 mm) fractions for the examined biochars, except for Biochar B, which had no significant differences between fractions (Figure 1). Similarly, to Ca, the Mg levels were higher in the fractions of < 1 mm and 1–2 mm for the Biochar A, whereas in Biochar C and Biochar D, greater Mg contents were observed in their fractions < 1 mm. The total contents of Mg in Biochar B were the same among fractions (Figure 1). No differences were evident between the fractions for the total Na content, as presented in Figure 1.

**Figure 1.** *Cont.*

**Figure 1.** The effects of biochar (A, B, C and D) particle size on total macronutrients. Non-significant differences are indicated by ns, while significant differences (*p* < 0.05) among particle size for each Biochar are indicated by different Latin letters according to Duncan's Multiple Range test. Error bars show standard errors-SE.

#### 3.1.2. Total Micronutrients Cu, Mn and Zn

Higher levels of total Cu were found in the finer fractions of two biochars—A and D. There was a clear-cut trend with the fine fraction with higher levels of Zn for the examined biochars (Figure 2). This was similar to the results that we got with extractable Zn (Figure 4). Only in Biochar C, the trend of higher levels of Mn in the fine fraction was present (Figure 2), and the results did not reflect in extractable Mn (Figure 4).

**Figure 2.** The effects of biochar (A, B, C, D) particle size on total micronutrients. Non-significant differences are indicated by ns, while significant differences (*p* < 0.05) among particle size for each Biochar are indicated by different Latin letters according to Duncan's Multiple Range test. Error bars show standard errors-SE.


is (pre-screening). Values are means ± standard errors of two measurements made on pH, electrical conductivity

**Table 1.**

Characterization

 of commercial

 biochars as

#### *3.2. Biochars' Extractable Elemental Contents*

3.2.1. The pH and Extractable Macronutrients N, K, P, Ca, Mg, Na and SO4

The pH was higher in the fine fraction for all four biochars (Figure 3). The pH levels revealed the same values for the medium (1–2 mm) and coarse (2–4 mm) fractions for Biochar A, Biochar B and Biochar D.

**Figure 3.** The effects of biochar (A, B, C, D) particle size on the pH value. Significant differences (*p* < 0.05) among particle size for each Biochar are indicated by different Latin letters according to Duncan's Multiple Range test. Error bars show standard errors-SE.

Levels of NH4-N and NO3-N were very low, with values < 2 mg L−<sup>1</sup> (data not presented). There was a clear trend of increasing K in the fine fraction (Figure 4). The two biochars, C and D, based on wood waste material, had a similar trend. Biochar B from paper fibre and cereal husk had higher (2535 mg L<sup>−</sup>1) levels of K (Figure 2).

The finer fraction (<1 mm) had significantly higher levels of extractable P in three (A, B and C) of the four biochars, while no significant differences were found among 1–2 mm and 2–4 mm fractions. Biochar D had very low values of extractable P in all the three fractions and there was no significant effect of biochar fraction (Figure 4).

Regarding extractable Ca, there was also a clear significant trend of the fine fraction containing higher levels calcium, with the woody materials, Biochars C and D, being very similar. The level of calcium in the Biochar A was high (up to 1094 mg L−<sup>1</sup> at < 1 mm), while in the Biochar B, even in the fine fraction it was very low (up to 271 mg L−<sup>1</sup> at < 1 mm) (Figure 4).

The fine fraction had significantly higher levels of magnesium but there was no significant difference between 1–2 mm and 2–4 mm. As with potassium the levels, in the biochar derived by bamboo (B), Mg was high, reaching 145.7 mg L−<sup>1</sup> of extractable Mg (Figure 4).

Only in two types of the biochars, namely, A and D, did the fine fraction have significantly higher levels of Na, while biochars B and C did not differ among the fractions examined (Figure 4). There were no significant differences between the fine and coarser fractions among the biochars in relation to extractable SO4 (Figure 4).

**Figure 4.** The effects of biochar (A, B, C, D) particle size on extractable macronutrients. Non-significant differences are indicated by ns, while significant differences (*p* < 0.05) among particle size for each Biochar are indicated by different Latin letters according to Duncan's Multiple Range test. Error bars show standard errors-SE.

#### 3.2.2. Extractable Micronutrients (Zn, Mn, Fe, B and Cu)

There was a clear significant trend of higher levels of Zn in the fine fraction for all four biochars, but the differences between 1–2 mm and 2–4 mm were not significant (Figure 5). Biochars derived by bamboo revealed the higher (4.72 mg L−1) Zn levels in the < 1 mm fraction. Manganese levels were particularly low in Biochar A and Biochar C, while both Biochar B and Biochar D revealed high (ranged from 18.82 to 20.85 mg L<sup>−</sup>1) levels of Mn in their fine fractions (Figure 5).

Higher levels of extractable Cu were found in the finer fractions of two biochars, A and D, those being the same biochars with higher levels of total Cu (Figure 2). There was no clear trend regarding the four biochars in the fractions regarding the Fe and B levels (Figure 5).

**Figure 5.** The effects of biochar (A, B, C, D) particle size on extractable micronutrients. Non-significant differences are indicated by ns, while significant differences (*p* < 0.05) among particle size for each Biochar are indicated by different Latin letters according to Duncan's Multiple Range test. Error bars show standard errors-SE.

#### *3.3. Correlations*

The pH value was correlated with total Ca (r = 0.518; Figure 6), and with total Mg (r = 0.364; Figure 6), respectively. However, the correlation was poor with extractable Ca (r = 0.272; Figure 6) and extractable Mg (r = 0.191; Figure 6). There was a strong correlation between total K and total N (r = 0.870) at level of *p* ≤ 0.01 (Figure 7). There was a very strong correlation between total Ca and total Mg (r = 0.862) at level of *p* ≤ 0.001 (Figure 7), indicating when Ca is high or low, the Mg content would follow a similar pattern. There was a very strong correlation (*p* ≤ 0.001) between extractable Ca and total Ca (r = 0.711) and total Mg (r = 0.823, Figure 7). Moreover, there was a very strong correlation (*p* = 0.001) between extractable K and total Mg (r = 0.941). There was fairly a strong correlation (*p* = 0.001) between extractable Zn and extractable SO4 (r = 0.616), extractable Fe (r = 0.602), extractable Cu (r = 0.857), respectively. Additionally, there was a strong correlation (*p* ≤ 0.001) between extractable Mn and extractable Zn (r = 0.795), and a fairly good correlation (r = 0.640) between extractable Mn and extractable Cu.

**Figure 6.** Correlation analysis of pH with total and extractable elements from four biochars over three fractions.

**Figure 7.** *Cont.*

**Figure 7.** Correlation analysis of total elements from four biochars over three fractions.

#### **4. Discussion**

It is well known that feedstock, pyrolysis temperature and residence time have a major effect on pH and nutrient content of a biochar [1,14,35]. We investigated four commercial biochars and found a high variation between them with regard to total nutrients. This variation can be partially explained on the basis of different processing conditions and differences in feedstock. However, due to the propriety nature of the materials used in the current study, exact information on processing and feedstocks are lacking. In the present study, the levels of total N, P and K were in the same range as previously reported [20,28,32,49]. Biochar A was an exception, having particularly high nutrient levels. The feedstock of this material was different from the other three, as it was based on paper fibre and cereal husk.

The finer fraction has higher levels of total macronutrients, and to a lesser extent, micronutrients, for the three-wood based biochar. The percentage of finer fraction of a biochar is of importance through the dependence of nutrients on the particle size. Regarding the biochars used in the present study, their particle sizes have been investigated previously; they have fine particles less than 1 mm in the range of 23% to 64% [30]. Biochar B in that study had 49.5% less than 1 mm. Other authors who investigated biochars derived from woody material, reported levels of less than 1 mm of 40%–81% [29], 21%–79% [37], 38%–73% [20] and 2%–98% [38]. It is most likely that the partial removal of the fine fraction (<1 mm) could make it suitable as a component of a growing media. This could be explained on the basis that different particle size of feedstock may lead to differential heating during pyrolysis and the finer fraction could have been more carbonized than the coarser fraction. Kloss et al. [35] have shown that the levels of cations Ca, Mg and K increase with greater carbonization. They, however, found that total N levels fall as carbonization increases. However, other authors have found that N levels can increase with carbonization [41,50–52].

He et al. [41] studied the chemical properties of biochar fractions of 2–5 mm, 1–2 mm and 0.5–1 mm but also seven fractions below 0.5 mm made from pine wood. In the three fractions, which were somewhat similar to our fractions, they found an increase in total N, Ca and Mg contents in the finer fraction. The authors argued that longitudinal and transverse heterogeneity of biochar and the dominant cleavage during the preparation process may be responsible for the significant differences in properties initiated by particle size. In our investigation, N, P and K of Biochar A derived from cereal husk and paper fibre, were higher in the coarser fraction.

Generally, all the extractable nutrients were higher in the fine fraction. The fact that the availability of nutrients based on extraction with BaCl2/DTPA increases as the fraction size decreases can be explained to some extent, as it is understandable, due to the greater surface area of the fine material that allows the extractant to extract more nutrients. It may also be due to higher bulk density.

In the present study, we showed that particle size of the same biochar can have different levels on pH and total and extractable macro and micro nutrients in the different fractions (i.e., higher in the fine fraction) in the context of its use as a component of a growing media. This would ultimately affect the acidity and macro and some micronutrient content and availability of peat-biochar growing media due to fact that each fraction, particularly the fine fraction has significantly different levels of nutrients and acidity.

Numerous studies have been published showing the beneficial effects on crops due to the addition of biochar to peat; however, only in a limited number of studies has the particle size of the biochar been presented. There have been a few publications that showed the particle size of the biochar has a marked effect on physical properties of the growing media [27,32,38]. Our results indicate that in the use of biochars in growing media, particle size should be taken into serious consideration not only for physical properties, but just as importantly, chemical properties. This information should be essential when formulating nutrients/lime addition and subsequent nutrient management during cropping.

The higher nutrient availability from the four biochars in fine fractions as found in the present study, are in agreement with previous reports [39]. However, Angst and Sohi [39] examined the effect of particle size in context of a soil application of biochar, which is different from the growing media conditions. They found that the availability and the release of limited number of nutrients, namely, magnesium, potassium and phosphorus, were affected by particle size with the fine fraction material showing greater water-soluble contents of these nutrients, and better release. The fraction sizes they looked at were smaller than ours, as they studied fractions of 0.15 μm to 0.60 μm, 0.60 μm to 1.8 mm, 1.8 mm to 4 mm and > 4 mm. For K and Mg, the differences between the finer fractions were relatively small but the differences were greater between the fine fractions, that was, < 4 mm, and > 4 mm. For P the differences in the finer fraction were even less and the difference between the finer fraction and > 4 mm was less than for K and Mg [39]. The extraction solution might affect the nutrients extracted derived by different fractions of biochar. Angst and Sohi [39] carried out six extractions and for K there was little change but for Mg and P the amount and the difference between the fractions increased. They used water as an extractant, unlike our extractant which had a cation in it. One is aware that biochar has a high cation exchange capacity and also has the ability to bind nitrate nitrogen, and as such a stronger extraction, which includes a cation and contains DTPA, which may give a bigger difference between the fractions and perhaps better picture on plant availability in growing media. In any case the extractant used here has been strongly correlated the plant uptake of some macro nutrients and micronutrients in a growing media situation [30,43].

Due to the nature of biochar coming from woodchips, thereby the need for crushing and relatively easy break down of particles, variation in particle size is likely to occur between batches, and this could possibly have reflected variation in physical and chemical properties of growing media, when biochar is added to peat. Our findings point to the need to be vigilant about particle size for each batch. However, the use of dolomitic lime was added not only to neutralize the acidity of peat but also to supply essential nutrients such as calcium and magnesium. The pH values in biochar can vary according to the pyrolysis temperature, as biochars produced from sewage sludge at low temperatures (300 and 400 ◦C) were acidic, whereas at high temperatures (500 and 600 ◦C), they were alkaline [53].

The levels for the Cu, Zn and Mn were in the same range as found by Altland and Locke [28] and Bedussi et al. [32]. The average extractable K, Ca and Mg as a percentage (%) of the total K, Ca and Mg for Biochar A, Biochar B, Biochar C and Biochar D were as follows: for K—Approximately 18%, 34%, 31% and 48%, respectively; for Ca—24%,16%,16% and 19%, respectively; and for Mg—14%, 9%, 11% and 28%, respectively. Angst and Sohi [39] found greater water extractability of K compared to Mg, probably reflecting the feedstock.

It is well known that most biochars have high pH-values [31], and peats have very low pH-values, ranging from 3 to 4 [54]. Therefore, peat requires additional lime, e.g., in form of a dolomitic lime, to adjust the pH to values to around 5.5, while other substrates, e.g., wood fibres, do not need it [55]. A number of researchers suggested that biochars could be used to replace lime application [30,34]. However, surprisingly, biochar-peat blends can contain up to 80% biochar without raising the pH above 7 [34]. Unfortunately, these studies do not give any information on particle size; one can surmise

it was a coarse material. Zaccheo et al. [33] reported that particle size can have an effect in increasing the pH of growing medium. Furthermore, the same authors found that the finer fraction, 0–3.3 mm, was more effective at raising the pH of a peat-biochar mixture [33]. This is in agreement with our findings, as the finer fraction had a higher pH and a higher content on Ca and Mg. However, in the same study of Zaccheo et al. [33], the premise that dolomitic lime or ordinary lime is added not only to increase the pH but also supply Ca and Mg, which are essential plant nutrients with very low levels in peat, was not taken into consideration [32]. In our study, despite the minor differences in pH, the levels of extractable Ca varied enormously between Biochar A with very high extractable Ca and Biochar B with very low extractable Ca. In addition, most biochars contain high levels of K. This could depress and reduce Mg uptake by plants, due to the cation antagonism among them. Increased K-rates, varied from 4.61–5.39 g kg−<sup>1</sup> by different biochars, were also reported by Gasco et al. [31], being in agreement with the findings of the present study.

Although generally there is good relationship between pH and total calcium and magnesium content, in our case the relationship would not be good enough to predict Ca and Mg levels. This is related to the poorer relationship of pH and available and extractable Ca and Mg content in biochars. The example of a good relationship or otherwise between Ca and Mg is important, as in peat dolomitic lime is added not only to adjust pH but to supply Ca and Mg. Extractable levels of Ca are generally very low in peat. The good relationship found in our study indicates that biochar is a similar material to dolomitic lime rather than calcitic lime. The good relationship between extractable Ca and total Ca gives confidence on the supply of Ca, whereas a poor relationship would invoke doubt, either from total Ca or extractable Ca: extractable Ca gives an indication of short term availability, while total Ca gives an indication of long term availability. P levels are too low to have any significance as a biochar is added at low rates with peat. However, K would be of importance. For instance, data from Bedussi et al. [32] shows that the finer biochar from poplar using pyrogasification has higher contents of total N, P, Ca, Mg and many micronutrients. The poplar biochar had higher contents of total Ca and Mg, despite having a lower pH. In addition, Bedussi et al. [32] measured water-soluble N, P and K which were higher for these nutrients, but found lower levels of Ca and no difference for Mg. Therefore, a blanket recommendation or a suggestion that biochar can replace lime in peat-based media needs serious rethinking. We feel it is essential that the total Ca and total Mg and available Ca and available Mg, should be analysed in biochar and taken into account when recommending lime application rates if biochar has been added to peat. This is in agreement with our previous works. As the biochar component increases in a peat/biochar-mixtures, this is affecting nutrient levels in plants; for example, the levels of Mg content in a leaf dropped due to the antagonistic effect of K on Mg [30]. This is primarily due to the excessive level of K, which is a feature of most but not all biochars. Additionally, in our resent study, K, P and Cu accumulation and Mg deficiency in cabbage leaves were related to the biochar presence and feedstock [10]. In that study, the biochar's feedstock, rate and the addition of fertilizers could affect the cabbage seedling performance.

#### **5. Conclusions**

The results of the present study have shown that within the same batch of biochar, total macronutrient and micronutrient levels are different in different fraction sizes. One can clearly state from our data that the use of a biochar when added to peat and when it is considered as a substitute for lime, due to its high pH, may not be valid in many cases, as Ca levels in biochars do not always equate to a pH value. Extractable Ca and perhaps total Ca in biochars need to be considered when advising biochar rates and lime rates to peat. Moreover, the particle size of biochar is very important regarding electrical conductivity and nutrient availability, especially K. This has far reaching consequences regarding nutrient imbalances in growing media and in formulating a base dressing and a liquid feeding programme. To our knowledge, this is the first time that it has been reported for biochar in context of growing media. The variation in total nutrients in biochar fractions of the same biochar was an unexpected result for a growing media. Our recommendation is to partially take out the biochar fraction < 1 mm if the EC levels and nutrients are particularly high and use it for other purposes. The reduction of EC and extractable K would, thus, make the biochar suitable as a component of a growing medium.

**Author Contributions:** Conceptualization, M.P. and N.T.; data curation, A.C., M.P. and N.T.; formal analysis, M.P., N.M. and A.C.; funding acquisition, A.K. and N.T.; investigation, M.P., and N.T.; methodology, M.P., A.C. and N.T.; project administration, M.P. and N.T.; resources, N.M. and A.C.; software, A.C. and N.T.; supervision, M.P. and N.T.; validation, M.P., N.S.G. and N.T.; visualization, A.C.; writing—original draft, M.P. and A.C.; writing—review and editing, M.P., N.S.G. and N.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Bord na Mona Horticulture Ltd and Cyprus University of Technology under the project OPTIBIOCHAR and Cyprus University of Technology Open Access Author Fund.

**Acknowledgments:** The authors are grateful to the project OPTIBIOCHAR that has been developed under the Cooperation Programme Cyprus-Ireland, co-funded by the Bord na Mona LtD and Cyprus University of Technology.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **Appendix A**

**Figure A1.** Biochar (A, B, C, D) particle size illustration.

## **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **Biochar Type and Ratio as a Peat Additive**/**Partial Peat Replacement in Growing Media for Cabbage Seedling Production**

## **Antonios Chrysargyris 1, Munoo Prasad 1,2,3, Anna Kavanagh <sup>3</sup> and Nikos Tzortzakis 1,\***


Received: 20 September 2019; Accepted: 23 October 2019; Published: 29 October 2019

**Abstract:** Biochar has been proposed mainly as a soil amendment, positively affecting plant growth/yield, and to a lesser degree for growing media. In this study, four commercial grade biochars (A-forest wood; B-husks and paper fiber; C-bamboo and D-fresh wood screening), mostly wood-based materials, were selected. Initial mixtures of peat (P) with different Biochar type and ratios (0-5-10-15-20%) were selected for cabbage seedling production. Biochar material had high K content and pH ≥ 8.64 which resulted in increased pH of the growing media. Biochar A and C at 20% reduced cabbage seed emergence. Biochar A, B and D maintained or improved plant growth at low ratio (i.e., 5–10%) while all Biochars increased N, K and P content in leaves. Biochars A and D were further examined at 7.5% and 15% with the addition of two doses of minerals (1-fold and 1.5-fold). Biochar A and D, initially stimulated seed emergence when compared to the control. High dose of fertilizer favored plant growth in Biochar A at 7.5% and Biochar D at 15%. Leaf stomatal conductance was decreased at Biochar A+Fert at 7.5% and Chlorophyll b content was decreased at Biochar A+Fert at 15%. The presence of Biochar A increased the antioxidant activity (as assayed by 2,2-diphenyl-1-picrylhydrazyl-DPPH). Lipid peroxidation was higher in plants grown with fertilized peat and Biochar A at 15%, activating antioxidant enzymatic metabolisms. Potassium, phosphorous and copper accumulation and magnesium deficiency in cabbage leaves were related to the Biochar presence. Wooden biochar of beech, spruce and pine species (Biochar A) at 7.5% and fertilized biochar of fruit trees and hedges (Biochar D) were more promising for peat replacement for cabbage seedling production.

**Keywords:** biochar; peat; growth; cabbage; *Brassica*; emergence

## **1. Introduction**

Biochar production is a process of dry pyrolysis of organic matter, whereby plant or animal-based organic materials are treated under high temperatures ranging from 450 to 600 ◦C, under the absence of oxygen or low oxygen conditions [1,2], while lower temperature (300 ◦C) for biochar production has been reported [3]. Primary material for biochar production is mainly wastes derived from intensive sectors such as agriculture, food, forest residues and wood industries with significant contribution to environmental management and recycling, decreasing the greenhouse gas (GHG) emission and sequester carbon [4–7]. Biochar (i.e., 70%) use in agriculture as an alternative container substrate adds value to the bioenergy process with significant reduction (up to 54%) of the cost for the use of peat-based substrates [8]. Moreover, biochar can substantially improve the soil adsorption capacity for heavy metals like Cd [9] and alleviate salinity stress in crops with significant protection of the environment [10]. Nowadays, attention has been focused on the potential biochar use in growing media formulation, attracting research interest [3].

Biochar is constantly receiving increasing attention through its usage for soil modifications, as it increases crop yields and retains or improves soil fertility [11]. At the same time, effective applications are questionable, as farmers need to combine biochar application with sustainable fertilizers and water input [5]. Biochar efficacy on yield increase was attributed to the application of the material in unfertile/barren lands, rather than to fertile soils [12] and the biochar co-application with fertilizers has been suggested [13].

Compared to the commonly used peat for growing media, biochar has high pH, increased surface area, excellent water and nutrient retention properties [14] and contains different forms of N and P (i.e., ammonium, orthophosphoric), considerable amount of K [2,15]. Moreover, biochar is highly resistant to biological degradation and preserve great longevity in soil [16]. Adding biochar in soil, it can assist to maintain nutrients, release and regulate contaminants, reduce the CO2 emission to the atmosphere, boost soil physical, chemical, and biological characteristics, and enhance microbial biomass and diversity [16–19]. Biochar particle size can affect various growing media physicochemical properties, including bulk density, total pore space and available water and air for the roots [20]. However, biochar efficacy and quality (particularly surface chemical properties and the size of the pores) relies on the feedstock and the production process [21]. Substrates with low biochar rates, i.e., 10% sewage sludge biochar in lettuce [22] and 10% wood-derived biochar in pepper and tomato [23], promoted plant growth. However, higher biochar ratio had contradicting effects with either increased plant biomass and height i.e., 60–80% conifer wood biochar in *Euphorbia* × *lomi* [24] or suppress growth i.e., 60–100% pinewood biochar in poinsettia [25] and 80–100% pinewood biochar in tomato and basil [26].

The section of seedling and potting horticultural plant production has been improved enormously over the last few years [27–29]. The ability of handling the mixtures of growing media by choosing the mineral levels and the raw or composted material is driven to a final substrate formation with desirable physicochemical properties. Biochar has shown the potential to be added in growing media, combined with various materials such as peat [3,30], compost [8,31], coir [3] and vermicomport [32]. Little information is accessible for the physiological responses of plants following biochar applications, as biochar is mainly acting as a soil conditioner and thus mitigating the effects of climate change [33].

Peat has traditionally been used as the major growing media component in Europe, followed by coir, perlite, bark, and compost [34]. Widely used for its well-known properties (high cation exchange capacity-CEC, low nutrient levels, low pH, suitable water holding capacity and air capacity), peat production in Europe exceeds 40 million m<sup>3</sup> [30,35]. However, on top of the high cost of the energy used for the extraction and transportation of peat to long distances (mainly produced at northern Europe), all these procedures are adding much to carbon footprint and increase environmental constrains. Thus, there is an increased ecological concern arising from the peat extraction, including conservation policies and identification of alternative components that could be appropriate for nursery enterprises [34,36].

Based on the favorable outcomes derived from the preliminary studies with different biochars and peat mixtures, the aim of this research was: (a) to assess the impact of biochar substitution in peat on extractable nutrients, (b) to assess four commercial biochar products as a peat diluent (growing medium) as demonstrated by plant growth, physiology and nutrient content, and (c) to evaluate the fertilizer dose and biochar ratio in peat on plant metabolism and nutrient content.

#### **2. Materials and Methods**

#### *2.1. Biochars and Plant Material*

In the current study, seeds of cabbage (*Brassica oleracea* L. var. *capitata*) were used for seedling production. Four commercial grade biochars were selected, three from Europe and one from China. They were of the following feedstocks: Forest wood e.g., beech, spruce and pine from Germany (Biochar A), husks and paper fiber wood screenings from tree branches at a ratio of 1:1 (*v*/*v*) from Germany (Biochar B), a three-year old wild high mountain bamboo (Biochar C), and fresh wood screenings (0–20 mm) from tree and shrub cuttings mainly from urban areas and farms (fruit trees, hedges, hedgerow management) from Switzerland (Biochar D). Biochars were generated using either the Pyreg equipment for Biochar B and C at 400–600 ◦C, and Biochar D at 500–600 ◦C, or Schotteredorf process for Biochar A at 700 ◦C with retention time of 15–30 min. However, owing to business sensitivity, additional data about Biochar production details is not known. A high-quality professional grade H4-H5 on von Post scale peat (P) was used as a control and as basic material to which the biochar was added. The selected biochars were assessed for their chemical characteristics [28], as for pH [37], Electrical Conductivity (EC) in water extract at 1:5 (v:v) ratio [38], and calcium chloride/DTPA (CAT) extractable (1:5 v:v) potassium (K) and phosphorus (P), ammonium (NH4-N), nitrate (NO3-N), and total extractable N (NH4-N+NO3-N) [39]. In brief, Biochar A had pH of 9.57; EC of 0.613 mS cm<sup>−</sup>1; P of 2 mg L−<sup>1</sup> and K of 1087 mg L<sup>−</sup>1; Biochar B had pH of 8.83; EC of 0.420 mS cm−1; P of 2 mg L−<sup>1</sup> and K of 376 mg L−1; Biochar C had pH of 8.64; EC of 0.450 mS cm−1; P of 21 mg L−1, K of 755 mg L−1, and NH4-N of 1 mg L−1; and Biochar D had pH of 9.55; EC of 0.410 mS cm−1; P of 3 mg L−<sup>1</sup> and K of 745 mg L<sup>−</sup>1. Biochars had negligible amount of NO3-N. Peat physicochemical characteristics have been reported previously [30]. In brief, peat had pH of 3.13; EC of 0.034 mS cm−1; NH4-N of 17 mg L<sup>−</sup>1; NO3-N of 3 mg L<sup>−</sup>1; K of 8 mg L−<sup>1</sup> and Oxygen Uptake Rate of 5.5 mmol O2 kg−<sup>1</sup> organic matter per hour.

#### *2.2. Preparation of Growing Media*

Two individual experiments were implemented in the present study. In the first experiment (Exp. I), the examined biochars mixed into the peat in different ratio. Therefore, the four biochars (A, B, C and D) were added at the rates of 0%, 5%, 10%, 15% and 20% to the peat resulting to 17 mixtures (treatments) including control treatment of peat (100% P). Then mixtures were brought to N, P and K levels (with standard fertilizers; 1-fold) to 170 mg N L−<sup>1</sup> as ammonium nitrate, 70 mg P L−<sup>1</sup> as triple superphosphate and 100 mg K L−<sup>1</sup> as potassium sulphate respectively for the peat-biochar mixtures and limed peat (dolomitic lime at 4 g L<sup>−</sup>1) and adequate amount of trace elements. The CAT extractable N, P and K that derived from the biochars were considered and the levels of fertilizers have been adjusted accordingly. There were almost insignificant amounts of N, some P and excess of K in most cases. No K was added into the mixture in case of K excess.

In the second experiment (Exp. II), the two more promising biochars and ratios were further selected for investigation with the application of additional fertilizers (1.5-fold). Therefore, the A and D biochars were selected at the rates of 0%, 7.5%, and 15% to the peat under 1-fold (N of 170 mg L<sup>−</sup>1, P of 70 mg L<sup>−</sup>1, and K of 100 mg L−1) or 1.5-fold (N of 255 mg L−1, P of 105 mg L−1, and K of 150 mg L<sup>−</sup>1) of fertilizers, resulting to 10 mixtures (treatments) including control (100% peat). Then mixtures were brought to adequate N, P and K levels, as described in Exp. I. The examined treatments and chemical analysis for both experiments are presented in Tables 1 and 2.





 (*<sup>n</sup>* 3) rows followed by the same letter are not significantly different, *p* ≤

#### *2.3. Seed Emergence*

Cabbage seeds were sown (1 cm depth) in plastic seedling trays. Each treatment had 9 and 18 modules for Exp. I and II, respectively, of 40 cm3 volume capacity each. Three seeds were placed in each module. Irrigation was performed daily with equal amount of water for all growing media, in order to cover the watering needs of the young seedlings. During seedling growth in the nursery, no fertilizers were applied. Max and min temperatures were 25 ± 2 ◦C and 20 ± 2 ◦C, respectively. Day light hours was L:D 16:8 with light flux density 300 <sup>μ</sup>mol PAR m−<sup>2</sup> <sup>s</sup>−<sup>1</sup> <sup>±</sup> 20.

A daily observation on seed emergence took place and seeds were recorded emerged when the hypocotyls were appeared. Mean emergence time (MET) was calculated as described previously [40].

#### *2.4. Vegetative Growth and Mineral Content*

Following a growing period of four to six weeks, seedlings growth-related parameters were recorded in six seedlings/treatment. Plant height and leaf number of the seedlings were measured. Leaf stomatal conductance was measured by using a ΔT-Porometer AP4 (Delta-T Devices-Cambridge, Burwell, Cambridge, UK). Leaf chlorophyll fluorescence (chlorophyll fluorometer, opti-sciences OS-30p, UK) was measured on two fully developed, light-exposed leaves per seedling. Leaves were incubated in the dark for 20 min prior to Fv/Fm measurements. Leaf chlorophyll content was assayed in six replicates/treatment either by SPAD meter or photometrically. Chlorophylls were extracted with dimethyl sulfoxide (DMSO) and Chlorophyll a (Chl a), Chlorophyll b (Chl b) and total Chlorophylls (total Chl) content was determined [28]. Seedlings were sampled above substrate surface, upper plant part was weighed (g), dried at 85 ◦C and then dry weight (g) was measured.

Mineral content in the upper part of the seedlings (including leaves and shoots) was determined on four replicates/treatment (two pooled plants/replicate). Plant tissue was dried to constant weight (at 65 ◦C for 3 day) and sub samples (~0.5 g) were ashed (at 500 ◦C for 5 h) and acid (2 N HCl) digested [41]. Nitrogen (N) content was determined with Kjeldahl (BUCHI, Digest automat K-439 and Distillation Kjeldahl K-360) digestion method. Phosphorus content was determined with spectrophotometer (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA), and K, Mg, Ca, Na, Fe, Cu, Zn, and B by an atomic absorption spectrophotometer (PG Instruments AA500FG, Leicestershire, UK) for plant tissue analysis or by inductively coupled plasma atomic emission spectrometry (ICP-AES; PSFO 2.0 (Leeman Labs Inc., Mason, OH, USA) for growing media analysis. Plant mineral content were expressed in g kg−<sup>1</sup> and mg kg−<sup>1</sup> of dry weight, for macronutrients and micronutrients, respectively. Biochar-based media minerals were expressed in mg L<sup>−</sup>1.

#### *2.5. Total Phenolics and Antioxidant Capacity*

In the Exp. II, methanolic extracts of four replicates (two pooled plants/replicate) of cabbage grown in different biochar types and ratio used for the determination of total phenolics and total antioxidant activity. The Folin–Ciocalteu method was used for the total phenolics content as described in Tzortzakis et al. [42] and results were expressed as gallic acid equivalents (mg GAE per g of fresh weight). For antioxidant capacity, two assays were used, the ferric reducing antioxidant power (FRAP) and the 2,2-diphenyl-1-picrylhydrazyl (DPPH), as described previously [43]. Results were expressed as trolox equivalents (mg trolox per g of fresh weight).

#### *2.6. Lipid Peroxidation, Hydrogen Peroxide, and Enzymes Antioxidant Activity*

Four replicates (each replicate was a poll of two plants) for each treatment were used for damage index and antioxidant enzymes activity. Lipid peroxidation and hydrogen peroxide (H2O2) content were assessed according to Loreto and Velikova [44] and De Azecedo Neto et al. [45]. The results were expressed as μmol H2O2 per g of fresh weight, while lipid peroxidation was calculated through the malondialdeyde (MDA) content (nmol of MDA per g of fresh weight).

The enzymes antioxidant activity for superoxide dismutase (SOD), for catalase (CAT) and for peroxidase activity (POD) was assayed as described previously [43]. Results were expressed as enzyme units per mg of protein. The protein content was determined by using bovine serum albumin (BSA) as a standard.

#### *2.7. Statistical Analysis*

Data were tested for normality and then statistically analyzed using analysis of variance (ANOVA) by SPSS v21.0 (SPSS Inc., Chicago, IL, USA) program. The significance of the differences between average values was based on Duncan's Multiple Range test (DMRT) at *p* ≤ 0.05, following one-way ANOVA. Values are means ± standard error (SE).

#### **3. Results**

#### *3.1. Growing Media Properties*

The biochar raw material had, in general, very high pH (ranging from 8.64 to 9.57) and considerable levels of EC (ranging from 0.410 to 0.613 mS cm−1). Therefore, adding biochars in ratios from 5% to 20% (Exp. I) increased the pH value of the acidic (pH of 4.97) peat-based material (Table 1). Moreover, biochar-based media had lower EC compared to the control (fertilized peat). The examined Biochars (A, B, C and D) had limited amounts of NH4-N and NO3-N, and this reflected the decreased levels found on the biochar-based growing media. Similar to ammonium and nitrate levels, the low (~2 mg L<sup>−</sup>1) P amounts of Biochar A, B and D reflected the decreased levels of P content in the growing media. However, Biochar C had P of 21 mg L−1, and as such, the P levels in the growing media increased for the ≥15% Biochar C. Interestingly, K levels of raw Biochars ranged from 376 to 1087 mg L−<sup>1</sup> affected the K content in the examined biochar-based growing media, and the values increased as the Biochar ratio increased from 5% to 20% (Table 1).

Following the selection of Biochars for the Exp. II, a detailed mineral composition of the examined Biochars (A and D), ratios (7.5% and 15%) and fertilizers dose (1-fold and 1.5 fold) presented in Table 2. The additional fertilizer (1.5-fold) increased, as expected, the levels of N, K, and P at the 100% fertilized peat compared with the control (P100). Growing media containing Biochar A at 15% and Biochar D at 7.5% and 15% at both fertilizers (1-fold and 1.5-fold) levels had decreased NO3-N compared to the control treatment (100% peat). The level of NH4-N increased with the presence of Biochars A and D except for the Biochar A at 7.5%, with more pronounced content at the fertilized media (Table 2). Potassium levels were increased at Biochar A-based growing media and at fertilized Biochar D media (i.e., Biochar D+Fert at 7.5% and 15%) compared to 100% peat media. Increased Ca levels were found at Biochar A-based media while Biochar D-media had reduced Ca levels. Magnesium levels were decreased in both Biochars-based media. Boron, Zn, Na and Cu levels increased in case of Biochar D presence and reduced (for Cu) in case of Biochar A. Phosphorous and Mn levels increased at the Biochar-based media, while increased P levels were found also at the 100% fertilized peat. Iron content decreased in general with the presence of Biochars A and D, with exception the Biochar D+Fert at 15% media (Table 2).

#### *3.2. Experiment I*

#### 3.2.1. Seed Emergence

In Exp. I, four biochars in four ratios were primary evaluated for cabbage seedling production. Biochar A and C at 20% reduced cabbage seed emergence compared to 100% peat (P100) as control treatment after 8 day (Figure 1A,C). Biochar B did not affect seed emergence (Figure 1B), and Biochar D (10–20%) decreased seed emergence at the first 3 day but no differences were obtained thereafter comparing with the control (Figure 1D). In general, low biochar ratios (5–10%) stimulated seed emergence for Biochar A and D compared to the control treatment for the first 3rd days. Mean germination time is shown in Figure 1E, and it was found that Biochar C and Biochar D at ≥10% delayed the seed emergence as they had higher MET comparing to control treatment (P100). Biochar A and B did not affect the MET (Figure 1E).

**Figure 1.** Cabbage cumulative seedling emergence and mean emergence time (MET) in peat with different biochar types (A, B, C, D) and ratio (0-5-10-15-20%). Biochar type is distinguished by different pattern at MET. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line present the levels of control treatment (100% peat). (**A**) is referring to Biochar A, (**B**) is Biochar B, (**C**) is Biochar C, (**D**) is Biochar D, (**E**) is mean emergence time for all biochars.

### 3.2.2. Plant Development

Biochar A and B at 5% increased but Biochar B at 20% decreased cabbage height compared to plants grown in 100% peat (Table 3). Biochar A (at 10%), Biochar B (at 5–10%), and Biochar D (at 5–10%) increased seedling fresh weight, while 20% of Biochars B and D and 10% of Biochar C decreased seedling fresh weight. Increasing Biochar ratio into the growing media, resulted in decreased plant dry weight compared to the control. The number of leaves produced did not differ among types and ratio of biochar. Biochars C and D at high levels affected negatively the cabbage root length (Table 3). Leaf chlorophyll fluorescence and content (SPAD units) were differently affected by the biochar type and ratio, with often more pronounced decreases at the higher biochars levels.




The use of biochars in the growing media significantly increased N, K and P content in cabbage leaves. Nitrogen increased (up to 80.7%, 91.3%, 103.2% and 116.7%), potassium increased (up to 284.4%, 276.5%, 459.0% and 234.8%) and phosphorus increased (up to 42.4%, 59.7%, 65.4% and 68.3%) for Biochars A, B, C and D, respectively, in relation to control (Table 3).

#### *3.3. Experiment II*

### 3.3.1. Seed Emergence

Following the Exp. I evaluation, two Biochars (A and D) under two ratio (7.5% and 15%) were further selected, including two mineral doses (1-fold and 1.5-fold). Biochar A and D improved seed emergence initially when compared to the control, while no differences were found after 4 day (Figure S1). Neither the biochar type nor the biochar ratio and applied fertilizers affected the mean emergence time for cabbage seeds.

## 3.3.2. Plant Growth and Physiology

Biochar A+Fert at 15% and Biochar D at 7.5% decreased plant height, comparing with the control treatment, while the greater plant height was found at the Biochar D+Fert at 15% (Table 4). Biochar A+Fert at 7.5% and Biochar D at 15% (independently of the fertilizers dose) increased seedling fresh weight, while Biochar A+Fert at 7.5% and Biochar D at 15% revealed increased dry weight. No differences were found on leaf number produced on biochar-based media and control (Figure S2), while the higher leaf number was found at the Biochar A+Fert at 7.5% and Biochar D+Fert at 15%.


**Table 4.** Effects of peat (P 100) with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on cabbage plant growth (height in cm, upper fresh weight in g, upper dry weight in g, root length in cm).

Values (*n* = 6) in columns followed by the same letter are not significantly different, *p* < 0.05.

Leaf stomatal conductance decreased at Biochar A+Fert at 7.5% and Chlorophyll b content decreased at Biochar A+Fert at 15%. No major differences were found on leaf SPAD measurements, the content of Chlorophyll a and total Chlorophylls in cabbage seedling subjected to different biochar types, ratios and fertilizer application (Table 5).


**Table 5.** Effects of peat (P 100) with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on leaf chlorophyll content (SPAD units), leaf stomatal conductance (cm s<sup>−</sup>1) and chlorophylls (Chl a, Chl b, total Chls) content (mg g−1).

Values (*n* = 6) in columns followed by the same letter are not significantly different, *p* < 0.05.

#### 3.3.3. Total Phenolics and Antioxidant Activity

Total phenolic content did not change much among the different treatments with the exception of Biochar D at 7.5% which revealed the highest content of phenolics (Figure 2A). Biochar A presence increased the antioxidant activity (as assayed by DPPH) of cabbage, while in case of Biochar D, DPPH increased at Biochar D at 7.5% and at Biochar D+Fert at 15% (Figure 2B). FRAP antioxidant activity revealed increased values in Biochar A at 15% and Biochar D at 7.5% (Figure 2C).

**Figure 2.** *Cont*.

**Figure 2.** Effects of peat (P 100) with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on cabbage total phenols and antioxidant activity. (**A**) total phenols, (**B**) DPPH, (**C**) FRAP. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line present the levels of control treatment (100% peat).

#### 3.3.4. Lipid Peroxidation, Hydrogen Peroxide, and Enzymes Antioxidant Activity

Lipid peroxidation (MDA) increased at 100% fertilized peat compared to the non-fertilized (control) treatment (Figure 3A). Additionally, MDA increased at 7.5% of Biochar D and for Biochar A+Fert at 15% when compared to the relevant control treatments. This increase indicates cellular damage and increased stress of the plants due to the applied treatment. The production of hydrogen peroxide increased in Biochar A and D (at 7.5% and 15%), and this increase was maintained in fertilized Biochar A, but not in fertilized Biochar D (Figure 3B). In order for the plants to detoxify the increased stress, CAT antioxidant enzymatic activity was increased for Biochar A treatments (Figure 3D). SOD activity decreased for Biochar A at 15%, Biochar D at 7.5%, Biochar D+Fert at 7.5% and Biochar D+Fert at 15%, compared to the control (Figure 3C). POD activity at the fertilized peat (PFert 100) and Biochar A+Fert at 15% maintained a similar levels as the 100% peat but decreased in all other treatments (Figure 3E).

**Figure 3.** *Cont*.

**Figure 3.** Effects of peat (P 100) with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on cabbage lipid peroxidation, hydrogen peroxide and antioxidant enzymes activity. (**A**) H2O2, (**B**) Lipid peroxidation (MDA), (**C**) SOD, (**D**) CAT, and (**E**) POD. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line present the levels of control treatment (100% peat).

#### 3.3.5. Mineral Content

The addition of Biochar A with fertilizers, as expected, increased the N accumulation in cabbage seedlings and the effects were more pronounced with 7.5% of Biochar A+Fert (Figure 4A). However, the low fertilized Biochar D reduced the N content in cabbage, while plants grown with Biochar D+Fert at 15% had increased N accumulation compared to the relevant plants grown in 100% peat. Interestingly, Biochar A increased the K accumulation in seedlings, while both Biochar ratios and fertilizer addition, increased the K accumulation. However, Biochar D needed to be fertilized and used at 15% into the mixture in order to increase the K accumulation in cabbage seedling to levels similar to the control (P-100) (Figure 4B). A similar trend to K was found for the P accumulation in the plant tissue (Figure 4C). Calcium content in cabbage increased with Biochar A at 15% (independently of the fertilizers dose) and Biochar D+Fert at 7.5% and at 15%, but was reduced with Biochar A at 7.5% (Figure 4D). Magnesium content decreased with Biochar and the effects were more pronounced in high ratio of 15% (Figure 4E). Sodium accumulation was higher with Biochar D at 15% (independent of the fertilizers dose) and lower for Biochar A at 7.5% (independent of the fertilizers dose) and 100% fertilized peat (Figure 4F). Biochar presence decreased the Fe content in cabbage while the fertilizer alleviated this effect, as Fe content was in similar levels to peat-based substrates (absence of Biochar) (Figure 4G). Copper increased with the presence of Biochar and/or fertilizers while Zn was fluctuated among the examined treatments (Figure 4H,I).

**Figure 4.** Effects of peat (P 100) with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on cabbage macro (**A**–**F**) and micronutrient (**G**–**I**) content. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line present the levels of control treatment (100% peat).

#### **4. Discussion**

Biochar can actively restore carbon to the soil, affecting environmental parameters such as carbon footprint, and therefore, is attracting research interest for a wide range of applications in the environment, agriculture and horticulture fields [46]. In the current work, biochar application was evaluated as a growth medium amendment, as different types of biochar can have different properties and cause various effects on plants. Biochar from woody feedstock with higher lignin content and higher surface area showed different sorption abilities on metals [47]; Biochar A had the best performance in the present study, and that could be a possible explanation. However, further studies are needed before final conclusions are made. Therefore, the successful application is related to the biochar type (raw material), the ratios and to the levels of fertilizers. It is known that biochar has been effectively produced from various organic materials including municipal solid wastes (garden pruning waste), agricultural (straw, greenhouse crop residues, olive-mill waste, vineyard by-products), food waste, digestate and even sewage sludge [46,48,49]. Additionally, according to reports "not all biochars are produced in the same way". Even biochars from the same source (wood-based materials), as examined in the present study, can have different impacts on plant growth and cultivation strategies, and present biochar-specific and site-specific effects on plants [50].

Biochar has mainly been studied in applications in soil but recently, during the last 10 years, there has been a big increase in research studies and publications in the area of peat substitution by biochar [22,23,36,46,51]. Biochar addition in different ratios, as presented in Exp. I, improved growing media properties, with pH increases to more adequate levels, compared to the acidic peat-based materials, for vegetables seedling production and provided considerable amount of basic nutrients, including K and P. The increased pH in the current study is in agreement with previous reports on Biochar-based material from forest waste [6], tomato crop green and wood waste [30,52], wheat straw [53], and hardwood waste [54]. Increasing the pH due to the biochar addition is an advantage for acidic soil or growing media (as it is for peat in the current work) applications, with biochar acting as a liming material and possibly replacing the calcium oxide which is used for pH increment [55,56]. However, the low biochar ratio used in the present study, maintained pH values between 5.0 and 7.0, as the ideal substrate pH for the majority of vegetables is between 5.8 and 6.8. Additionally, biochar-based media had lower EC compared to the standard fertilized peat (control), in accordance with previous studies [30]. This outcome has very significant consequences, as materials that are commonly used for peat dilution often have elevated EC levels, such as composted green waste. Those materials could be used at a higher ratio, in combination with biochar, as high EC is very often the limiting factor for these materials to be added. The EC value is an important variable for growing media preparation and stability ranged between 1.5 and 2.0 dS m−<sup>1</sup> [57]. The EC can either represent efficient nutrient support or saline conditions with adversely effects on seed germination and seedling growth [58]. However, lower initial EC values are not of consideration as substrates are commonly fertilized after plants transplanting [53]. Tailor-made fertilization is important for sustainable and successful plant growth. Therefore, increasing the fertilizers 1.5 times increased levels of minerals (i.e., K, N, P) available for the plant's growth needs. However, such nutrient enrichment can possibly create antagonistic impacts on cations such as Ca and Mg, or other effects such as increased Ca levels in Biochar A-based media and decreased Ca levels in Biochar D-media. In that case, periodical fertigation of a supplementary (hydroponic) nutrient solutions to balance the deficient levels of specific minerals could be examined. A successful case was mentioned in pot ornamental production growing in paper-waste as a substrate, supported by a hydroponic nutrient solution [29]. Previously, we had addressed the possible explanation for the decreased/low levels in nitrate and in P [30], whereas Altland and Locke [59] reported P release from biochar made from rice husks, with additional studies to be needed to explain the mechanism involved.

Seed emergence and MET in low biochar ratios (5–10%) growing media remained at similar levels with the control treatment (100% peat), while Biochars at 20% decreased seed emergence. Moreover, cabbage growing in low biochar-based media improved growth (i.e., height, fresh weight) for Biochars A, B and D. Chlorophyll fluorescence revealed low values in <15% Biochar C and in >10% in Biochars B and D, impacting the chlorophyll production, efficiency of PSII photochemistry and photosynthetic rate [28]. Increased biochar rates (i.e., 50%) resulted in decreased seed germination in myrtle and mastic seeds [6] and in tomato [30]. Solaiman et al. [1] who studied the impact of five different chars under five levels, on three plant species (wheat, mung bean, subterranean clover) indicated the early seed germination and seedling growth and this was depended on the char material and ratio. The use of biochar considerably improves seedlings' early growth [60] but some biochar may have substances that could adversely influence seed germination and early growth [1]. Seed emergence decrease was found in Biochars A and C at the rates of 20% in the present study.

Following the Exp. I, the examined Biochar ratios and types were further selected for evaluation. Biochar A and D improved seed emergence initially compared to the control. Fast and consistent seed emergence is an important issue for increased crop production, product quality, and eventually elevated profits.

In general, additional fertilizers could support plant growth with increased fresh weight at 7.5% for Biochar A and at 15% for Biochar D, observing also greater dry weight. Leaf number did not change among treatments and the decreased seedling height in case of Biochar A+Fert at 15% and Biochar D at 7.5% is not necessarily negative, as shorter (dwarf) plants are often desirable due to easy handling, transport and storage under nursery enterprises. Similar to our findings, Kim et al. [61] reported a 150% increase in shoot dry weight of kale (*Brassica olereaseae* L. var. acephala) when Biochar from rice husk was added at 5% to coir dust, perlite and verlmiculite. Vaughn et al. [62] and Steiner and Harttung [63] researched biochars for horticultural production as a substitute for peat and found no impacts on dry weight of plants. Tian et al. [51] and Mendez et al. [3] mixed biochar with compost to grow calathea and biochar with peat to grow lettuce, respectively, and revealed greater plant quality compared to those cultivated in single substrates, while Belda et al. [6] reported that the plant's response to biochar is affected by the plant species itself. No major changes were observed in plant physiology attributes in general in the present study. Leaf stomatal conductance decreased at Biochar A+Fert at 7.5% and chlorophyll b content decreased at Biochar A+Fert at 15%. The decrease in stomatal conductance and the greater water use efficiency after application of biochar shows the ability of biochar to mitigate stress from the water deficit [5].

Total phenolic content did not change much among the different treatments with the exception of the increased phenolic content in case of Biochar D at 7.5%. However, antioxidant activity increased in several cases. Interestingly, DPPH decreased when fertilizer was added with Biochar D at 7.5% but increased in case of 15% Biochar D with fertilizers, indicating the induced stress of the added minerals in the high biochar content, following MDA increment. Total phenolics and antioxidant activities increases were also found in biochar-treated *Andrographis paniculata* (kalmegh) [5]. Plants have restricted protective processes, including the production of stress response proteins and synthesis of antioxidant enzymes (includes SOD, POD and CAT) in order to overcome reactive oxygen species (ROS) accumulation [64]. The increase of MDA observed with the additionally fertilized peat (PFert 100), with Biochar A+Fert at 15% and Biochar D at 7.5% indicates cellular damage and increased stress of the plants. This was further supported with hydrogen peroxide increases and the activation of CAT antioxidant enzymes activity to detoxify the ROS accumulation [43]. The high ROS accumulation is related to intensive damage of cellular proteins, nucleic acids and lipids [65].

Although K has no direct toxicity impacts on plants, elevated K concentrations can trigger deficiencies in Mg and Ca, and plant growth reduction [66], whereas this was evident with the Biochar A mixtures that caused substantial Mg content decrease, but plant growth decrease was not observed in cabbage seedling production. Therefore, K content was increased in cabbage grown in Biochar A-based media with more pronounce effects at high ratio and/or fertilizer, while Biochar D needed to be fertilized and used in 15% into the mixture in order to obtain K levels like the control. Phosphorus accumulation followed the K trend for the examined growing media. Kim et al. [61] also reported increase of N, P and K content in kale shoots when Biochar was mixed at various ratios with

the growing media. Similarly, increased K and P contents were found when *Syngonium podophyllum* was grown in different Biochar-based media, and this was related to the higher levels of these elements in the growing media [53]. Calcium content was found to be reduced with Biochar A at 7.5% indicating antagonistic effect with the K presence. However, Ca content with 7.5% Biochar A treatment was maintained to similar levels with the control, only when fertilizers at 1.5-fold were used. The high Biochar A ratio (i.e., 15%) increased the Ca content and this reduced Mg levels. In general, Mg and Fe contents were decreased with Biochar addition and the effects were more pronounced at a high ratio of 15%. The decrease SPAD units in Biochars C and D at high ratios, reflected the leaf discoloration and the decreased Mg and Fe levels, both involved in chlorophyll metabolism.

## **5. Conclusions**

In conclusion, Biochar at a low ratio (5–10%) increased plant growth (fresh weight, height), while at 20%, it reduced cabbage seed emergence and plant height. The addition of Biochars supported the mineral accumulation in seedlings, as more available minerals could be absorbed by the plants. The production of seedlings with low height could be of benefit for nurseries, when they want to produce draft plants and where irrigation is overhead. This helps transportation and storage conditions. An increased stress occurred when a high ratio of Biochar was used (i.e., 20%), while lower ratios (5–10%) benefited plant growth-related parameters. Seeding at 20% of Biochar should be avoided as the seed emergence is decreased with higher MET. Biochars from forest wood (A) and woody feedstock (D) are quite promising materials. Finally, it seems to be preferable to use a wooden biochar of beech, spruce and pine species manufactured at 700 ◦C with the Schotteredorf process, and the produced Biochar (A) to be utilized at 7.5% ratio for cabbage seedling production. If fresh wooden biochar (D) of fruit trees and hedges are used, manufactured at 500–600 ◦C with the Pyreg equipment, then additional fertilizer is needed. However, different species need to be evaluated accordingly.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/9/11/693/s1, Figure S1: Cabbage cumulative seedling emergence in peat with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses, Figure S2: Cabbage seedling production in peat with different biochar types (A, D) and ratio (7.5%, 15%) and mineral doses.

**Author Contributions:** Conceptualization, M.P. and N.T.; methodology, M.P. and N.T.; software, N.T.; validation, A.C., M.P. and N.T.; formal analysis, A.C. and M.P.; investigation, A.C. and A.K.; resources, A.K. and N.T.; data curation, A.C. and M.P.; writing—original draft preparation, A.C.; writing—review and editing, M.P. and N.T.; visualization, A.K.; supervision, M.P. and N.T.; project administration, M.P. and N.T.; funding acquisition, A.K. and N.T.

**Funding:** This research was funded by Bord na Mona Horticulture Ltd. and Cyprus University of Technology under the project OPTIBIOCHAR and Cyprus University of Technology Open Access Author Fund.

**Acknowledgments:** The authors are grateful to the project OPTIBIOCHAR that has been developed under the Cooperation Programme Cyprus-Ireland, co-funded by the Bord na Mona LtD and Cyprus University of Technology.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **Biochar Type, Ratio, and Nutrient Levels in Growing Media A**ff**ects Seedling Production and Plant Performance**

## **Antonios Chrysargyris 1, Munoo Prasad 1,2,3, Anna Kavanagh <sup>3</sup> and Nikos Tzortzakis 1,\***


Received: 21 July 2020; Accepted: 16 September 2020; Published: 18 September 2020

**Abstract:** Biochar can be used as an alternative component in growing media, positively affecting plant growth/yield, but also media properties. In the present study, two commercial grade biochars (BFW-forest wood; and BTS-fresh wood screening), mainly wood-based materials, were used at 7.5% and 15% (*v*/*v*), adding nutrient in two levels (100% and 150% standard fertilizer level-Fert). Biochar affected growing media properties, with increases on pH and changes on the nutrient content levels. Biochar BFW enhanced the emergence of seeds in comparison to the control. Increased fertilizer levels benefited plant yield in BFW and BTS at 7.5%, but not at 15%. Leaf stomatal conductance was reduced at 150% fertilized biochars (BFW + Fert and BTS + Fert) at 7.5%, while total chlorophylls increased at BTS + Fert at 7.5% and 15%. The addition of biochars decreased the antioxidant activity in the plant. Lipid peroxidation in lettuce was increased in most cases with the presence of biochars (BFW, BTS) and 150% fertilization, activating antioxidant (superoxide oxidase and peroxidase) enzymatic metabolisms. The addition of Biochars in the growing media increased the content of nutrients in seedlings, as plants could absorb more available nutrients. Biochar of beech, spruce, and pine species (BFW) at 7.5% was more promising for substituting peat to produce lettuce seedlings. However, examining different species (tomato, leek, impatiens, and geranium) with BFW at 7.5%, the results were not common, and each species needs to be evaluated further.

**Keywords:** biochar; peat; growth; lettuce; emergence; nursery production; container; extractable nutrients; plant nutrient content

## **1. Introduction**

Biochar is produced with dry pyrolysis of the organic matter, in which plant or animal-based organic materials are subjected to high temperatures (450 to 600 ◦C), under hypoxia or anoxia environment [1,2], whereas lower temperature (300 ◦C) have been reported for biochar production [3]. The initial organic material for biochar production is mainly wastes coming from intensive sectors, such as forest residues and wood industries, agriculture and food, and greatly contribute to the environmental management and recycling, reducing the greenhouse gas (GHG) emission and increasing carbon sequestration [4–7]. The use of biochar in the agriculture sector as an alternative container growing media adds value to the bioenergy business process [8]. Moreover, biochar can alleviate salinity stress in crops with important environmental, agriculture, and economic benefit [9]. Biochar has demonstrated the potential for inclusion in growing media, together with different materials such as peat [3,10], compost [8,11], coir [3] and vermicompost [12].

Nowadays, there is a great deal of interest in the use of biochar as a peat replacement as evidenced by increasing scientific publications. This is supported by research and review articles [13–18]. The main objectives of this research have been to investigate if biochar could replace either totally or partly peat. Dilution rates as high as 50 to 75% mixed with peat or even at 100% have been tried with mixed results [19–21]. The objective of these peat replacement trials was often to show that at these high rates, biochar/peat mixtures performed as good as peat mostly as evidenced by plant height, fresh and dry weight [21,22]. Increased biochar ratio had conflicting impacts, either by increasing plant growth (higher biomass and height) i.e., 60–80% conifer wood biochar in *Euphorbia* × *lomi* [23] or by suppressing plant growth i.e., 60–100% pinewood biochar in poinsettia (*Euphorbia pulcherrima* L.) [20] and 80–100% pinewood biochar in tomato (*Solanum lycopersicum* L.) and basil (*Ocimum basilicum* L.) [24]. However, the researchers did not always take into consideration the economic viability of the use of biochar as a peat replacement. At the cost from \$67–177 per m3 (bulk density of 0.3 kg L<sup>−</sup>1) of biochar in the UK [25] and the cost of peat at \$25–30 per m3 in Europe, biochar may not be a viable substitute as a peat replacement at high ratios, unless the mixtures of peat and biochar would outperform 100% professional grade peat. Handling and incorporation of biochar as a growing media constituent would also need to be considered. There are certain extraneous advantages in mixing biochar with peat due to its properties other than based on saving on peat and crop performance. For instance, the use of biochar would lead to reduction or elimination of certain inputs such as lime and certain fertilizers, e.g., potassium [2,10,26] and thus lead to savings. Use of biochar could get carbon credits, a monetary advantage, and improve the efficiency of nutrient inputs through reduced leakage of nutrients [27] to the environment, an additional savings.

There are reports that biochar may have bio-stimulant properties due to its ability to change gene expression (i.e., transcription of auxin- and brassinosteroid-related genes in *Arabidopsis*) of the plants and the presence of gibberellic acid in biochar and these could lead to changes the morphological–physiological aspects of the plants [28,29]. Prendergast-Miller et al. [30] showed that biochar is attracting roots, resulting in its partitioning between bulk and rhizosphere soil and thus, biochar directly regulates the acquisition of plant root nutrients as a source of nutrients. Fewer studies investigated the impacts of biochar on different plant growth characteristics that affect yield, such as seed germination and the architecture of shoots or roots [1,30,31]. A few papers have applied low rates of biochar at 1–15% to peat [1,16,32,33] and have found a positive response, not only on fresh and dry weight, but also morphological and physiological changes [28,29,34]. We also know from previous publications that all biochars do not behave in a similar way ("not all biochars are made equal") and crop response depends on crop species. Therefore, biochar properties are significantly affected by feedstock, temperature, and residence time. Thus, even with the same feedstock, the properties such as surface area and pore volume will vary with temperature [35]. In addition, the application of biochar and the effects of the production of greenhouse seedlings or subsequent growth of seedlings have been less reported [34,36]. In the recent years, the field of seedling and potting horticultural plants has been significantly increased [37–39]. Since biochar from different resources has different characteristics due to potential phytotoxicity, some may have adverse effects on plant growth. Phytotoxicity evaluation is important to a successful soil/soilless amendment with bioenergy by-products such as biochar [40], and seed germination testing is a reliable procedure for biochar phytotoxicity tests.

Biochar presence in soil can modify the abundance, activity, and community structure of microorganisms in the soil. Adding cotton straw biochar (i.e., 4.5 t ha−1) in desert soil increased microbial respiration and carbon and nitrogen biomass with increases of activity of key enzymes involved in carbon and nitrogen transformation [41]. Gomez et al. [42] reported that low rates of biochars increased the microbial abundance with Gram-negative bacteria dominating the microbial community; however, high biochar rates (i.e., 49 t ha−1) inhibited microbial activity and reduced extractable phospholipid fatty acid [43]. Therefore, biochar amendment impact is greatly depending on biochar properties and soil characteristics and the possible biochar-microorganism interaction mechanisms include toxicity and emission of volatile organic compounds acting directly on soil microorganisms or indirectly by changing soil properties and enzymatic activities [44].

Biochar application can reduce N and P uptake from the plants and additional fertilization may be required for successful plant growth [34]. Moreover, the effects of biochar on plant growth may differ with plant species since different plants may have different growing conditions or different tolerances to certain stresses. The present study was conducted with two biochars from different raw materials mixed with professional grade peat at low rates (a) to assess the impact of biochar substitution in peat on extractable nutrients and phytotoxicity for lettuce (*Lactuca sativa* L.) seed emergence, (b) to assess the impact of biochar on lettuce plant growth, physiology, and nutrient content, (c) to evaluate the fertilizer dose and biochar ratio in peat on plant metabolism and nutrient content, and (d) based on the optimum biochar applications on lettuce, four more plant species were examined for their performance on biochar application.

#### **2. Materials and Methods**

#### *2.1. Biochars and Plant Material*

In the present study, lettuce (*Lactuca sativa* L. var. Nogal; Hazera, Israel) seeds were used for seedling production. Two commercial grade biochars of the following feedstocks were used: forest wood, e.g., beech, spruce, and pine from Germany (BFW), and fresh wood screenings (0–20 mm) from tree and shrub cuttings mainly from urban areas and farms (fruit trees, hedges, hedgerow management) from Switzerland (BTS). Biochars were produced using either the Pyreg equipment (Verora, Edlibach, Switzerland) for Biochar BTS at 500–600 ◦C, or Schottdorf–Meiler equipment (Carbon Terra, Wallerstein, Germany) for Biochar BFW at 700 ◦C with retention time of 15–30 min. Nonetheless, due to business sensitivity, additional data on the specifics of Biochar production are not known. A high-quality industry standard professional grade H4–H5 on von Post scale fertilized/limed peat (P) from Bord na Mona (Newbridge, Ireland) was used as a reference (control) and as basic substrate to which the biochar was added. The selected biochars had roughly the same particle size to eliminate/reduce the effect on physical properties as far as possible. Biochars were evaluated for their chemical properties [38], as for pH (EN 13,037 2002) [45], Electrical Conductivity (EC) in water extract at 1:5 (*v*/*v*) ratio (EN 13,038 2002) [46], and barium chloride/DTPA (CAT) extractable (1:5 *v*/*v*) potassium (K) and phosphorus (P), ammonium (NH4-N), nitrate (NO3-N), and total extractable N (NH4-N + NO3-N) (EN 13651-2002) [47]. Barium Chloride was substituted for Calcium Chloride at the same concentration, in order to determine Ca in the extract. The surface area determined with the Brunauer, Emmett, and Teller method [48]. Peat physicochemical characteristics have been reported previously [10]. Physicochemical properties of biochars BWF and BTS and peat are presented in Table 1.

**Table 1.** Physicochemical properties of biochars and peat.


#### *2.2. Preparation of Growing Media*

In the present study, two individual experiments were carried out. In the first experiment (Exp. I), the biochars (BFW, BTS) mixed into the peat into two ratios (7.5% and 15% *v*/*v*) and took the mixtures to N, P, and K levels (with standard fertilizers; 100%) to 170 mg N L−<sup>1</sup> as ammonium nitrate, 70 mg P L−<sup>1</sup> as triple superphosphate and 100 mg K L−<sup>1</sup> as potassium sulphate respectively for the peat-biochar mixtures and limed peat (dolomitic lime at 4 g L<sup>−</sup>1) plus addition of standard level of trace elements to all treatments (Table S1). The CAT extractable N, P, and K that derived from the biochars were considered and the levels of fertilizers have been adjusted appropriately. In most cases, there were nearly insignificant amounts of N, some P, and excess of K. No K has been added into the mixture of 100% fertilization in the case of K excess (K greater than that applied in standard fertilizer i.e., >100 mg K L<sup>−</sup>1). The application of additional fertilizers (150%) named as "+ Fert" was also examined. Therefore, the BFW and BTS biochars were selected at the rates of 0%, 7.5%, and 15% to the peat under 100% standard rate (N of 170 mg L<sup>−</sup>1, P of 70 mg L−1, and K of 100 mg L−1) or 150% standard rate (N of 255 mg L<sup>−</sup>1, P of 105 mg L−1, and K of 150 mg L−1) of fertilizers, resulting in 10 mixtures (treatments) including control (100% peat). The examined treatments, their interactions, and chemical analysis for growing media are presented in Tables 2 and 3.

**Table 2.** Effects of different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-F) on lettuce plant growth, physiology, and nutrient content.


\*, \*\*, \*\*\* Significant difference at *p* ≤ 5%, 1%, and 0.1% following three-way ANOVA. ns: non-significant.


**Table 3.** Effects of peat (P) with di fferent biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.)substratemineralscontent.

 on

In the second experiment (Exp. II), the most promising treatment of biochar was further selected for investigation and examined in four different plant species of high importance and marketability interest for seedling production. Tomato (*Solanum lycopersicum* cv. Fi Akron), leek (*Allium porrum* cv. F1 Stamford), geranium (*Pelargonium* × *hortorum* cv. Fi Horizon), and impatiens (*Impatiens walleriana* cv. F1 New Guinea Divine Orange) seeds were used. Plant growth, nutrients, and physiology-related attributes were examined.

#### *2.3. Seed Emergence*

Both Exp. I and Exp. II investigated the emergence of seeds. Seeds were sown in plastic seedling trays (1 cm depth). Each treatment had 18 modules, each with a volume of 40 cm3. Each module was seeded with three seeds. Irrigation took place daily with equal amount of potable water for all growing media. No fertilizers were applied during seedling growth. The recorded maximum and minimum temperatures were 25 ± 2 ◦C and 20 ± 2 ◦C, respectively. Day light hours was L:D 16:8 with light flux density 300 <sup>μ</sup>mol PAR m−<sup>2</sup> <sup>s</sup>−<sup>1</sup> <sup>±</sup> 20.

Seed emergence was observed daily and seeds were marked emerged with the hypocotyl's appearance. Mean emergence time (MET) was calculated as described previously [49].

#### *2.4. Plant Growth and Nutrient Content*

In six seedlings/treatments, seedlings were recorded with growth-related parameters following 4–6 weeks of plant growth. The seedling height and the number of leaved produced were recorded. The stomatal conductance of leaves was measured with a ΔT-Porometer AP4 (Delta-T Devices-Cambridge, Burwell, Cambridge, UK) [38]. Leaf chlorophyll fluorescence (chlorophyll fluorometer, opti-sciences OS-30p, UK) was measured on two fully developed, light-exposed leaves per seedling. Following leaves incubation in the dark for 20 min, the Fv/Fm ratio was measured [38]. Leaf chlorophyll content was assayed in six replicates/treatment. Chlorophyll was extracted with dimethyl sulfoxide (DMSO) and Chlorophyll a (Chl a), Chlorophyll b (Chl b) and total Chlorophylls (total Chl) content was determined [38]. Seedlings were sampled above substrate surface, the upper plant part was weighed (g), dried at 85 ◦C and then the dry weight (g) was recorded.

Nutrient content was measured in the seedling's upper part (including leaves and shoots) on four replicates/treatment (two pooled plants/replicate). Dried plant tissue (at 65 ◦C for 3 d) was used (~0.5 g) and was ashed (at 500 ◦C for 5 h) and acid (2 N HCl) digested [50]. Nitrogen (N) content was determined with Kjeldahl (BUCHI, Digest automat K-439 and Distillation Kjeldahl K-360) digestion method. Phosphorus content was determined spectrophotometrically (Multiskan GO, Thermo Fisher Scientific, Waltham, MA, USA), and K, Mg, Ca, Na, Fe, Cu, and Zn by an atomic absorption spectrophotometer (PG Instruments AA500FG, Leicestershire, UK) for plant tissue analysis or by inductively coupled plasma atomic emission spectrometry (ICP-AES; PSFO 2.0, Leeman Labs INC., Mason, OH, USA) for growing media analysis [10,38]. Plant nutrient content was expressed in g kg−<sup>1</sup> and mg kg−<sup>1</sup> of dry weight, for macronutrients and micronutrients, respectively, and growing media mineral content was expressed in mg L<sup>−</sup>1.

#### *2.5. Total Phenols Content and Antioxidant Capacity*

In the Exp. I, methanolic extracts of four replicates (two pooled plants/replicate) of lettuce grown in different biochar type and ratio were used for total phenols content and total antioxidant activity determination. Total phenols content was determined as previously described [51] and results were expressed as gallic acid equivalents (mg GAE per gram of fresh weight). For antioxidant capacity, two assays were employed, the ferric reducing antioxidant power (FRAP) and the 2,2-diphenyl-1-picrylhydrazyl (DPPH), as described previously [52]. Results were expressed as trolox equivalents (mg trolox per gram of fresh weight).

#### *2.6. Lipid Peroxidation, Hydrogen Peroxide, and Enzyme Antioxidant Activity*

In the Exp. I, four replicates (two pooled plants/replicate) of each treatment were used for the evaluation of damage index and antioxidant enzymatic activity. Lipid peroxidation (assayed through the malondialdeyde-MDA content) and hydrogen peroxide (H2O2) content were measured [53,54]. Results were expressed as μmol H2O2 per gram of fresh weight, and nmol of MDA per gram of fresh weight.

The enzymes antioxidant activity for superoxide dismutase (SOD), for catalase (CAT) and for peroxidase activity (POD) was assayed as described previously [52]. Results were expressed as enzyme units per mg of protein. The protein content was determined with bovine serum albumin (BSA), as a standard.

#### *2.7. Statistical Analysis*

A three-factor (Biochar type, Biochar rate and Fertilizer) factorial experiment was carried out. Results were statistically analyzed with a three-way analysis of variance (ANOVA) with the IBM SPSS v.22 software for Windows. The Duncan's Multiple Range test (DMRT) was used for comparing means in case of the effect of factors and their interaction, at *p* ≤ 0.05, following one-way ANOVA. Mean values ± standard error (SE) of three biological replications (*n* = 3) for growing media and of four biological replications (*n* = 4) for plant-related analysis were used.

#### **3. Results**

Table 2 presents the effects of biochar type, biochar ratio, fertilizer, and their interaction on growing media and plant-related parameters. Biochar type affected significantly growing media parameters (EC, pH, organic matter, organic carbon, K, Ca, Na, SO4, Fe, Cu, Zn, Mn, and B at *p* < 0.001; N at *p* < 0.01) and plant (Fv/Fm, MDA, N, K, P, Ca, Mg, Na, Cu, and Zn at *p* < 0.001; height, chlorophylls, FRAP, and SOD at *p* < 0.05). Biochar ratio affected significantly growing media parameters (EC, organic matter, organic carbon, N, K, Mg, Na, SO4, Fe, Zn, and Mn at *p* < 0.001; P and B at *p* < 0.01) and plant (Fv/Fm, MDA, CAT, POD, N, K, Ca, Mg, Cu, and Zn at *p* < 0.001). Fertilizer affected significantly growing media parameters (EC, organic matter, organic carbon, N, K, P, SO4, and Cu at *p* < 0.001; Ca, Fe at *p* < 0.05) and plant-related parameters (height, fresh weight, dry weight, H2O2, POD, P, Ca, Mg, Na, and Cu at *p* < 0.001; leaf number, chlorophyll b, MDA, and N at *p* < 0.01; chlorophyll a, total chlorophyll, CAT, Fe, and Zn at *p* < 0.05).

Considering the interaction of the examined factors, Biochar type × Biochar ratio (T × R) affected significantly growing media (pH, organic matter, organic carbon, N, K, Na, SO4, Fe, Zn, and Mn at *p* < 0.001; B at *p* < 0.01; EC, P and Mg at *p* < 0.05) and plant (H2O2, MDA, CAT, K, P, Na, Cu, and Zn at *p* < 0.001; FRAP at *p* < 0.01; height, stomatal conductance, chlorophylls, total phenols, DPPH, Mg, and Fe at *p* < 0.05). Biochar type × Fertilizer (T × F) affected significantly growing media (SO4, Fe, and Cu at *p* < 0.001; EC and P at *p* < 0.01) and plant (H2O2, MDA, N, K, P, Na, and Zn at *p* < 0.001; Fv/Fm and CAT, at *p* < 0.01; stomatal conductance, chlorophylls, and Cu at *p* < 0.05). Biochar rate × Fertilizer (R × F) affected significantly growing media (SO4, Fe, and Cu at *p* < 0.001; K, at *p* < 0.01) and plant (MDA, N, K, P, Na, and Zn, at *p* < 0.001; height, stomatal conductance, CAT, Ca, Mg, at *p* < 0.01; dry weight, and Cu, at *p* < 0.05). Biochar type × Biochar rate × Fertilizer (T × R × F) affected significantly growing media (P, SO4, Fe, and Cu at *p* < 0.001; K at *p* < 0.01; pH and Ca at *p* < 0.05) and plant (Fv/Fm, N, P, and Zn at *p* < 0.001; height, MDA, and K at *p* < 0.01; leaf number, SOD, and CAT at *p* < 0.05).

### *3.1. Growing Media Properties*

The growing media properties from different mixtures based on different biochars types (BFW or BTS), ratios (7.5% and 15%), and fertilizer level (100% and 150%) are shown in Table 3. The addition of NPK-fertilizer at a level of 150% increased the EC and the levels of N, K, and P at the 100% fertilized peat (P + Fert) in comparison to the control (P). Fertilized substrates (+ Fert) of 150% revealed higher

EC values compared to the 100%. The addition of BFW and BTS decreases the EC, more with the former and more at the higher rate of biochar. This was present at both rates of fertilizer (Table 3). BTS-based media had lower organic matter compared with the BFW. Adding BFW at 15% increased pH value compared to lower ratio (i.e., 7.5% BFW) or BTS-based media. The adding of BFW and BTS at 15% into the growing media decreased N content comparing to the control (peat). This was also evidenced at the 150% fertilized BFW and BTS even at lower ratio, i.e., at 7.5%, but also at the 15%. Potassium was increased at BFW-based growing media (independently of the fertilization), but decreased at the 150% fertilized BTS media (i.e., BTS + Fert at 7.5% and 15%) in comparison to the relevant control (peat or peat + Fert, respectively). Phosphorus increased at 7.5% of BFW and BTS compared to peat, increased at BTS + Fert at 15%, and decreased at BFW + Fert at 15% compared to the peat + Fert treatment. The addition of BTS into the growing media decreased the Ca, but increased the Na and B levels, independently of the fertilization and/or biochar ratios. Magnesium and sulfur levels were decreased in BFW- and BTS-based media. The addition of BFW decreased Cu while the addition of BTS increased Cu levels compared to the control. The Zn levels were increased in BTS-based media and in case of BFW + Fert at 7.5%. Iron content decreased in BFW at 15%, but increased in BTS + Fert at 15%. Manganese levels were increased in BFW-based media compared with the relevant control (peat or peat + Fert, respectively) (Table 3).

#### *3.2. Experiment I*

#### 3.2.1. Seed Emergence

Biochar BFW at 7.5% increased seed emergence after 4 days compared to control (peat). Biochar BTS did not change the emergence of lettuce seeds (Figure 1). Neither the type nor the ratio of the examined biochars and applied fertilizers (100% or 150%) affected the mean emergence time for lettuce seeds (data not shown).

**Figure 1.** Lettuce cumulative seedling emergence in peat (P) with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with Fertilizers-Fert.). Error bars show SE (*n* = 4).

#### 3.2.2. Plant Growth and Physiology

Biochar BFW and BTS either at 7% or at 15% reduced plant height, when compared with the control treatment (Table 4). The tallest seedlings were found at the BTS + Fert at 7.5% treatment. Fertilization at 150% and biochar type were affecting upper seedling fresh weight as BTS + Fert at 15% and BFW + Fert at 7.5% and at 15% decreased seedling fresh weight compared with the peat + Fert treatment. BFW + Fert at 15% and BTS + Fert at 15% decreased dry weight when compared to control and 7.5% of 150% fertilized Biochars (BFW + Fert and BTS + Fert). The number of leaves produced was similar in plants grown on biochar-based media and control (Figure S1), while the higher number of leaves was obtained at the BTS + Fert at 7.5% and relevant control (Peat + Fert) (Table 4).


**Table 4.** Effects of peat (P) with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on lettuce plant growth (height in cm, upper fresh weight in g, upper dry weight in g, root length in cm).

Values (*n* = 6) in columns followed by the same letter are not significantly different, *p* < 0.05.

The stomatal conductance of leaves was increased at BFW + Fert at 15% (Table 5). Leaf chlorophyll fluorescence decreased with the biochars (BFW or BTS) presence at both 7.5 and 15% ratios. However, fertilization at 150% increased chlorophyll fluorescence only in the case of BFW at 7.5%. Chlorophyll a content increased at BTS at 7.5% and at BTS (at 7.5 and 15%) + Fert compared to relevant controls (100% fertilized peat in the first case and 150% fertilized peat in the latter). Chlorophyll b content was also increased at BTS + Fert, which resulted in increased total chlorophylls content at BTS at 7.5% and 15% (Table 5).

**Table 5.** Effects of peat (P) with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-Fert.) on lettuce leaf stomatal conductance (cm s<sup>−</sup>1), chlorophyll fluorescence (Fv/Fm), and chlorophylls (Chl a, Chl b, total Chls) content (mg g<sup>−</sup>1).


Values (*n* = 6) in columns followed by the same letter are not significantly different, *p* < 0.05.

3.2.3. Total Phenol Content and Antioxidant Activity

Total phenol content decreased at 7.5% BFW and 15% BTS, independently of the fertilization scheme (Figure 2A). BFW presence at 7.5%, independently of the fertilization, decreased the antioxidant activity (as assayed by DPPH and FRAP) of lettuce, while in the case of BTS, DPPH and FRAP were decreased at 100% fertilized BTS at 7.5% and 15% (Figure 2B,C).

**Figure 2.** Effects of peat (P) with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-F.) on lettuce total phenols and antioxidant activity. (**A**) total phenols, (**B**) DPPH, and (**C**) FRAP. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line presents the levels of control treatment (100% peat).

#### 3.2.4. Lipid Peroxidation, Hydrogen Peroxide, and Enzyme Antioxidant Activity

The 150% fertilized peat (peat + Fert) revealed increases in the plant lipid peroxidation (MDA) when compared to the standard rate of fertilized (control) treatment (Figure 3A). Moreover, MDA increased at 7.5% and 15% BFW and at BFW + Fert at 15% in comparison to peat. In the case of BTS, MDA content increased at 7.5% and 15% BTS as well as the BTS + Fert at 7.5%, but MDA decreased at BTS + Fert at 15% compared to relevant controls. The MDA increases were followed by the increased trend of production of hydrogen peroxide in most cases (Figure 3B). Antioxidant enzymes have fluctuated among the treatments, so that the plants can detoxify the elevated stress. SOD activity increased for BTS + Fert at 7.5% and BTS at 15%, when compared to the relevant control (Figure 3C). CAT antioxidant enzymatic activity was decreased for BFW at 15% (independently of the fertilization) treatments and for BTS at 15% (Figure 3D). POD activity at the 150% fertilized peat (P + Fert) increased compared to the 100% peat (Figure 3E). POD activity decreased for the BFW at 15%, BTS at 7.5%, BTS at 15% compared to peat. Fertilized (at 150%) Biochars (BFW and BTS) decreased POD activity compared with the relevant control (peat + Fert).

**Figure 3.** Effects of peat (P) with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-F) on lettuce lipid peroxidation, hydrogen peroxide and antioxidant enzymes activity. (**A**) Lipid peroxidation (MDA), (**B**) H2O2, (**C**) superoxide dismutase (SOD), (**D**) catalase (CAT), and (**E**) peroxidase activity (POD). Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line presents the levels of control treatment (100% peat).

#### 3.2.5. Nutrient Content

The addition of 150% fertilizers in peat, increased the N, P, Mg, and Na but decreased Ca and Cu accumulation in lettuce seedlings (Figure 4A–I). The BTS at 7.5% increased further the N accumulation in lettuce, while BFW at 15%, BTS at 15%, and 150% fertilized BFW and BTS at both rations decreased the N content in lettuce in comparison to the plants grown in control (Figure 4A). BFW significantly increased the accumulation of K in seedlings, as both the Biochar ratios and the fertilizer presence, increased the K content in plants. Nonetheless, BTS decreased in general the K content in lettuce and was necessary to be fertilized at 150% and used at 15% into the growing media so to increase the K content in lettuce seedling to levels comparable to the control (peat) (Figure 4B). The P content decreased at 15% BFW and BTS at 7.5% and 15% compared to peat, while BFW + Fert and BTS + Fert resulted in decreased P content in lettuce, independently of the ratio of 7.5% or 15% used (Figure 4C). A similar trend to P was observed in the plant tissue for the Mg accumulation (Figure 4E). Calcium content in lettuce was accumulated less with BFW at 7.5% and at 15% and BTS at 7.5% compared to peat, but Ca content was increased with BTS + Fert at 15% (Figure 4D). Sodium was accumulated more in plants grown at BTS at 15% (no matter the dosage of fertilizer) and lesser at BFW at 15% (independent

of the fertilizers levels) and BFW + Fert at 7.5% compared to the relevant controls (Figure 4F). Biochar presence at 7.5% for BTS and at 15% for BFW and BTS decreased the Fe content in lettuce, while the 150% fertilization alleviated this effect, as Fe content was in comparable levels to peat-based growing media (without Biochar) (Figure 4G). Copper levels were increased with the adding of BTS and/or fertilizers, but decreased with the BFW at 7.5% (Figure 4H). A similar tendency to Cu was observed for Zn accumulation with exception the decreased Zn content with the BTS at 7.5% compared to the control treatment (Figure 4I).

## *3.3. Experiment II*

## 3.3.1. Seed Emergence

In tomato, no differences were found on seed emergence percentage and MET (Figure S2). The first seed emergence took place on day 4, while all seeds were emerged on the 5th day. In leek, seed emergence percentage increased (up to 60%) in Biochar-based media when compared to the control (100% peat) and significant differences were found after the 11th day (Figure S2). There were not any delays on the emergence time as the MET was similar for control and Biochar-based media. In geranium, a slight increase on seed emergence was found in Biochar-based media however, the effects ended up not to be significant at the end (Figure S2). No differences were found on MET among the examined growing media. In impatiens, the first seed emergence took place on the 9th day and the emergence was completed on the day 16 (Figure S2). Seed emergence on the Biochar-based media was significantly increased up to the 12th day compared with the relevant emergence on the control treatment. MET was decreased in biochar-based growing media. In tomato, seed emergence decreased in biochar-based media after the 4th day, while MET was the same for peat and biochar-based media (Figure S2).

## 3.3.2. Effects on Plant Growth, Physiology, and Nutrient Content

In tomato, BFW + Fert at 7.5% increased plant height and K accumulation, but decreased leaf stomatal conductance and P levels compared to plants grown in peat (control) (Figure 5). In leak, Biochar presence increased seedling dry weight, and the levels of chlorophylls (Chlorophyll a, Chlorophyll b and total Chlorophylls), but decreased the nutrient accumulation by decreasing N, K, and P levels. In geranium, plants grown in Biochar-enriched growing media revealed higher plant height, dry weight, and K content compared to the control treatment. In impatiens, Biochar presence increased leaf number, P, and K content, but decreased chlorophyll fluorescence, Chlorophylls content, and N content compared to the peat (control) (Figure 5).

**Figure 4.** Effects of peat (P) with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and mineral doses (with standard or with additional Fertilizers-F) on lettuce macro- and micronutrient content. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). Dotted line presents the levels of control treatment (100% peat).

**Figure 5.** *Cont*.

**Figure 5.** Effects of peat (P; light grey) with BFW at 7.5% (dark grey) with additional Fertilizers-Fert. on tomato, leek, geranium, and impatiens plant growth, physiology, and nutrient content-related parameters. Significant differences (*p* < 0.05) among treatments are indicated by different letters. Error bars show SE (*n* = 4). ns: not significant.

#### **4. Discussion**

In the present study, biochar type and ratio as well as fertilization levels were examined for lettuce seedling production and affected lettuce growth and physiology-related attributes with the levels of fertilization impacting plant performance. However, optimized biochar and fertilizer application do not have the same impacts on different plant species and selection should be on a plant-species based strategy. Biochar production is of different organic materials including urban wastes (garden pruning waste), agricultural waste (straw, residue from greenhouse crops, olive-mill waste, by-products from vineyards), food waste, digestate, and even sewage sludge [55–57], and different sources of organic material result in different biochar quality, as presented in Table 1.

In the present study, the increased pH is in line with numerous studies on Biochar-based material derived from forest waste [5], tomato crop green and wood waste [10,58], wheat straw [14], and hardwood waste [59]. The increase of pH following biochar addition is beneficial for acidic soil or growing media (as is the case for peat in the present work), with biochar serving as a liming agent and likely replacing the calcium oxide used to increase the pH [60,61]. However, there is need for caution, as we have shown that biochars of similar pH can have different levels of extractable Ca [62]. The low ratio of biochars employed in the present study retained pH values between 5.0 and 7.0, since the ideal substrate pH for peat substrate is between 5.0 and 5.5 [63]. Moreover, biochar-based media had lower EC than standard fertilized peat (control) and was related to rate of application with the higher rate (15%) to be more effective than 5% and the biochar with higher surface area being more effective, in accordance with previous reports [10]. This finding has very serious implications, as materials widely used for peat dilution have usually high levels of EC, such as compost from green waste [64]. Those materials could be used in combination with biochar at a higher ratio since high EC is commonly a limiting factor to be added to these materials. Wang et al. [65] also found that Biochar derived from woody feedstock with higher lignin content and greater surface area revealed different metal sorption capabilities. However, additional studies on biochar type and rate applied on other crops, and at different growth stage apart from seedlings, are required before final conclusions are made. We had previously discussed the possible reason for decreased/low levels of nitrate and of P [10], while Altland and Locke [66] documented P release of biochar from rice husks, with further studies required to understand the mechanism involved. There are various reasons for the retention of nitrate and phosphate in biochar amended growing media. Biochar provides refugee for soil

microorganism to influence the binding of carbon and anions [67]. Another explanation could be that due to improvement of the root milieu due to addition of biochar e.g., due to soluble organic carbon and this leads in increased microbial activity. Accelerated metabolism of soil biota turns the inorganic nitrogen into organic form, hence less available N and less N uptake [68]. The reduction of phosphate availability and uptake could be due to increased availability of Ca and Mg, due to addition of biochar [69]. Finally, increased surface area and net surface charge may also be responsible [70]. Kammann et al. [71] hypothesized that surface ageing plus non-conventional ion-water bonding in micro- and nano-pores enhanced the capture of nitrates in the biochar particles. Amending (N-rich) bio-waste with biochar may promote its agronomic value and reduce nutrient losses from bio-wastes and agricultural soils.

Biochar BFW at 7.5% stimulated seed emergence compared to the control, but this was not evidenced at the BTS. One possible explanation for that is that BFW had increased K levels compared with BTS, and priming effect of extra K on seed germination is already documented in Chinese cabbage [72]. Under nursery conditions, the consistent and fast emergence of seeds is required to meet the increased demands of healthy plant material, delivered to the producers.

Lettuce seedling height was decreased with the presence of BFW or BTS, independently of the biochar ratio, which can be of benefit for a nursery, as shorter plants can be handled, transported, and stored easier than taller plants. BFW and BTS were nutrient rich (EC ranged from 0.41–0.61 mS cm<sup>−</sup>1) materials, but in the present study, they were used in low levels (7.5–15% *v*/*v*) from one hand, while the fertilized peat used, was a more nutrient rich (higher EC) component compared to biochar-based growing media. Therefore, the reduced plant height might be attributed to the decreased available nutrients to the roots and/or due to the different particle size/surface area and extractable nutrients, i.e., Ca at the biochar-based media [62]. Therefore, the decreased plant height cannot be considered of disadvantage at all. Based on that, plant fresh weight was not affected by the biochar type and ratio used. However, when fertilization took place, plant fresh biomass decreased in both BFW and BTS, dry weight decreased at 15% fertilized biochar's media, indicating an increased stress condition due to the overloaded fertilizers particularly K, without affecting the number of leaves produced per plant. High ratio of biochar (up to 20%) and different biochar type can affect the cabbage seedling production, as biochar from bamboo or from husks and paper fiber wood screenings affected negatively plant growth and successful cabbage seedling production [34].

Growing media with low biochar rates, for example 10% sewage-sludge derived biochar in lettuce [73] and 10% wood-derived biochar in pepper (*Capsicum annuum* L.) and tomato [74], promoted plant growth. Similar to our findings based on the low biochar ratio used (i.e., 7.5%), Kim et al. [75] reported increased dry weight of kale (*Brassica olereaseae* L.) shoots with the use of low biochar levels (i.e., 5% rice husk biochar) into a coir dust-based media. However, other studies reported no effects of biochar on plant dry weight in tomato and marigold (*Tagetes erecta* L.) [22] and sunflower (*Helianthus annuus* L.) [21]. This fluctuation of results among studies might be related to the different biochar sources (wood, straw) and method of production, thus quality, the ratio used (from 5% up to 75%), and the plant species [5]. Mendez et al. [3] mixed biochar with peat to grow lettuce revealed higher quality plants in comparison to those cultivated in single substrate. Changes in chlorophyll fluorescence is affecting the efficiency of PSII photochemistry and the plant photosynthetic performance [38], and this was evidenced at biochar-based media that revealed decreased leaf chlorophyll fluorescence and plant height. Leaf stomatal conductance increased at BFW+Fert at 15%. *Andrographis paniculate* (kalmegh) herb grown in biochar-based media mitigated drought stress altering plant metabolism, decreasing the stomatal conductance and increased the water use efficiency under such conditions [7].

Total phenols content did not differ much among the examined treatments, except for the decreased total phenols content in BFW at 7.5% and BTS at 15%. Lettuce antioxidant activity (DPPH, FRAP) decreased in most cases of BWF and BTS presence indicating the reduced capacity of the plant to withstand oxidative stress and which is less appreciated and accepted by consumers and markets/industry, who are seeking added value products of high antioxidant status [76]. In contrast, total phenols content and antioxidant activity were increased in the case of high biochar ratio and/or fertilizer for cabbage, as reported by Chrysargyris et al. [34]. Even though total phenols and antioxidants remained low, lettuce plants were subjected to oxidative stress with the presence of biochars (BFW, BTS) and/or fertilizers as revealed by the increased levels of MDA, indicating lipid peroxidation and cellular damage. This was further supported with the increased levels of H2O2 and the activation of SOD and POD antioxidant enzymes to detoxify the reactive oxygen species (ROS) accumulation [52]. Plants responded to oxidative stress by activation of protecting mechanism, producing stress response proteins and antioxidant enzymes (including SOD, POD, and CAT) to resolve the accumulation of ROS [77].

Increased K levels are not directly toxic to plants however, the increased K levels can cause antagonism and resulted in Mg and Ca deficiencies with decreases in plant growth [78]. In the present study, decreased Mg levels in lettuce were evidenced in biochar-based media, those media that had lower Mg and higher K content compared to control media (peat). Similarly, lower Mg was recorded when Biochars was added to a peat growing media [10]. Moreover, K accumulated in lettuce produced in BFW-based media with more pronounced effects at high ratio and/or fertilizer, whereas BTS had to be fertilized at 150% and mixed at 15% to obtain K levels such as control. Similarly, when *Syngonium podophyllum* was grown in various Biochar-based media, increased K content was found and this was related to the higher level of this element in the growing media [14]. Fertilized (150%) biochar-based media and BFW and BTS at high ratio, decreased N and P content in lettuce, being in accordance with findings of our previous studies [10]. In contrast, Kim et al. [75] documented increase contents of N and P in kale shoots when Biochar was added at different ratios with the growing media. Calcium content was decreased in lettuce grown in BFW (at 7.5–15%) and BTS (at 7.5%) -based media and fertilization at 150% were needed to overcome this decrease. Biochar presence in general decreased the Fe content in lettuce but fertilization alleviated this effect, as the content of Fe was similar to that of peat-based substrates (without Biochar). Copper and Zn content in lettuce were increased by adding BTS and/or fertilizers in the growing media with more profound effect at the higher ratio of 15%, and this is reflecting the increased Cu and Zn levels into the growing media.

Following selection of BFW at 7.5% with 150% fertilization, seed emergence was improved in leek and impatiens, but not in tomato and geranium. In tomato, geranium, and impatiens BFW + Fert at 7.5% increased several plant-growth related parameters and nutrient accumulation, mainly of K compared to the control. However, in some cases, such as tomato, leaf stomatal conductance decreased in plants grown in biochar-based media.

#### **5. Conclusions**

In conclusion, Biochar increased plant growth (biomass, height) at a low ratio (7.5%), while it reduced the emergence of lettuce seed and plant height at 15%. The addition of Biochars provided nutrients in the seedlings, because the plants could absorb more available nutrients. Production of low-weight seedlings may be beneficial to nurseries when they want to produce dwarf plants and overhead irrigation. This helps to conditions for transport and storage. It seems better to use a wooden biochar of beech, spruce, and pine species produced at 700 ◦C with the Schotteredorf process and to use the resulting Biochar (BFW) at a ratio of 7.5% for the production of lettuce seedlings. Different species however need to be assessed accordingly. These results showed clearly the ability of biochars to reduce EC depending on rate of application and biochar surface area. This finding is very significant as most materials used to dilute peat e.g., composted greenwaste have high EC and the limiting factor on the rate of peat dilution are their high EC. These results also clearly showed that any investigation into the suitability of adding biochar to peat as peat replacement and/or biostimulant, must consider the nutrient implications of this addition to the plant. In addition, bringing the nutrient to the levels based on calculation e.g., peat growing media, may not be enough as there is a strong interaction between biochar and N and P availability as based on our substrate analysis and plant nutrient uptake data. This area of work needs attention when experiments are conducted to evaluate biochar as an addition to peat and to other growing media.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/9/1421/s1, Table S1: Growing media composition. Figure S1: Lettuce seedling production in peat with different biochar types (BFW, BTS) and ratio (7.5%, 15%) and nutrient doses. Figure S2. Effects biochar BFW at 7.5% with additional Fertilizers-Fert. on tomato, leek, geranium, and impatiens seed emergence and mean emergence time.

**Author Contributions:** Conceptualization, M.P. and N.T.; methodology, M.P. and N.T.; software, N.T.; validation, A.C., M.P., and N.T.; formal analysis, A.C. and M.P.; investigation, A.C. and A.K.; resources, A.K. and N.T.; data curation, A.C. and M.P.; writing—original draft preparation, A.C.; writing—review and editing, M.P. and N.T.; visualization, A.K.; supervision, M.P. and N.T.; project administration, M.P. and N.T.; funding acquisition, A.K. and N.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Bord na Mona Horticulture Ltd. and Cyprus University of Technology under the project OPTIBIOCHAR.

**Acknowledgments:** The authors are grateful to the project OPTIBIOCHAR that has been developed under the Cooperation Programme Cyprus-Ireland, co-funded by the Bord na Mona LtD and Cyprus University of Technology. Moreover, we are grateful to the reviewer for their substantial and critical contribution during the review/evaluation process.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Utilization of Olive Oil Processing Waste Composts in Organic Tomato Seedling Production**

**Yüksel Tüzel 1,\*, Kamil Ekinci 2, Gölgen Bahar Öztekin 1, ˙ Ibrahim Erdal 2, Nurhan Varol <sup>3</sup> and Özen Merken <sup>4</sup>**


Received: 6 May 2020; Accepted: 31 May 2020; Published: 4 June 2020

**Abstract:** Olive oil byproducts show differences according to the olive oil extraction systems, which are called olive mill solid wastes, olive oil wastewater and olive oil wastewater sludge. Three different kinds of composts, including two-phase and three-phase olive mill solid wastes, and olive oil wastewater sludge were produced with separated dairy manure, poultry manure, and straw. The composts obtained from two-phase and three-phase olive mill solid wastes and olive oil wastewater sludge were named as two-phase, three-phase, and water sludge composts, respectively. They were separately enriched by rock phosphate and potassium salt. These composts were mixed with peat in a ratio of 0%, 25%, 50%, 75%, and 100% (*v*/*v*). Tomato seeds were sown in all mixtures on 3 February 2016. All the seeds were sown into 2 trays and each plug included 2 replicates. The trays were left in a germination room for 3 days, then moved to a heated greenhouse which is specialized for growing seedlings, and the seedlings were grown there for 3 weeks. The results showed that increasing compost ratios in the growing medium and also the enrichment of the growing medium increased organic matter content, electrical conductivity, and macro and micro nutrient concentrations. The germination period lasted longer with increasing compost ratios. The shoot length was lower at a compost ratio of over 50% excluding water sludge compost, which reacted to over 75%. The highest plant dry weights were obtained in the plants grown on the media with compost ratios of 50%, 25%, and 25% for water sludge compost, enriched two-phase compost, and enriched three-phase compost, respectively. We concluded that the composts obtained from two-phase and three-phase olive mill solid wastes and olive oil waste water sludge can be used without any need of enrichment and a ratio of 25% was found appropriate in most of the measured properties.

**Keywords:** *Solanum lycopersicum*; olive oil waste; two-phase; three-phase; water sludge

## **1. Introduction**

Turkey is one of the most important olive- and olive oil-producing countries among the Mediterranean countries with a production of 1,500,467 tonnes of olive in 2018 [1]. The share of organic olive production among the total production in 2018 was 14.22% [2]. Two-phase or three-phase olive oil processing systems are used for the extraction of olive oil and both systems generate large amounts of by-products, which are called two-phase and three-phase olive mill solid wastes, olive oil waste water and olive oil wastewater sludge [3]. In Turkey, a survey study showed that all the shares of producers running three-phase, two-phase, and traditional cold stone pressed olive oil production systems were 71%, 27%, and 2%, respectively [4].

Olive oil production produces a large amount of solid and liquid wastes each year. Three-phase olive mill solid wastes contain broken seeds of olive. Olive mill wastewater contains 83%–96% water, 3.5%–15% organic matter, and 0.5%–2.0% mineral salts, depending on factors such as olive varieties, growing conditions, soil and climatic conditions, extraction methods, etc. [5]. Both effluents pose environmental problems since they exhibit highly phytotoxic and antimicrobial properties mainly due to phenols and they are not easily biodegradable [6–8]. Therefore, olive processing wastes have been considered as soil and water pollutants and cannot be used directly for agricultural purposes [7]. Within the framework of the measures taken by the Ministry of Environment and Urbanization of the Republic of Turkey, it is recommended that factory owners accumulate olive mill wastewater in lagoons or open ponds, evaporate their water, and utilize olive oil waste water sludge as the least risky solution for the environment. Additionally, factories should convert their processing systems into a two-phase system. Chowdhury et al. [9] reported that two-phase systems produce a lignocellulosic olive humid husk, which is a watery solid by-product with high contents of water (56.6%–74.5%) and phenols (0.62%–2.39%).

As a result, it is necessary to utilize solid and liquid wastes from both systems. Numerous researchers indicate that composting of olive oil production wastes with manure and some other organic materials is the best way of recycling as agricultural material [10,11]. The composted olive oil processing solid waste can be utilized as organic inputs for soil fertility and plant nutrition in agricultural production.

Fertilization is the most important input necessary for the conservation and maintenance of soil fertility in crop production in organic agriculture. On the other hand, with the growth in agricultural production, the amount of organic wastes arising from agriculture-based industry is increasing day by day. By composting these resources, it is possible to obtain organic raw materials that are beneficial to the soil and to protect the environment [12]. At the same time, rational input can be provided in organic agriculture for plant nutrition. Cegarra et al. [11] stated that the final form of composted olive oil processing solid waste has a higher organic matter content and remarkable mineral elements without toxic elements.

Several studies were carried out on the applications of compost obtained from olive oil processing wastes in agricultural production. Raviv et al. [13] applied composts produced from solid and liquid wastes of olive oil mill on tomato seedlings. Michailides et al. [14] produced compost from three-phase olive pomace waste and olive leaves and tested it on lettuce yield. Killi and Kavdır [15] carried out a study on the effects of compost produced from three-phase pomace waste on tomato yield. Diacono and Montemurro [16] conducted a study on the effects of composts obtained from two-phase olive pomace on the yield of organic emmer crop. However, none of these studies carried out a comparative study as to the effects of composts obtained from two-phase and three-phase olive mill solid wastes and olive oil waste water sludge separately on the growth performance of *Solanum lycopersicum* L. seedlings in growing medium.

The purposes of this study were to evaluate composts obtained from two-phase and three-phase olive mill solid wastes and olive oil waste water sludge, to determine the effects of enrichment of composts, and to compare different compost rates on organic tomato seedling production.

#### **2. Materials and Methods**

This study was conducted during the years of 2015 and 2016. Composts were produced at Composting Facility in Olive Research Institute, Ministry of Agriculture and Forestry. Then, they were tested in seedling production at the Horticulture Department of the Faculty of Agriculture at Ege University, Izmir, Turkey (38◦27 17" N, 27◦14 17" E). Organically certified seeds of tomato cultivar ' ¸Sencan-9 (provided from Ataturk Central Horticultural Research Institute, Yalova, Turkey) were used for the study. Compost materials were obtained from the mixture of olive oil processing wastes from two and three phase systems (two-phase and three-phase olive mill solid wastes, olive oil waste water and olive oil waste water sludge) with separated dairy manure, poultry manure, and wheat straws. All input materials were obtained from the organically certified farms. The optimized mixing ratios for 3 different kinds of composts determined at Composting Laboratory at the Department of Agricultural Machinery and Technology Engineering in Isparta University of Applied Sciences (Table 1) were produced based on dry weight (Table 1) were produced (based on dry weight) [17].


**Table 1.** The optimized mixing ratios for 3 different kinds of composts used in this research.

An aerated static pile composting method was used for composting the wastes (Figure 1). Piles with a width of 2 m, a length of 3 m and a height of 1.50 m were formed. Rutgers aeration strategies [18] were performed for aeration of piles for 360 days, which is in agreement with those reported in the study of Chowdhury et al. [9]. Although the composting process was monitored for temperature, pH, electrical conductivity (EC), moisture, and organic matter contents, C, N, and heavy metals ratios, and total phosphorus, they are not reported here. At the later stages, 0.38 kg of cotton seed meal per one kg of dry matter of the initial compost was added to each compost pile to enrich the composts at day 330 of composting (maturation and stabilization stages). Additionally, 0.16 kg of rock phosphate and 0.02 kg of raw potassium salt [19] per one kg of dry matter of the initial compost was added to each compost pile for the enrichment of composts at day 360 of composting. Composting lasted for 425 days including the maturation and stabilization periods. This prolonged period was due to the enrichment (E) process of composts. Therefore, the enriched versions of each compost were labeled as E2P, E3P, and EWS. Powder sulfur was applied at the fourth month of composting to reduce the pH value in the piles. For this purpose, 8 g of powder elemental sulfur was applied to one kg of dry compost [20].

**Figure 1.** Aerated static pile composting system.

Peat provided from Denizli local peat bogs (Turkey) and composts (2P, 3P, WS, E2P, E3P, and EWS) were used as organic substrates in the growing media with compost ratios (%, *v*/*v*) of 0%, 25%, 50%, 75%, and 100% with local peat. Neither lime nor any nutrient was added into the peat.

Tomato seeds were sown in all growing media on 3 February 2016. All the seeds were sown into 2 trays with 128 plugs in each. Each plug included 2 replicates. After sowing, the trays were left in a germination room at a day/night temperature of 24/24 ◦C and 80% relative humidity (RH) under dark conditions for 3 days, then moved to a heated greenhouse (15/24 ◦C and 70%RH) which is specialized for growing seedlings and the seedlings were grown there for 3 weeks. The seedlings were fertilized with liquid farmyard manure (Botanica, Camli Yem Besicilik, Izmir, Turkey) (2 cc L<sup>−</sup>1, EC:1.32 dS m<sup>−</sup>1, pH:4.6) every day with a boom system based on the previous results of Tuzel et al. [21]. In this period, the germination rate and germination period of the seeds were noted. The germination rate was calculated by counting the number of germinated seeds in the cells and expressed as %. The germination period was determined as the number of days required for 50% seed emergence.

When the seedlings were ready for planting in a month, they were harvested from each replicate containing 20 seedlings of treatments in order to measure shoot and root biomass. The roots were washed and cleaned from the growing medium and separated from the shoots. The root and shoot (stem and leaf) samples were weighed for fresh weight (g) and dried for 48 h in a thermo-ventilated oven at 65 ◦C. Then, these dried samples were weighed for dry weight (g) and dry matter was calculated as (%). The longest root length (from top to bottom) was measured with a tape meter and the average result was expressed in cm. The distance between the starting point of the roots and the tip of plant leaves was measured again with a tape meter (cm) and the values were used as seedling height. Stem diameter was also measured above the root collar of the seedlings between nodium with digital caliper (mm).

Minolta colorimeter (CR-400, Minolta Co., Tokyo, Japan) was used to determine leaf color as CIE L\* a\* b\*. The obtained values of a\* and b\* were used to calculate hue angle [*h*◦ = tan−<sup>1</sup> (b\*/a\*)] and chroma [C\* <sup>=</sup> <sup>√</sup> (a\*<sup>2</sup> <sup>+</sup> b\*2)], which determine the saturation and the essential components of the color (red, yellow, blue, and green), respectively [22]. The total chlorophyll index was measured with a chlorophyll meter (SPAD-502Plus, Konica Minolta, Chiyoda-Tokyo, Japan) and expressed as SPAD.

In order to determine plant nutrient concentrations, the seedlings were harvested after the experiment period over the soil surface. Then, they were washed with tap water and distilled water to clean surface residues, dried at 65 ◦C until constant weight, and were grounded. The samples were wet digested with a microwave digestion system and then filtered up to 50 mL with de-ionized water for P, K, Ca, Mg, Cu, Zn, and Mn measurement. Except for P, other nutrients in the supernatant were measured using an atomic absorption spectrophotometer (AAnalyst 400, Perkin Elmer, Waltham, MA, USA). Phosphorus was determined calorimetrically using the spectrophotometer (TU1880 Double Beam UV-VIS, PG Instruments, Leicestershire, UK). In order to determine the N concentration, the samples were wet digested in 250 mL macro-Kjeldahl tubes using concentrated H2SO4 and Khjeldahl tablet at 350–400 ◦C. After digesting the samples with NaOH (40%), NH4-N was fixed in H3BO3 (2%) and titrated with 0.1 N H2SO4 [23]. The same procedures and methods were applied to determine the mineral compositions of composts and peat used in the growing media and their mixtures as in plant analysis. The organic matter content of the dry samples of materials was analyzed after incinerating the samples at 550 ◦C as recommended by the US Department of Agriculture and the US Composting Council [24]. The pH and EC of the fresh samples were extracted by shaking at 180 rpm for 20 min at a solid:water ratio of 1:10 (*w*/*v*) [25], and were measured using pH (pH 720, WTW, Weilheim, Germany) and EC (Multi 340i, WTW, Weilheim, Germany) meters.

The experimental design was randomized blocks with 4 replicates (*n* = 20). A factorial analysis was performed with the composts (WS, EWS, 2P, E2P, 3P, E3P) and ratios (0%, 25%, 50%, 75%, and 100% with local peat and the interaction between these 2 factors. The data were subjected to analysis of variance to determine any statistically significant differences by using the JMP statistical analysis package program (SAS Institute, Cary, NC, USA). The Tukey range test was conducted at a 5% significance level.

#### **3. Results**

#### *3.1. Physical and Chemical Properties of Substrates*

Some physical and chemical characteristics of the seedling growing media were determined before seed sowing (Tables 1–3). The organic matter, content of the media was 38.45% at the initial stage. However, when the compost ratio was increased from 0% to 100% in the growing media at the start, the organic matter contents increased with the rate of 38.49%, 28.32%, 41.40%, 19.25%, 62.21% and 67.70% for WS, EWS, 2P, E2P, 3P and E3P, respectively. The highest organic matter (64.48%) was determined for E3P with a compost ratio of 100%. EC of the local peat was 1.11 dS m−<sup>1</sup> before seed sowing. By the use of composts, EC values increased dramatically in particular when the composts were enriched and used with 75% and/or more. The pH of the growing media changed between 5.60 and 7.38. The pH decreased with an increasing compost ratio in the growing medium (Table 2).



Means within each column followed by the same letters are not significantly different according to the Tukey test. \* and \*\*\*: significant at 0.01 < *p* ≤ 0.05 and *p* ≤ 0.001, respectively.

The main and interaction effects of the treatments on the N, P, K, and Mg concentration of the growing medium before seed sowing were found to be significantly different. The initial N concentration (0.81%) of the growing media increased due to the increase in the compost ratio from 0% to 100% at the start with the rate of 34.57%, 70.37%, 16.05%, 53.09%, 61.73% and 111.11% and for WS, EWS, 2P, E2P, 3P and E3P, respectively. Higher compost ratios produced higher P and K concentrations of the media. The average Ca concentrations of WS, EWS, 2P, E2P, 3P, and E3P were 2.18%, 2.33%, 2.26%, 2.51%, 3.10%, and 2.25% at the start, while the Mg concentration changed between 0.45% and 0.82% (Table 3).


**Table 3.** Macro nutrient concentrations of the growing medium before seed sowing.

Means within each column followed by the same letters are not significantly different according to the Tukey test. ns, \*, \*\* and \*\*\*: nonsignificant, significant at 0.01 < *p* ≤ 0.05, 0.001 < *p* ≤ 0.01 and *p* ≤ 0.001, respectively.

The type of composts and ratios also affected the Zn, Mn, and Cu concentrations of the growing media at the start of the experiment. The Zn concentration varied between 68.2 and 432.4 mg kg<sup>−</sup>1, the Mg and Cu concentration varied between 107.8–287.8 mg kg−<sup>1</sup> and 36.6–55.0 mg kg−<sup>1</sup> before seed sowing (Table 4).

#### *3.2. Germination Period and Rate*

The number of days from seed sowing until germination was 4.25 days in local peat (0%) and increased in all composts with increasing compost ratios in the growing medium particularly in the enriched treatments. The use of a compost ratio of 25% in the growing medium shortened the number of days compared with other compost ratios, but extended 11.8%, 17.6%, 5.9%, 17.6%, 5.9% and 111.8% in WS, EWS, 2P, E2P, 3P and E3P, respectively, compared to local peat, while the extension rate was 41.2%, 252.9%, 117.6%, 194.1%, 152.9%, and 264.7% for a compost ratio of 100% compared with local peat (Table 5). The germination rate also showed the same tendency and decreased with increasing compost ratios, but the ratio changed dramatically in the enriched growing medium (Table 5).


**Table 4.** Micro nutrient concentrations of the growing medium before seed sowing.

Means within each column followed by the same letters are not significantly different according to the Tukey test. ns, \* and \*\*: nonsignificant, significant at 0.01 < *p* ≤ 0.05 and 0.001 < *p* ≤ 0.01, respectively.



\* "0" is accepted as germination rates lower than 50%. Means within each column followed by the same letters are not significantly different according to the Tukey test. Capital letters show significant differences in mean values of composts and compost ratios in peat; lowercase letters indicate significant differences in interaction.

#### *3.3. Seedling Growth*

The effects of the treatments on the lengths of shoots and roots and stem diameter were found to be significantly different (Table 5). The shoot length changed between 16.33 and 4.65 cm. A compost ratio of up to 50% in the growing medium promoted the shoot length, but an increasing compost ratio had an impact on shoot growth excluding the compost ratio of 75% in WS. The shoot length sharply decreased in E2P and E3P. The root length was similar in the treatments, but it decreased by 35% in E3P. The stem diameter also showed similarities to the other measured parameters and decreased in the enriched treatments with an increasing compost ratio (Table 6).


**Table 6.** Effects of treatments on growth.

Means within each column followed by the same letters are not significantly different according to the Tukey test. Capital letters show significant differences in mean values of composts and compost ratios in peat; lowercase letters indicate significant differences in interaction.

The treatments affected the dry weights of the roots significantly. Although root dry/fresh weights increased with compost ratios in the growing medium, this tendency did not continue with increasing ratios. Particularly, the values of E2P and E3P with compost ratios of over 25% showed less root growth (Figure 2).

The main and interaction effects of the treatments on shoot growth were also found to be significantly different. The results showed that the highest dry weights were in WS, while the lowest values were determined for the seedlings grown in E3P. Increasing doses of compost ratios of more than 25% and enrichment had negative effects on seedling dry weights (Figure 3).

#### *3.4. Chlorophyll Index*

The treatments affected the chlorophyll index values (SPAD) significantly. However, there was a slight reduction in WS, 2P, and 3P with increasing compost ratios, whereas the chlorophyll index increased in the enriched compost treatments (Table 7).

**Figure 2.** Main (**a**,**b**) and interaction (**c**) effects of the treatments on root dry weight. Means within each column followed by the same letters are not significantly different according to the Tukey test.

**Figure 3.** Main (**a**,**b**) and interaction (**c**) effects of the treatments on shoot dry weight. Means within each column followed by the same letters are not significantly different according to the Tukey test.


**Table 7.** Effects of the treatments on chlorophyll index values (SPAD).

Means within each column followed by the same letters are not significantly different according to the Tukey test. Capital letters show significant differences in mean values of composts and compost ratios in peat; lowercase letters indicate significant differences in interaction.

#### *3.5. Leaf Color*

The main effects of composts and compost ratios on the "L\*" value of leaf color were significant. The lowest "L\*" was in the growing medium composed of local peat. Additionally, "L\*" was lower in the enriched composts. The compost ratios only affected the "a\*" value and the treatments showed significant difference when compared with peat usage. However, the "b\*" value was affected by the main and interaction effect of the treatments and the b\* values of the enriched composts were lower. The value of "h" changed according to the compost ratios and peat usage and 2P with a compost ratio of 75% gave the lowest hue value. However, "C\*" had the same tendency with the "b\*" value (Table 8).


**Table 8.** Effects of treatments on leaf color.


**Table 8.** *Cont.*

Means within each column followed by the same letters are not significantly different according to the Tukey test. Capital letters show significant differences in mean values of composts and compost ratios in peat; lowercase letters indicate significant differences in interaction.

#### *3.6. Nutrient Concentration*

Individual effects of composts and compost ratio with local peat and their interactions on the N and P concentrations of the seedlings showed a similar effect. Based on the interactions, both nutrient concentrations containing enriched composts with compost ratios of 50%, 75%, and 100% for EWS and E2P and with compost ratios of 25%, 50%, and 75% for E3P were higher than those of the composts without enrichment. The mean plant nutrient concentrations of compost rates significantly varied from 2.71% (a compost ratio of 25%) to 3.54% (a compost ratio of 75%) for N, and from 0.15% (0%) to 0.76% for P (Table 9). As for the plant Ca concentration obtained from composts × compost ratio interactions, increasing the compost ratios resulted in a decrease of Ca in plant tissue. This result implies that 100% local peat as seedling substrate had the highest Ca concentration. These results can also be obtained from the compost ratio comparison. The mean values showed that the Ca concentrations obtained from E3P were higher than those obtained from other composts. The plant Mg concentrations showed a similar tendency to Ca. Namely, except for E3P with a compost ratio of 25%, all the other plant Mg concentrations measured from the plugs with 100% local peat were higher. Furthermore, higher compost ratios generally led to a decrease in the plant Mg concentrations. The same trend was recorded from the means of compost ratios. While the lowest Mg concentrations were determined from 2P, there was not a significant variation among the means of the other composts (Table 9).



**Table 9.** *Cont.*

Means within each column followed by the same letters are not significantly different according to the Tukey test. Capital letters show significant differences in mean values of composts and compost ratios in peat; lowercase letters indicate significant differences in interaction.

The plant Zn concentrations were significantly affected by individual factors and their interactions (Table 10). The Zn concentrations increased with increasing compost ratios. The Zn concentrations of the seedlings grown on the enriched composts were usually higher than those of the other composts without enrichment and the highest values were measured from E3P with a compost ratio of 75% and E3P with a compost ratio of 100% with the values of 325 and 226 mg kg−<sup>1</sup> Zn in seedling tissue. Compared to the control (0%), the plant Zn concentrations showed more than threefold increment with increasing compost ratios up to 75%. The means of composts showed that Zn levels determined from the enriched compost were higher than those obtained from non-enriched composts. The highest Zn concentration was measured from the plants growing on E3P. The individual effects of composts and compost ratio showed a significant effect on the Mn and Cu concentrations (Table 10). While the seedling Mn concentrations increased with the compost ratio, the plant Cu concentrations decreased, but no significant differences were observed among compost ratios between 25% and 100%. The results show that the enriched composts seemed to be more effective than the non-enriched composts on the plant Mn concentrations. Additionally, WS was statistically in the same group. The Mn concentrations obtained from 2P and 3P substrates had the lowest values. Similarly, the plant Cu concentrations measured from the enriched composts were higher than those measured from the non-enriched composts and the highest Cu concentration was determined from the plant grown on E3P. WS had the lowest effect on the plant Cu concentration.


**Table 10.** Effects of the treatments on micro element concentrations of leaves.

Means within each column followed by the same letters are not significantly different according to the Tukey test. Capital letters show significant differences in mean values of composts and compost ratios in peat; lowercase letters indicate significant differences in interaction.

#### **4. Discussion**

Seedlings are grown in a limited volume of containers, however, materials and rates utilized in formulations of growing medium affect the physical, chemical and/or biological properties of medium [26], which is also directly linked with seedling quality. Growing medium provides physical support, aeration, supply of water, and nutrients [27]. In our experiments, the enrichment of the growing medium and also increasing the compost ratio increased organic matter content, electrical conductivity, and macro and micro element concentrations. The origin of compost also affects the nutritional features of growing medium. Furthermore, olive oil processing wastes are rich in nutrients with a higher electrical conductivity [28,29]. Although there were slight changes in organic matter content before planting, P, Ca, Mg, Zn, and Mn decreased during the seedling growth due to plant consumption. However, the increase in N was most probably due to the ongoing mineralization affected by the composition and the characteristics of the material, temperature, and water content [30].

The germination rate changed between 14.45% and 96.88% and decreased by the enrichment of the growing medium in particular in EWS and E2P when the compost ratio was 75% and over, while the germination rate declined in E3P after a compost ratio of 50% and with the increasing compost ratio in the growing medium. However, the germination period also lasted longer with the enrichment of the growing medium and increasing compost ratios. Sánchez–Monedero et al. [31] also reported a lower germination rate and a delay in seedling emergence when the relative proportion of the compost increased in the growing medium, leading to higher EC. The rate and duration of germination are affected by the physical and chemical properties of the growing medium, the rate of ingredients, the requirement of crop species, and crop management including irrigation, fertigation, and the use of beneficial microorganisms as well as environmental conditions [32].

In terms of germination rate, two composts made from olive pomace waste and green waste were used as growing medium components at four ratios (20%, 45%, 70%, 90%, *v*/*v*) and compost made of green waste with ratio 20% and 45% and olive pomace waste with ratio of 20% showed the best performances [29]. Perez-Murcia et al. [33] tested the addition of increasing quantities of composted sewage sludge to peat (0%, 15%, 30%, and 50%, *v*/*v*), and increasing sewage sludge treatments (especially 30% and 50%) reduced the germination of lettuce and broccoli, but in cauliflower seedlings, an increment of germination was observed for the 15% and 30% treatments compared with the control. A compost ratio of 25% for composted rose oil processing [34] and for olive oil production wastes [35] was found appropriate in terms of the rate and duration of germination for organic tomato seedling production which is in harmony with our results.

Healthy seedling growth is a prerequisite for the success of crop production [36]. The shoot length was lower in compost ratios over 50% excluding WS, which reacted to over 75%. Shoot length and stem diameter decreased by the enrichment of the growing medium over 50% compost rate in EWS and E2P. The longest root lengths were also affected by the enrichment of medium excluding WS and EWS which could be also be related to the washing process. The development of shoot, root, and stem was the poorest in E3P. The nutrient contents of the growing media were higher in the ones with higher compost ratios and the enriched ones (Table 2), but the EC values were also high in those ones. The highest average EC value was in E3P treatment, resulting in the poorest shoot, root and stem development.

Tomato is moderately sensitive to salinity and salinity threshold of tomatoes is 2.5 dS m−<sup>1</sup> [37]. Increasing salinity in the rhizosphere restricts root cell growth and increases root lesion, resulting in a reduction in root elongation rate and lateral root growth. Additionally, a reduction in photosynthesis and tissue expansion and the inhibition of cell division affect leaf and shoot growth [38]. Maggio et al. [39] found that high EC (approx. 9.6 dS m−1) caused a sharp increase in the values of root and shoot abscisic acid (ABA), which coincided with the reduction of stomatal resistance to ABA, a different partitioning of Na ions between young and mature leaves, and the increase of root to shoot ratio [39]. In our experiment, morphological measurements (a decrease in shoot length, stem diameter, shoot and root biomass with an increasing compost ratio and enrichment process, poor growth particularly in under E2P and E3P) and SPAD readings, which showed the greenness or the relative chlorophyll concentration of leaves and the highest root to shoot dry matter ratio (in E3P), confirm the effect of salt stress on the seedlings.

The highest plant dry weights were measured from the plants grown on the media with compost ratios of 50% and 25% for WS and E2P, respectively. The variation of the results could be explained in terms of the chemical composition of the composts [40–42]. However, some other properties such as humic and fulvic acid and some other hormones like substances may also have positive effects on plant growth, and thus dry weight [43]. The decrease of dry weight with an increase higher than 50% in compost ratio either enriched or not might be due to the toxicity of some fenolic compounds on plant growth [44,45]. In order to prevent the toxic effect of WS, it was reported to follow the changes occurring in phenols and biotoxicity during composting. Moreover, Zenjari et al. [46] indicated that toxicity disappeared after 2 months of composting. Many studies conducted with different plants grown on different composting materials proposed rates of WS in composting between 25% and 67% [31,47]. The enrichment of 2P (E2P) with P and Ca due to different materials, especially rock phosphate, may have a positive effect on plant growth and dry weight.

The results show that all the composts, either enriched or not, and compost ratios had significantly different effects on most of the plant nutrient concentrations. If a general evaluation is made for the plant N, P, and Zn concentrations, it can be clearly seen that these nutrient concentrations in plants grown on the enriched composts were higher than the non-enriched composts. A number of studies showed that pre-mixing rock phosphate with agro-wastes followed by composting increased the P availability to plants [48–51]. Local peat seems to be the best medium in terms of the plant Ca and Mg concentrations. However, it is quite clear that the dilution effect played a very important role especially for Ca, as dry weights obtained from 100% local peat containing plug were quite low when compared to most of the media. It is well-documented in the literature that nutrients are diluted in plant tissues with plant growth and concentrated with growth retention [41,52].

Among the tested compost ratios, a ratio of 25% was found appropriate in most of the measured properties. However, compost ratios could be increased by up to 50% in the case of water sludge use. Previous research results also propose a rate starting from 25% up to 67% in different crops (such as poinsettia with olive mill wastes [53]; tomato with municipal solid waste compost [47]; broccoli, onion, and tomato with sweet sorghum bagasse, pine bark, and either urea or brewery sludge [31]; lettuce, chard, broccoli, and coriander with exhausted grape marc and cattle or poultry manure [54]). The chemical and physical properties of compost affect the compost ratio in the growing medium [47] and nitrogen has the greatest effect on transplant growth [55]. In our experiment, the higher EC level of the growing medium when enriched and/or included higher compost ratio affected plant growth starting from the seed germination stage. These results are in harmony with the results of our experiments conducted with composts containing rose oil processing wastes [34] and olive oil production wastes [35].

Peat is the most common substrate in seedling production. Although peat-based growing media are allowed in organic production, peat substitution in plant nursery activity and, in particular, in organic seedling production is a debated issue [56] since peat utilization contradicts numerous fundamental principles of organic agriculture. EGTOP (Expert Group for Technical Advice on Organic Production) advises that its use in growing media should be limited to a maximum of 80% by volume, as normally 20%–30% of peat by volume in growing media for professional use could be replaced by compost [57]. Our results showed that composts based on olive mill wastes and olive oil wastewater sludge could be used in the growing medium of vegetable seedlings and there is no need to enrich the medium, which results in a much higher electrical conductivity and higher costs.

Future studies should focus on the enrichment of composts with the effective microorganisms to improve soil fertility and facilitate the nutrient uptake from the soil.

#### **5. Conclusions**

In conclusion, the composts obtained from two-phase and three-phase olive mill solid wastes and olive oil wastewater sludge can be used without any need of enrichment and a ratio of 25% was found appropriate in most of the measured properties. However, compost ratios could be increased by up to 50% in the case of water sludge compost use.

**Author Contributions:** All the authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Y.T., K.E., G.B.Ö., ˙ I.E., N.V. and Ö.M. The manuscript was written by Y.T., K.E., G.B.Ö. and ˙ I.E. All the authors read and approved the final manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was carried out in the framework of the project "Developing of Input Production Methods for Utilization in Organic Plant Production", which was approved by The Scientific and Technological Research Council of Turkey (TUB˙ ITAK) with project number 11-G−055.

**Acknowledgments:** The authors are grateful to Gulay Be¸sirli for organic seed supply.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Biochar and Vermicompost Amendments A**ff**ect Substrate Properties and Plant Growth of Basil and Tomato**

## **Lan Huang 1, Mengmeng Gu 2, Ping Yu 3, Chunling Zhou <sup>4</sup> and Xiuli Liu 5,\***


Received: 12 December 2019; Accepted: 1 February 2020; Published: 4 February 2020

**Abstract:** The suitability of biochar (BC) as a container substrate depends on the BC mix ratio and plant species. Mixes with mixed hardwood BC (20%, 40%, 60%, and 80%, by volume) and vermicompost (VC; 5%, 10%, 15%, and 20%, by volume) were evaluated as container substrates on basil (*Ocimum basilicum* L.) and tomato (*Solanum lycopersicum* L. 'Roma') plants compared to a commercial peat-based substrate (CS). The CS made up the rest of the volume when BC and VC did not add up to 100%. The total porosity of all mixes with BC, VC, and CS (BC:VC:CS mixes) was similar to the control. Mixes with 80% BC had lower container capacity than the control. At 9 weeks after transplanting, the leachate pH of all the BC:VC:CS mixes was higher than that of the control, except for mixes of 20%BC and 5%VC with the rest (75%) being CS (20BC:5VC:75CS) and 20BC:10VC:70CS with tomato plants. The soil plant analysis development (SPAD) readings in BC:VC:CS mixes were similar to or higher than the control except for tomato plants in 80BC:5VC:15CS, 80BC:15VC:5CS, and 80BC:20VC:0CS mixes. Plants in BC:VC:CS mixes had similar growth indexes and total dry weight with respect to those in 100% CS, with the root DW of basil plants in 60BC:15VC:25CS being the highest among all treatments. Therefore, the BC (20%, 40%, 60%, or 80%, by volume) and VC (5%, 10%, 15%, or 20%, by volume) mixes had the potential to replace CS for container-grown plants, with the estimate wholesale price for 80BC:5VC:15CS was only 61.6% that of the control.

**Keywords:** container; growing media; nursery production; carbon; peat moss; bioenergy

## **1. Introduction**

Biochar (BC), attracting increasing interests in recent years for its use in agriculture, can be used to replace some components of commonly used container substrates [1–3]. Biochar could be made from the pyrolysis [4,5] or gasification [6,7] of biomass. The main purpose of the bioenergy production process, pyrolysis and gasification, is to produce syngas or bio-oil [4,8,9], and BC is the by-product. The application of BC in other fields including agriculture provides extra benefits to the bioenergy producers. Biochars can be made from green waste [10,11], wheat straw [12,13], wood [13,14], and rice hull [15], and are renewable and quickly generated [16]. Biochars produced from various raw materials or production conditions would be different and thus cause diverse effects when being incorporated in

container substrates [17,18]. Meanwhile, BCs made from the same feedstock but with different fraction sizes could have different pH and nutrient levels [19].

The incorporation of BC in container substrates has many benefits. Ecological issues caused by extracting the most commonly used horticultural substrate constituent peat from peatlands has increased the necessity of using alternative growing media components including BC in the near future [20]. Research has shown that peatlands have been drained for peat use in agriculture for a long time, which led to the loss of carbon to the atmosphere [21]. Drained peatlands would cause the release of 1.91 Gt CO2-eq. contemporary annual greenhouse gas emission, and peatland rehabilitation is strongly needed [22]. Due to the environmental concern of using peat, the use of BC in containers substituting peat could be a more sustainable choice for the horticultural industry. Biochar in container substrates could increase water-holding capacity [23] and reduce nutrient leaching [24]. Furthermore, the incorporation of BC in peat-based substrate could increase substrate electrical conductivity (EC) and mineral nutrients uptake [25]. Many of the BCs used were alkaline and thus could be used to raise the pH of acidic substrate [10,26,27].

Although BC in container substrates has a lot of benefits, different plant species and BC mix ratios in the container may lead to different results. Kadota and Niimi [28] concluded that mixing 10% or 30% (by volume) BC to the basal medium (substrate with peat, vermiculite, soil, and sand at the ratio of 2:1:1:1, by volume) caused enhanced zinnia (*Zinnia linearis* Benth) shoot growth but no positive effects on marigold (*Tagetes patula* L.) or scarlet sage (*Salvia splendens* Ker Gawl.). Mixes with potato anaerobic digestate and acidified wood pellet BC (1:1, by volume) increased tomato (*Solanum lycopersicum* L.) dry weight (DW) but decreased marigold DW compared to those in the 1:1 peat:vermiculite control [29]. Mixes of 50% or 70% (by volume) sugarcane bagasse BC with the rest being bark-based container substrates led to decreased tomato total DW but no negative effects on basil DW compared to the control [30]. In addition to plant species, different percentages of BC mixed with other substrates components also led to diverse results. Gu et al. [31] has shown that the BC rate in pine bark mixes was positively correlated with gomphrena (*Gomphrena globosa* L.) fresh weight (FW) and DW. Housley et al. [32] found that the pansy (*Viola hybrida* Schur) aboveground DW was increased in the mixes with pine bark, coir, clinker ash, and coarse sand when incorporated with 2.5% (*w*/*w*) Sydney blue gum (*Eucalyptus saligna* Sm.) wood chip BC, while suppressed when incorporated with 10% (*w*/*w*) BC, compared to the control. Webber et al. [33] also showed that the amendment of 50% (by volume) pneumatic bagasse BC could increase squash (*Cucurbita pepo* L.) plant DW, but the amendment of 25% (by volume) BC caused no negative effect on plant DW in comparison with the control.

Vermicompost (VC) is produced by using worms to digest and thus break down organic matter, such as sewage sludge [34], animal waste [35–37], and crop residues [38]. Vermicompost is finely textured and rich in nutrients [39,40], and it has good water-holding capacity [41]. Beneficial effects have been shown in a lot of studies in which VC was used in containers with other substrates. Atiyeh et al. [40] concluded that VC addition in container substrates would enhance plant growth. The swamp rose mallow (*Hibiscus moscheutos* L.) grown in containers with VC showed improved plant DW [42]. Vermicompost mixed with coir at a ratio of 2:1 (*w*/*w*) as container substrate increased Swiss chard (*Beta vulgaris* L.) plant height and FW [43]. The substitution of peat-based growing media with VC (10%, 20%, 30%, 40%, and 50%, by volume) and BC (4%, 8%, and 12%, by volume) did not cause any negative effect on the shoot DW of petunia (*Petunia hybrida* E.Vilm.) and pelargonium (*Pelargonium peltatum* L.) except pelargonium in mixes of 4% BC and 50% VC [44]. It was also shown that the plant size, flower production, and root growth capacity of petunia and pelargonium in BC (4%, 8%, and 12%, by volume) and VC (10%, 20%, and 30%, by volume) mixes was similar to or higher than those in the peat-based substrate control [45]. However, VC made from different parent materials could have different properties [46]. The addition of VC in container substrate may not always cause positive effects on plant growth. Liu et al. [47] have shown that the DW of vegetative and flower organs and growth index of pepper (*Capsicum annuum* L.) in BC (70%, 80%, and 90%, by volume) mixes with VC were lower than those in the Sunshine #1 Mix, which is a peat moss-based substrate. Therefore, although a specific percentage of certain VCs could be used as container substrates to grow plants, caution is also needed due to the complexity of VC.

Tomato (*Solanum lycopersicum* L.) and basil (*Ocimum basilicum* L.) are two widely consumed plants in the horticultural industry. Tomato is an important source of various antioxidant vitamins including ascorbic acid, tocopherols, vitamin C, and carotenoids [48]. Tomato plants are considered as "heavy feeder", requiring medium to high fertility [49]. Basil is referred to as the "king of the herbs" [50]. The essential oil of basil is used in various food products, perfumes, insecticides, medicines, and industrial products [51,52]. Basil is sensitive to high fertility [53]. It was shown that high fertility (500 mg N L<sup>−</sup>1) reduced the basil leaf area, when compared to the ones growing with 100 mg N L−<sup>1</sup> [54].

Few research studies have investigated combinations of BC with VC as container substrates. Due to the high cost of VC and the proneness to use more BC in containers to replace the commonly used peat-based substrate, low percentages of VC (5%, 10%, 15%, and 20%, by volume) and wide range of percentages of BC (20%, 40%, 60%, and 80%, by volume) were used in this experiment. The purpose of this experiment was to test the potential of the mixed hardwood BC and VC mixes as replacements for a commercial peat-based substrate (CS). The specific objectives were to (1) investigate the physical and chemical properties of the BC and VC mixes; and (2) compare the impacts of different mixes of BC with VC on container-grown basil and tomato plants to 100% CS.

#### **2. Materials and Methods**

#### *2.1. Plant Materials and Container Substrates Treatments*

Tomato 'Roma' (Morgan County Seeds, Barnett, MO, USA) and basil seeds (Johnny's Selected Seeds, Winslow, ME, USA) were sown in commercial propagation mix (Propagation mix; Sun Gro® Horticulture, Agawam, MA, USA) in plug trays on 28 October 2016. One tomato seed and four basil seeds were sown per cell (hexagon with side length of 2.6 cm; height: 4.2 cm; volume: 20 mL). Uniform basil and tomato seedlings were selected and transplanted into the experimental substrates in pots (depth: 10.8 cm; top diameter: 15.5 cm; bottom diameter: 11.3 cm; volume: 1330 mL) on 16 November 2016 after true leaves emerged. Each container contained one tomato seedling or four basil seedlings. Sixteen BC and VC mixes were formulated by mixing four rates of BC (20%, 40%, 60%, and 80%, by volume; a by-product of fast pyrolysis of mixed hardwood, Proton Power, Inc., Lenior City, TN, USA) with four rates of VC (5%, 10%, 15%, and 20%, by volume; Pachamama earthworm castings; Lady Bug Brand, Conroe, TX, USA). The CS (BM7 35BKS; Berger, Saint-Modeste, QC, Canada) made up the rest of the volume when the BC and VC did not add up to 100%. The CS was used as the control (Figure 1). The CS (Berger BM7 35BKS) used in this research consisted of 55% coarse peat moss, 35% pine bark, and 10% horticultural perlite. The wholesale price for the mixed hardwood BC was \$65.4 per cubic meter (Personal Communication). The wholesale price was approximately \$176.6 per cubic meter for the CS (Berger BM7 35BKS) and \$607.4 per cubic meter for the VC [55]. The estimated wholesale price for the 17 different substrates per cubic meter was shown in Figure 2. The estimated wholesale price for mixes of 20% BC and 10% VC by volume with the rest (70%) being the CS (20BC:10VC:70CS), 20BC:15VC:65CS, 40BC:15VC:45CS, 20BC:20VC:60CS, 40BC:20VC:40CS, and 60BC:20VC:20CS was higher than the 100% CS (control), while the other BC:VC:CS mixes were all cheaper than the control. The cheapest treatment (80BC:5VC:15CS) was only 61.6% of the price of the CS. The nutrient concentration (N, P, K, Ca, Mg, S, B, Ca, Cu, Fe, Mn, Na, and Zn) of the CS, BC, and VC were tested by the Texas A&M AgriLife Extension Service Soil, Water and Forage Testing Laboratory in College Station, TX, USA and shown in Table 1.

**Figure 1.** Seventeen formulated substrates including mixes of biochar (20%, 40%, 60%, or 80%, by volume) with vermicompost (5%, 10%, 15%, or 20%, by volume) and the 100% commercial peat-based substrates, Berger BM7 35BKS.

**Figure 2.** The estimated wholesale price (\$) for the 17 formulated substrates per cubic meter. The ratios on the *X*-axis indicate the percentage ratio of biochar to vermicompost to commercial substrate (by volume). The control was 100% commercial substrate (Berger BM7 35BKS).

**Table 1.** Nutrient analysis of the commercial substrate (CS, Berger BM7 35BKS), biochar (BC), and vermicompost (VC).


<sup>z</sup> Means within a column under each mean factor followed by the same letter are not significantly different according to the Tukey's HSD test at *p* < 0.05 (*n* = 4).

The pH of the CS, BC, and VC was measured by using a handheld pH-EC meter (HI 98129, Hanna Instruments, Woonsocket, RI, USA), and the EC was measured by using the Bluelab Combo Meter (Bluelab Corporation Limited, Tauranga, New Zealand) according to the pour-through extraction method [56]. The pH of the CS was 6.06, and the EC was 1.3 dS m<sup>−</sup>1, respectively. The pH of the BC was 11.18 and the EC was 2.0 dS m<sup>−</sup>1, respectively. The pH of the VC was 4.8 and the EC was 6.7 dS m−1, respectively. The total porosity, container capacity, air space, and bulk density of the BC were 84.7%, 60.3%, 24.4%, and 0.15 g cm−3, respectively. Particle size distribution of the BC was determined by passing 40 g BC through 2.8, 2, 1, 0.425, and 0.25 mm sieves, and the weight was measured to determine the percentage of each particle size. Percentages of the BC particles ranging from greater than 2.8 mm, 2.0 mm to 2.8 mm, 1.0 mm to 2.0 mm, 0.425 mm to 1.0 mm, 0.25 mm to 0.425 mm, and smaller than 0.25 mm in diameter were 47.9%, 19.4%, 19.4%, 9.1%, 2.0%, and 2.2% (*w*/*w*), respectively.

Six replications of the 17 treatments (16 BC:VC:CS mixes plus control) were arranged in randomized complete blocks in the greenhouse located on Texas A&M University campus, College Station, TX, USA to control the environmental variance. The temperature, humidity, and dew point in the greenhouse were monitored using Watchdog (Spectrum Technologies Inc., Paxinos, PA, USA). During the experimental period, the average greenhouse temperature, relative humidity, and dew point were 20.5 ◦C, 76.0%, and 15.4 ◦C, respectively. The basil plants were irrigated with 200 mg nitrogen (N) L−<sup>1</sup> (20N-4.4 P-16.6K) Peters® Professional (Everris NA Inc., Dublin, OH, USA) nutrient solution. The tomato plants were irrigated with 200 mg N L−<sup>1</sup> (20N-4.4P-16.6K) Peters® Professional nutrient solution from 0 to 3 weeks after transplanting (WAT) and changed to 300 mg N L−<sup>1</sup> from 4 WAT. The total N in Peters® Professional contains 8.1% ammoniacal N and 11.9% nitrate N. The pH of the 200 mg N L−<sup>1</sup> (20N-4.4P-16.6K) Peters® Professional nutrient solution was 6.1, and the EC was 1.0 dS m<sup>−</sup>1. The pH of the 300 mg N L−<sup>1</sup> (20N-4.4P-16.6K) Peters® Professional nutrient solution was 5.9, and the EC was 1.3 dS m<sup>−</sup>1.

#### *2.2. Substrate Physical Properties and Substrate Leachate pH*

Four replications of each substrate were tested to determine physical properties including the bulk density, total porosity, air space, and container capacity of the 17 substrates using the porometers of the North Carolina State University Horticultural Substrates Laboratory [57]. The substrate leachate pH was measured at 0, 2, 4, 6, and 9 WAT using a handheld pH-EC meter (HI 98129, Hanna Instrument, Woonsocket, RI, USA) according to the pour-through extraction method using the same amount of leachate for each test [56].

#### *2.3. Plant Growth and Development*

The plant growth index (GI) of each plant was measured at 0, 2, 4, 6, and 9 WAT, respectively. The height of the plant was measured from the medium surface to the highest point of the plant. The widest plant canopy width and its perpendicular width were measured. The plant GI was determined by the following formula: *GI* = *plant height*/*2* + (*plant width 1* + *plant width 2*)/*4* [58]. The leaf chlorophyll content of each plant was measured as soil plant analysis development (SPAD) values at 2, 4, 6, and 9 WAT, respectively using a portable SPAD 502 Plus Chlorophyll Meter (Spectrum Technologies, Inc., Plainfield, IL, USA). Plant leaves were too small to measure SPAD at 0 WAT. The leaf greenness of each plant was determined using the average of readings from three mature leaves.

At the end of nine WAT, plants were harvested to measure DW. For each tomato plant, the stems, leaves, root, and combined fruits and flowers were harvested separately. For each basil plant, the shoot and root were harvested separately. All the plant parts were oven-dried at 80 ◦C to constant weight before the DW measurements. The total DW of each plant was calculated by adding the DW of all parts of the plant.

#### *2.4. Statistical Analysis*

Data were analyzed with one-way analysis of variance (ANOVA) using JMP Statistical Software (version Pro 12.2.0; SAS Institute, Cary, NC, USA) to test the effect of different substrates on the physical and chemical properties and plant growth. The type of substrate was the main factor. Tukey's Honestly Significant Difference (HSD) tests were used for the comparison of means among treatments at *p* < 0.05. Tomato and basil plants were treated as independent studies and were not compared.

#### **3. Results and Discussion**

#### *3.1. Physical Properties of the Container Substrates*

The total porosity of all the BC:VC:CS mixes was similar to the control (Figure 3a). There was no difference between the container capacity of the mixes of BC (20% or 40%) with VC (5%, 10%, or 15%) and the control (Figure 3b). The container capacity of 80% BC mixes were significantly lower than the control, since BC had lower container capacity (60.3%) than that of CS (70.7%). The air space of 80BC:5VC:15CS, 60BC:20VC:20CS, and 80BC:20VC:0CS was higher than that of the control due to the high incorporation rate of BC with large particle size (Figure 3c), which increased the macropores and thus the air space. The fraction of BC with size greater than 2.8 mm (47.9%) was higher than that of CS (25.4%) and VC (1.2%). The past research showed the variable results of the substrates' physical properties after BC incorporation. Tian et al. [10] reported that the total porosity and container capacity of peat substrate with or without 50% (by volume) green waste BC were similar, while others found that the total porosity and container capacity of the substrates increased with the increasing BC rate [59–61]. Guo et al. [62] found that the air space increased as the pine wood BC rate increased. Yu et al. [63] indicated that the air space increased with the increasing mixed hardwood BC incorporation rate from 10% to 100% (by volume), but the trend was totally opposite for sugarcane bagasse BC. Another research showed that the substitution of peat with 10% (by volume) sewage sludge BC caused no difference on the air space in comparison with the 100% peat substrate control [64]. The effect of BC incorporation on a substrate's physical properties is BC-specific. Container substrates hold water in the micropores between or inside the container substrate components' particles [65]. The container capacity would be increased if the incorporation of BC leads to a higher fraction of micropores. Air space is the proportion of air-filled macropores after the water drains [65]. Air space is closely related to the particle size distributions of BC and the other substrate components, and the changed interporosity after BC incorporation affects air space. The effect of BC on total porosity is related to container capacity and air space, since total porosity is the sum of container capacity and air space.

**Figure 3.** Total porosity (**a**), container capacity (**b**), air space (**c**), and bulk density (**d**) (mean ± standard error) of the 17 different formulated substrates. The ratios on the *X*-axis indicate the percentage ratio of biochar to vermicompost to commercial substrate (by volume). The control was 100% commercial substrate (Berger BM7 35BKS). Means indicated by the same letter are not significantly different according to Tukey's Honestly Significant Difference (HSD) test at *p* < 0.05 (*n* = 4).

The bulk density of 15% VC mixes, 60BC:10VC:30CS, and 60BC:20VC:20CS was higher than that of the control (Figure 3d). The increased bulk density could be due to the high bulk density of VC (0.38 g cm<sup>−</sup>3) and BC (0.15 g cm<sup>−</sup>3) compared to the control (0.10 g cm−3). Similar to our results, a lot of research has shown that substitution of the commonly used substrate with BC could increase bulk density [13,23,29].

#### *3.2. Substrate Leachate pH*

Compared to the control, the substrate leachate pH in all BC:VC:CS mixes was increased, except for those of the 20BC:5VC:75CS and 20BC:10VC:70CS mixes with tomato plants at 9 WAT (see Table S1). The increased pH was probably due to the high pH of the BC (11.18) used in this experiment. The liming effect of BC was found in a lot of research [66–68]. In addition, substrate leachate pH tended to slightly decrease during the study (see Table S1), which was possibly due to the acidifying effect of the fertilizer 20N-4.4P-16.6K Peters® Professional (a potential acidity of 188 kg calcium carbonate equivalent per 1000 kg of the fertilizer). Therefore, the reason for the similar pH of mixes of 20BC:5VC:75CS and 20BC:10VC:70CS with tomato plants at 9 WAT with the control could be due to the low percentage of the BC incorporation rate and the relative large amount of nutrient solution applied to the tomato plants for 9 weeks 'washing down' the substrate leachate pH.

#### *3.3. Plant Growth and Development*

For basil, BC:VC:CS mixes caused no negative effect on the SPAD readings in comparison with the control at 2, 4, 6, or 9 WAT (see Table S2). For tomato, the SPAD readings of the plants in BC:VC:CS mixes were similar to or higher than those in the control at 2, 4, and 6 WAT, while at 9 WAT, the leaf SPAD readings of tomato plants grown in the 80BC:5VC:15CS, 80BC:15VC:5CS, and 80BC:20VC:0CS mixes were lower than those in the control (see Table S2). Similarly, Liu et al. [47] found that SPAD readings of bell pepper leaves in mixes of BC (70%, 80%, and 90%, by volume) with the rest being VC were lower than those in commercial substrates. The decreased leaf SPAD readings at 9 WAT could be caused by two reasons. First, it was shown that leaf SPAD readings was closely related to leaf N concentration, and lower SPAD readings indicated lower leaf N concentration [69]. The decreased SPAD readings could be due to the BC's ability to immobilize N [70]. Second, the increased substrate pH after the incorporation of the BC with high pH (11.18) could reduce iron (Fe) availability, causing decreased leaf greenness. It was shown that shoot Fe concentration was lower at substrate with higher pH [71], and leaf SPAD readings were significantly correlated with Fe availability, since Fe is essential for the chlorophyll synthesis [72]. The leaf chlorosis (as measured by chlorophyll concentration) could be more severe with less iron concentration [73]. The reason for the decreased leaf SPAD readings of the tomato plants only shown at 9 WAT was due to the nutrient deficiency of the leaves caused by the strong nutrient sink (fruits and flowers) at that stage, since all the tomato plants had flowers and fruits at 9 WAT.

However, the possible N binding of BC and reduced Fe availability caused by increased pH did not decrease the plant GI and DW of either tomato or basil plants in this research. The GIs of both basil and tomato plants grown in BC:VC:CS mixes were similar to those in the control at 9 WAT (see Figure S1). All basil plants grown in BC:VC:CS mixes had similar shoot and total DWs in comparison with the control (see Figure S2). The root DWs of basil plants in BC:VC:CS mixes were similar to or higher than those in 100% CS, with those in 60BC:15VC:25CS being the highest among all treatments (Figure 4). Similarly, all tomato plants grown in BC:VC:CS mixes had similar DWs (the combined flower and fruit, leaf, stem, root, and total DW) with respect to the control (see Figure S3). The reasons for the enhanced plant growth could be the VC's extra nutrient supply and the BC's nutrient-holding ability. Similar results were reported by Huang et al. [74], who indicated that tomato and basil plant growth in mixes of BC (60% or 70%, by volume) with either 5% VC or chicken manure compost with the rest being CS was similar to or higher than the control. Alvarez et al. [44] also found that the amendment of VC

(10%, 20%, 30%, 40%, and 50%, by volume) and BC (4%, 8%, and 12%, by volume) to a peat-based substrate did not adversely affect the petunia shoot DW.

**Figure 4.** Root dry weight (mean ± standard error) per basil plant harvested at 9 weeks after transplanting. The ratios on the *X*-axis indicate the percentage ratio of biochar to vermicompost to commercial substrate (by volume). The control was 100% commercial substrate (Berger BM7 35BKS). Means indicated by the same letter are not significantly different according to Tukey's HSD test at *p* < 0.05 (*n* = 6).

#### **4. Conclusions**

The mixes of mixed hardwood BC (20%, 40%, 60%, or 80%, by volume) made from fast pyrolysis and VC (5%, 10%, 15%, or 20%, by volume) used in this study had the potential to replace the CS to grow basil and tomato plants. Our results found difference in the substrate leachate pH between the 100% CS and BC:VC:CS mixes except for 20BC:5VC:75CS and 20BC:10VC:70CS with tomato plants at 9 WAT, which indicated the liming effect of the mixed hardwood BC used in this research. At 9 WAT, the leaf SPAD readings of tomato plants grown in 80BC:5VC:15CS, 80BC:15VC:5CS, and 80BC:20VC:0CS mixes were lower than those in the control, which was possibly due to the binding ability of BC or reduced Fe availability caused by increased substrate pH after BC incorporation. The growth index and total dry weight of basil and tomato plants in BC:VC:CS mixes were similar to those in the CS. Considering the cost of the alternative substrates, all the BC:VC:CS mixes (except for 20BC:10VC:70CS, 20BC:15VC:65CS, 40BC:15VC:45CS, 20BC:20VC:60CS, 40BC:20VC:40CS, and 60BC:20VC:20CS) in this experiment could be selected as the suitable ones to grow plants, with the 80BC:5VC:15CS being the cheapest and most recommended. This study is important for the future use of mixtures of BC with VC in container substrate for greenhouse and nursery plant production to provide a sustainable and environmentally friendly way to substitute peat use in agriculture and add value to the bioenergy process by using the by-product BC. Using the BC:VC:CS mixes with wholesale prices cheaper than the 100% CS could provide more economical ways to grow plants and benefit the growers. Tomato and basil plants were used as model plants in this study. Since results were similar for these two plants with different optimal growing conditions, these results could be applicable to many other plants. The results in this study can be only suitable for the specific mixed hardwood BC made from pyrolysis and VC (Pachamama earthworm castings; Lady Bug Brand) due to the complexity of BC and VC. More mixed hardwood BC incorporation percentages and other potential amendment candidates need to be tested for economic viability.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4395/10/2/224/s1, Table S1: The leachate pH of the 17 substrates with basil and tomato plants at 0, 2, 4, 6, and 9 weeks after transplanting (WAT), Table S2: The SPAD reading of basil and tomato leaves in 17 substrates at 2, 4, 6, and 9 weeks after transplanting (WAT), Figure S1: Cumulative growth index (mean ± standard error) per basil (a) or tomato (b) plant grown in 17 substrates at 0, 2, 4, 6, and 9 weeks after transplanting (WAT), Figure S2: Shoot (a) and total (b) dry weight (mean ± standard error) per basil plant harvested at 9 weeks after transplanting, and Figure S3: Leaves (a), stem (b), root (c), combined flower and fruit (d), and total (e) dry weight (mean ± standard error) per tomato plant harvested at 9 weeks after transplanting.

**Author Contributions:** Conceptualization, L.H., M.G., and X.L.; Data curation, L.H.; Formal analysis, L.H.; Investigation, L.H., P.Y., C.Z., and X.L.; Resources, M.G.; Supervision, M.G. and X.L.; Writing—original draft, L.H.; Writing—review and editing, M.G., P.Y., C.Z., and X.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** The authors thank the Agriculture Women Excited to Share Opinions, Mentoring, and Experiences (AWESOME) faculty group of the College of Agriculture and Life Sciences at Texas A&M University for assistance with editing the manuscript.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*

## **E**ff**ects of Mixed Hardwood and Sugarcane Biochar as Bark-Based Substrate Substitutes on Container Plants Production and Nutrient Leaching**

**Ping Yu 1, Lan Huang 2, Qiansheng Li 3, Isabel M. Lima 4, Paul M. White <sup>5</sup> and Mengmeng Gu 2,\***


Received: 31 December 2019; Accepted: 20 January 2020; Published: 22 January 2020

**Abstract:** Biochar (BC) has the potential to replace bark-based commercial substrates in the production of container plants. A greenhouse experiment was conducted to evaluate the potential of mixed hardwood biochar (HB) and sugarcane bagasse biochar (SBB) to replace the bark-based commercial substrate. A bark-based commercial substrate was incorporated with either HB at 50% (vol.) or SBB at 50% and 70% (vol.), with a bark-based commercial substrate being used as the control. The total porosity (TP) and container capacity (CC) of all SBB-incorporated mixes were slightly higher than the recommended value, while, the others were within the recommended range. Both tomato and basil plants grown in the BC-incorporated mixes had a similar or higher growth index (GI), leaf greenness (indicated by soil-plant analyses development), and yield than the control. The leachate of all mixes had the highest NO3–N concentration in the first week after transplantation (1 WAT). All BC-incorporated mixes grown with both tomato and basil had similar NO3–N concentration to the control (except 50% SBB at 1 and 5 WAT, and 50% HB at 5 WAT with tomato plants; 50% SBB at 5 WAT with basil plants). In conclusion, HB could replace bark-based substrates at 50% and SBB at 70% for both tomato and basil plant growth, without negative effects.

**Keywords:** biochar; NO3–N; plant; substrate; container; production

## **1. Introduction**

Both tomato and basil are important crops and 95% of tomato and basil are produced in soilless cultivation systems using different horticultural growing substrates [1]. Tomato is one of the most important horticulture crops, with a total production estimated to be at 164 MT worldwide [2]. Tomato can be grown in coconut fiber, and perlite alone or in mixture with peat, and produce good yields [3]. Additionally, 50% coco–peat mixed with 50% perlite was recommended for tomato seedling production [4]. Basil is an annual herb that is commercially important for its medical and culinary purposes [5,6]. Basil plants can be grown in 75% sphagnum peat moss mixed with 25% coarse perlite [7]. Additionally, the mix of 60% sphagnum peat and 10% biochar with compost, has proven to be suitable for basil production [8].

Container plant production has become a major source of N leaching and runoff that can be a potential contamination source [9,10]. Container plant production requires a large amount of fertilizer, with nitrogen as the key component, making container plant production a major source of N leaching or

runoff [9]. The leachate of N can be a potential contamination source for surface and underground water, resulting in environmental and health concerns [11]. NO3–N, the main form for plants absorption, contributes in large to the N leaching and runoff in soilless production systems.

Bark has become one of the most commonly used container organic components in horticulture [12]. The reason for bark being commonly used in horticulture is because it has suitable properties for container plants to grow well and it is easy to get access to [13,14]. Compared to peat moss, another most commonly used container component, bark, is a byproduct of the forestry industry, is less expensive because it is available locally and does not require extra shipping costs [15,16]. In the USA, Douglas fir bark is mainly used in the pacific northwest, while pine bark is mainly used in the southwest [17,18].

Although bark has been a good container component, besides peat moss, its inconstant and unpredictable supply in recent years has limited its usage in horticulture industry [16,19,20]. Bark supply competes with many other markets, including alternatives of industrial fuel, timber production, housing and paper market, all of which prevent bark from being a constant source for the horticulture industry [20–22]. Since the supply of bark is fluctuating and unpredictable, it would be beneficial for the horticulture industry to explore less expensive and more constant alternatives with similar properties [16,22].

Biochar (BC), a by-product from thermochemical biomass decomposition under an oxygen-depleted or oxygen-limited environment [23–25] with specific time and temperature conditions and from certain carbon-rich raw materials, can be a potential alternative to common substrates for plant growth, as has been documented in many trials [16,26–29]. Research has shown that BC can increase water and nutrient holding capacity, ameliorate substrate acidity, and provide suitable environments for plants [30–32]. It, thus, improves greenhouse crop growth, yield, and quality, under appropriate conditions [32–36].

Biochar has been considered to be a sustainable component of a growing substrate because it can be derived from various agriculture by-products, such as green waste [33], wood, straw [31,37–40], bark [41], rice hull [42], and wheat straw [31,43]. Additionally, due to the significant variation in pyrolysis conditions, the BC properties could vary significantly, and there is no universal standard for BC addition to plant production and BC's effects on container substrates vary, as a result [28]. Research on BC as a substrate amendment is still in its infant stage [29]. In this present study, a trial was conducted to determine whether two types of BCs had the potential to be a replacement of bark-based substrate amendments for container plant production.

### **2. Materials and Methods**

## *2.1. Plant Material*

Plant seeds (tomato, *Solanum lycopersicum* 'Red Robin™', Fred C. Gloeckner and Company Inc., Harrison, NY, USA; basil, *Ocimum basilicum*, Johnny's Selected Seeds, Winslow, ME, USA) were sown in 72-cell plug trays (one seed per cell, cell dimension: 5 cm\*4 cm\*4 cm, depth/length/width; volume: 55 mL) with a commercial germination substrate (BM2 Berger, Saint-Modeste, Quebec, Canada), on 26 February 2019. After the first pair of true leaves expanded, uniform seedlings were transplanted into 6-inch azalea pots (dimension: 10.8 cm\* 15.5 cm\*11.3 cm, depth/top/bottom diameter; volume: 1330 mL) with a commercial growing substrate (Jolly Gardener, Oldcastle Lawn & Garden Inc., Atlanta, GA, USA) that was incorporated with either sugarcane bagasse biochar (SBB, American Biocarbon LLC White Castle, LA, USA) at two different rates (50% and 70%; by vol.), or with mixed hardwood biochar (HB, Proton Power Inc. Lenoir City, TN, USA) at 50% (by vol.), on 27 March 2019.

The composition used in this study was chosen because a previous study had showed that 70% of HB can be successfully incorporated with peat moss based commercial substrates and with composts for tomato and basil production [29], and 50% of SBB can be used for petunia growth (not published). We wanted to do further tests of HB with different compositions, on tomato and basil, using tests of

SBB with different plant species. The main components for the commercial growing substrate was aged pine bark (55%; by vol.), the other ingredients in the substrate were Canadian sphagnum peat moss, perlite, and vermiculite. The commercial substrate was used as the control. The pH of SBB and of HB were 5.9 and 10.1, respectively (Table 1). The SBB and HB had electrical conductivity (EC) of 753 μS/cm and 1,058 μS/cm, respectively [44]. During transplanting, slow-release fertilizer Osmocote Plus (15N-4P-10K, Scotts-Sierra Horticultural Products Company, Marysville, OH, USA) was surface-dressed at the rate of 4.8 g/pot for basil and 7.7 g/pot for tomato. All mixes were placed in a greenhouse at Texas A&M University, College Station, TX, USA. The average greenhouse temperature, relative humidity, and dew point were 23.7 ◦C, 82%, and 19.6 ◦C, respectively.

**Table 1.** The pH, electrical conductivity (EC), total porosity (TP), container capacity (CC), air space (AS), and bulk density (BD) of biochars and the substrate mixes used in this study.


Note: SBB = Sugarcane Bagasse Biochar; HB = Mixed hardwood Biochar; and CS = Commercial bark-based growing mix; <sup>Z</sup> Recommended physical properties of container substrate by Yeager et al. [45].

#### *2.2. Measurements*

## 2.2.1. Potting Mix Physical and Chemical Properties

Mix physical properties—total porosity (TP), container capacity (CC), air space (AS), and bulk density (BD)—were measured according to North Carolina State University Horticultural Substrates Laboratory Porometer [46]. The leachate EC and pH were measured every other week, starting at one week after transplantation (1 WAT), with a portable EC/pH meter (Hanna Instrument, Woonsocket, RI, USA), according to the pour-through method [47].

Nutrient leachate was collected whenever EC and pH were measured and was stored in the refrigerator (4 ◦C) until analysis. A HQ440d Benchtop Meter and ISENO3181 nitrate electrode (Hach Company, Loveland, CO, USA) were used for leachate NO3–N measurements.

## 2.2.2. Plant Growth

Plant height and two widest canopy widths (width 1: horizontal, width 2: perpendicular) were measured at 1, 3, 5, and 7 WAT. The plant growth index (GI) was calculated according to the formula: GI = plant height/2 + (width 1 + width 2)/4 [26]. Plants' leaf greenness was measured at 1 WAT with a portable soil-plant analyses development (SPAD) meter, (SPAD 502 Plus Chlorophyll Meter, Spectrum Technologies, Inc., Plainfield, IL, USA). Each plant's leaf greenness was determined by taking averages of readings from three random mature leaves. Plant stem, leaf, and fruit were harvested separately. After being dried at 80 ◦C in an oven until a consistent weight was reached, their dry weights (shoot dry weight (SDW), leaf dry weight (LDW), fruit dry weight (FDW)) were measured. Plant roots were washed under running water, after harvest. Root length, root surface area, average root diameter, and the number of root tips were measured by using a root scanner (WinRHIZO, Regent Instruments Canada Inc., Quebec, Canada). Root dry weights (RDW) were determined after being dried at 80 ◦C in an oven, until a constant weight was reached. Total dry weights (TDW) were calculated by adding up the SDW, LDW, FDW, and RDW.

#### *2.3. Statistical Analysis*

This experiment was designed as a completely randomized block design with six replications for each mix. A one-way analysis of variance using JMP Statistical Software (version Pro 14.2.0; SAS Institute, Cary, NC, USA) was used for data analysis. All the means were separated by using Dunnett's test when treatments were significantly different from control at *p* < 0.05. A principle component analysis (PCA) was conducted to evaluate the relationship between the selected variables and were treated using R programing software (version 3.5.1).

#### **3. Results**

#### *3.1. Potting Mix Physical and Chemical Properties*

Most of the mixes' physical properties were within the recommended range [45], except for the SBB-incorporated mixes, which had a slightly higher TP and CC than the recommended value (Table 1). The 50% SBB mix had a slightly lower AS, as compared to the recommended value. All the mixes had slightly lower BD in comparison to the recommended value and the commercial mix had the lowest BD among all the mixes.

Tomato and basil plants grown in all BC-incorporated pots had similar EC as compared to the control, throughout the experiment, except for the tomato plants in 50% HB at 1 WAT (Figure 1). The 50% HB mixes with tomato plants had a significantly higher pH than the control at 1, 3, and 7 WAT (Figure 2A). The SBB-incorporated mix with tomato plants (50% at 1 WAT, 70% SBB at 7 WAT) had a significantly lower pH, compared to the control. Plants in all the other BC-incorporated mixes had a similar pH, throughout the experiment. Basil plants grown in 50% HB mixes had a significantly higher pH compared to the control, throughout the experiment (Figure 2B). However, basil plants grown in SBB-incorporated mixes (50% and 70%, at 5 and 7 WAT) had a significantly lower pH, compared to the control.

**Figure 1.** The EC (mean ± standard error) of potting mixes with 50% sugarcane bagasse biochar (SBB), 50% mixed hardwood biochar (HB), and 70% SBB (by vol.) mixed with bark-based commercial substrate (CS) with tomato (**A**) and basil (**B**) plants at 1, 3, 5, and 7 week(s) after transplanting ( WAT). \*indicated significant differences from CS using Dunnett's test at *p* ≤ 0.05.

**Figure 2.** The pH (mean ± standard error) of container mixes, with 50% sugarcane bagasse biochar (SBB), 50% mixed hardwood (HB), and 70% SBB (by vol.) mixed with bark-based commercial substrate (CS) grown with tomato (**A**) and basil (**B**) plants at 1, 3, 5, and 7 week(s) after transplantation (WAT). \*\*indicated significant differences from CS using Dunnett's test at *p* ≤ 0.01.

## *3.2. Leachate NO3–N*

The leachate of all BC-incorporated mixes (both with tomato and basil plants) had a similar or higher NO3–N concentration compared to the control. The leachate NO3–N concentration generally decreased from 1 WAT to 7 WAT, for each mix (Figure 3).

**Figure 3.** Leachate NO3–N (mean ± standard error) of tomato (**A**) and basil (**B**) plants grown in container mixes with 50% (by vol.) sugarcane bagasse biochar (SBB), 50% mixed hardwood biochar (HB), and 70% SBB mixed with bark-based commercial substrate (CS). (**A**,**B**) Amplified figure for tomato (**a**) and basil (**b**) from 5 WAT to 7 WAT. \*, \*\*indicated significant differences from CS using Dunnett's test at *p* ≤ 0.05 and *p* ≤ 0.01, respectively.

#### *3.3. Plant Growth*

In the BC-incorporated mixes, both tomato and basil plants had a similar or higher GI, in comparison to the control, throughout the experiment (Figure 4). Tomato plants in all BC-incorporated mixes had similar SDW and FDW (yield), compared to the control, however, tomato plants in SBB-incorporated mixes had significantly lower TDW, RDW, and LDW compared to the control (Figure 5A). Basil plants grown in all BC-incorporated mixes had similar RDW, SDW (except 50% HB), LDW, FDW, and TDW to the control (Figure 5B). The SPAD of tomato and basil plants grown in all BC-incorporated mixes was no different from the control (Figure 6).

**Figure 4.** Growth index (mean ± standard error) of plants tomato (**A**) and basil (**B**) grown in container mixes with 50% sugarcane bagasse biochar (SBB), 50% mixed hardwood biochar (HB), and 70% SBB (by vol.) mixed with bark-based commercial substrate (CS) at 1, 3, 5, and 7 week(s) after transplantation (WAT). \*indicated significant differences from CS, using Dunnett's test at *p* ≤ 0.05.

**Figure 5.** Total dry weight (Total DW = root dry weight (RDW) + shoot dry weight (SDW) + leave dry weight (LDW) + fruit dry weight (FDW); mean ± standard error) of tomato (**A**) and basil (**B**) grown in container mixes with 50% sugarcane bagasse biochar (SBB), 50% mixed hardwood biochar (HB), and 70% SBB (by vol.) mixed with bark-based commercial substrate (CS). \*indicated significant differences on the total DW from CS using Dunnett's test at *p* ≤ 0.05.

**Figure 6.** The soil-plant analyses development (SPAD) (mean ± standard error) of tomato and basil grown in container mixes with 50% sugarcane bagasse biochar (SBB), 50% mixed hardwood biochar (HB), and 70% SBB (by vol.), mixed with bark-based commercial substrate (CS).

Similar root length, average root diameter, and number of root tips were observed between tomato plants grown in all BC-incorporated mixes and the control (except 50% SBB), however, significantly smaller root surface area of tomato plants grown in all SBB-incorporated mixes were noticed (Table 2). Basil plants grown in all BC-incorporated mixes had significantly shorter root length but bigger diameter than the control. Basil plants in all BC-incorporated mixes had similar root surface area to

the control, yet those in 50% BC-incorporated mixes had significantly fewer root tips than the control (Table 2).

**Table 2.** The root development (mean ± standard error) of plants grown in potting mixes with 50% sugarcane bagasse biochar (SBB), 50% mixed hardwood biochar (HB), and 70% SBB (by vol.) mixed with bark-based commercial substrate (CS). \*, \*\*, and \*\*\*indicated significant differences from CS using Dunnett's test at *p* ≤ 0.05, *p* ≤ 0.01, and *p* ≤ 0.001, respectively.


#### **4. Discussion**

#### *4.1. Potting Mix Physical and Chemical Properties*

Despite the fact that BC can have various effects on substrate properties contingent on the types of feedstocks and the pyrolysis conditions of BC [28,48], many types of BC have been proven to be suitable replacements for commercial growing substrates, without negatively affecting the plant [28,35]. Biochar from fast pyrolysis (pinewood, 450 ◦C), for instance, could replace commercial substrate at up to 80%, providing suitable properties for the poinsettia and Easter lily growth [26,27]. Biochar from fast pyrolysis (mixed hardwood) could be suitable for tomato and basil plant growth, due to the proper properties it created [29]. Sugarcane bagasse BC and pinewood BC mixes had similar physical properties to commercial growing mix, allowing them to be acceptable for bean and cucurbit seedlings production, even though some of the TP and CC in the SBB-incorporated mixes were slightly higher than the recommended values [44]. Adding pruning residue BC (fast pyrolysis, 500 ◦C) to soilless mixes can render appropriate physical properties for vegetable production [35,49]. In this study, even though 50% SBB and 70% SBB mixes had slightly higher TP (81%, 89%, respectively) and CC (75%, 76%, respectively) than the recommended value (TP 50%–80% and CC 45%–65%) [45], the growth of tomato and basil plants was not affected, as observed in Webber's study [44].

Different initial BC pH (HB: 10.05, SBB: 5.94) resulted in differences in pH levels in the different BC mixes. Mixes with HB (50%, by vol.) and commercial bark-based substrates (initial pH: 6.81) had a pH lower than the initial HB but higher than the initial commercial bark-based substrate. The same was true for all SBB mixes. Since SBB had an acidic initial pH, adding 30% to 50% of the commercial substrate (pH: 6.81) resulted in mixes with a pH that was lower than the commercial substrate but was higher than the SBB.

## *4.2. Biochar E*ff*ects on Leachate NO3–N*

Plant species, plant stage, and substrate properties can influence NO3–N leaching [9,50,51]. Tomato, as a heavy feeder fertilizer crop, require more nutrients throughout the growing season than other lighter feeder fertilizer crops, such as snapdragon and bedding plants [52,53]. As a result of administering the same amount of fertilizer to different plant species due to their divergent nutrient requirements, the final NO3–N leaching varies. Additionally, the nutrients demand for plant at different stages also vary. During the growing period, plants' requirement for nutrients presents a skewed

"s" curve—vegetative periods need less nutrient yet when entering the flowering/fruit-set period, the demand for nutrients increases dramatically [54]. Nitrate leaching can be also affected by soil or substrate texture and normally, coarse textured mixtures lead to more nitrate leaching [55]. Substrate properties affecting nitrate leaching can explain why leachate from 50% HB (in both case of tomato and basil) had the lowest NO3–N concentration (except tomato at 5 WAT), among all mixes.

#### *4.3. Biochar E*ff*ects on Plants Growth*

Biochar can have positive, null, and negative effects on plant growth [26,56,57], contingent on plant species, BC types, incorporation rates, and the interactions of both. For instance, pinewood BC had positive effects on bell pepper growth [58], similar results were reported on Easter lily, poinsettia, and "Firework" *Gomphrena.* Mixed hardwood BC can positively affect tomato and basil plants growth [16,26,27,29]. The null and negative effects of BC (from tomato crop waste or wood pellet) on tomato plant growth have also been reported [56,57]. This study obtained similar results to some previous studies that found that BC does not negatively affect plant growth at high incorporation rates [16,26,27,29].

There are few studies with detailed information on BC–root systems [59]. Since roots are essential parts for water and nutrients uptake, plants with better roots were desired [59,60], and the effects of BC on root development is an eventuality. In this study, tomato plants grown in all the BC-incorporated mixes had similar root length, root surface area (except 50% and 70% SBB), average root diameter, and number of tips, in comparison to the control. Basil plants had similar root surface area to the control, which can explain why plants grown in BC-incorporated mixes performed as well as those in the control.

#### *4.4. Treatment Factors Determined Plants and Mix Properties*

As the effect of biochar on plants and mix properties can be complex and difficult to explain, given the fact that two types of biochars and multiple variables were included in this study, a principal component analysis (PCA) was used to depict variables shaped by different biochars with tomato (Figure 7A) and basil (Figure 7B) plants. For tomato plants, 88.9% of the variability was explained by the first two components (Figure 7A). PC1 accounted for 65.8% variance, with SBB differing from HB and CS. Sugarcane bagasse biochar was associated more with yield (FDW) and NO3–N leaching, while CS and HB was related more to plant growth (RDW, LDW, and GI). PC2 accounted for 23.1% variance, distinguishing the CS and BC mixes. Commercial substrate tended to be affiliated with plant biomass, however, BC mixes appeared to be related to nutrient leaching. For basil plants, the first two components explained 77.1% of the variability (Figure 7B). PC1 accounted for 42.9% variance, SBB 50% differing from HB and CS mixes. A 50% sugarcane bagasse biochar mix showed a greater association with NO3–N leaching and SDW, while CS, 70% SBB, and HB showed a greater relation to plant growth, including root parameters (RDW, root length (RL), root tip (RT), and root surface area (RSA)) and chemical properties of the mixes (EC, pH). PC2 accounted for 34.2% variance, distinguishing between the CS and BC mixes. Commercial substrates tended to affiliated with plant biomass, however, BC mixes appeared to be related to the chemical properties of the mixes (EC, pH, NO3–N).

**Figure 7.** Principal component analysis (PCA) depicting the relationships between selected variables and treatment factors with tomato (**A**) and basil (**B**). Selected variables are displayed by arrows and include plant growth parameters—SPAD, growth index (GI), fruit dry weight (FDW), leave dry weight (LDW), shoot dry weight (SDW), root length (RL), root dry weight (RDW), root diameter (RD), root surface area (RSA), and number of root tips (RT); substrate chemical parameters were pH, EC, and NO3–N leachate at different weeks. Treatment factors are displayed by filled grey circles: 50% sugarcane bagasse biochar (SBB 50), 50% mixed hardwood biochar (HB 50), 70% SBB (SBB 70) mixed with bark-based commercial substrate, and bark-based commercial substrate (CS).

#### **5. Conclusions**

The mixed hardwood biochar and sugarcane bagasse biochar used in this experiment could be used as bark-based substrate amendments for container plant production. The mixed hardwood biochar could replace the bark-based substrate at 50% and the sugarcane bagasse biochar at 70%, as growing mixes for tomato and basil production. More than 5.4 M ft<sup>3</sup> container substrates were used in horticulture industry in 2017 and the current container substrate major components—peat moss and bark are causing serious environmental concerns [61]. As can be seen from the results of this study, if mixed hardwood biochar or sugarcane bagasse biochar was chosen for greenhouse production, around 1.35 M ft<sup>3</sup> fewer peat moss or 1.94 M ft3 fewer bark could be used annually (assuming container substrate contains 50% peat moss or bark).

**Author Contributions:** P.Y. conducted the experiments, collected and analyzed the data, and wrote the manuscript with assistance from all other authors, mainly M.G., L.H. and Q.L. provided technical advice and assistance when the study was conducted. I.M.L. and P.M.W. revised and improved the manuscript. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** We would like to thank American Biocarbon LLC White Castle, Louisiana, Proton Power Inc. Lenoir City, Tennessee, for supplying the biochar for the experiment; Elizabeth Pierson and Kevin Crosby, for supplying experimental instruments; Patricia Goodson; all the students in CEHD 603 fall 2019 writing class; Charles L. Webber III and the Agriculture Women Excited to Share Opinions, Mentoring and Experiences (AWESOME) faculty group of the College of Agriculture and Life Sciences at Texas A&M University for assistance with editing the manuscript. The author would also like to thank the Texas A&M University Open Access to Knowledge Fund (OAK Fund, supported by the University Libraries and the Office of the Vice President for Research) for partially covering the open access publication fees.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article*

## **The Use of Dewpoint Hygrometry to Measure Low Water Potentials in Soilless Substrate Components and Composites**

**Jeb S. Fields 1, William C. Fonteno 2, Brian E. Jackson 2,\*, Joshua L. Heitman <sup>3</sup> and James S. Owen Jr. <sup>4</sup>**


Received: 23 July 2020; Accepted: 10 September 2020; Published: 15 September 2020

**Abstract:** Plant water availability in soilless substrates is an important management consideration to maximize water efficiency for containerized crops. Changes in the characteristics (i.e., shrink) of these substrates at low water potential (<−1.0 MPa) when using a conventional pressure plate-base can reduce hydraulic connectivity between the plate and the substrate sample resulting in inaccurate measures of water retention. Soilless substrate components *Sphagnum* peatmoss, coconut coir, aged pine bark, shredded pine wood, pine wood chips, and two substrate composites were tested to determine the range of volumetric water content (VWC) of surface-bound water at water potentials between −1.0 to −2.0 MPa. Substrate water potentials were measured utilizing dewpoint hygrometry. The VWC for all components or composites was between 5% and 14%. These results were considerably lower compared to previous research (25% to 35% VWC) utilizing conventional pressure plate extraction techniques. This suggests that pressure plate measurements may overestimate this surface-bound water which is generally considered unavailable for plant uptake. This would result in underestimating available water by as much as 50%.

**Keywords:** available water; coconut coir; dewpoint potentiometer; peat; pine bark; pine tree substrate; substrate processing; surface-bound water; unavailable water; wood substrate

## **1. Introduction**

Traditionally, substrate scientists separate the water storage capacity of a soilless substrate into two categories, available water (AW; water that is available for plant uptake) and unavailable water (water that is bound tightly to soil surfaces and is unavailable for plant uptake). Soil and substrate scientists separate the availability of water as a function of water potential, as water within the substrate matrix is held at various tensions by a combination of matric and gravitational potentials. To absorb water from the substrate matrix, plants exert suction which must overcome the water tension. As the substrate volumetric water content (VWC) and water potential decreases (tension increases) the water becomes less available for plant uptake. The water potential at which the substrate transitions from AW to unavailable is not exact, but instead plant water availability is gradually reduced as the substrate dries and substrate water potential becomes more negative [1]. Water that is less available for plant uptake is most often tightly bound to particle surfaces, known herein as surface-bound water (SBW).

Water typically becomes less available for common agricultural crops in soil at potentials between −1.0 and −2.0 MPa [2]. Often, soils and substrate researchers use a water potential of −1.5 MPa as the potential at which water becomes plant unavailable for calculation. It is understood that many plant species can survive in soils with water potentials well below −1.5 MPa. However, the change in actual water content as water potential becomes more negative with drying beyond −1.5 MPa is typically negligible, with miniscule losses in water content accounting for substantial drops in water potential thereafter. Denmead and Shaw [3] reported that plants started to reduce transpiration levels at water potentials as high as −0.2 MPa, and Caron et al. [4] indicated that horticultural crops grown in containerized peat-based soilless substrates begin to show stress signals at substrate water potentials as high as −0.003 MPa. However, utilizing substrate VWC at substrate water potentials of −1.5 MPa as an estimated transition value, scientists can estimate substrate AW as a proxy for substrate water storage capacity, the water held at water potentials between container capacity (CC; the maximum volume occupied by water in a soilless substrate after drainage) and the water held at substrate water potentials of −1.5 MPa [5]. This calculation is useful for practical management considerations and to compare substrates.

To determine relationships between soil (or substrate) water potential and VWC, Bouyoucos [6] described an apparatus that introduces suction upon soil samples. This idea was refined by Richards and Fireman [7] who applied pressure and employed the use of porous plates that allowed water to be extracted at a given pressure representing tension. The most commonly used and described method of measuring soil water potential in situ is through the use of tensiometers [8]. However, water potentials below approximately −0.085 MPa cannot be measured with tensiometers due to vaporization and cavitation of water within the device [9]. More recent work has extended the range of tensiometers by employing a polymer in place of water [10]; however, the range remains limited to soil water potentials much greater than −1.5 MPa.

Presently, the most commonly cited method for measuring moisture content (MC) at low water potentials in soilless substrates is the pressure outflow apparatus method described by Cassel and Nielsen [11]. The pressure outflow apparatus is a modified version of Richards and Fireman's pressure plates [7]. This method determines soil or substrate MC at a specified water potential and can be conducted in a relatively short period of time. However, previous research has pointed out inaccuracies with this method due to a loss of hydraulic connectivity between the water in the sample and the water in the plate when the soil or substrate dries [12–15]. This occurs because water moves in porous media primarily by displacement. Therefore, when the water column is interrupted, water movement ceases, preventing equilibrium between the applied pressure and the soil or substrate water potential.

Dewpoint hygrometry has also been used to measure water potentials of porous substances [16]. A dewpoint potentiometer utilizes hygrometry to determine water potentials of porous media via a chilled mirror to measure the dew point temperature in the headspace above a sample [17]. Recent research has shown the effectiveness of the dewpoint potentiometer for determining water potentials below −0.1 MPa for soils [18,19] and soilless substrates [20]. Moreover, dewpoint potentiometer measurements have demonstrated inaccuracies in pressure plate measurements for mineral soils [21,22]. Curtis and Claassen [23] compared dewpoint hygrometry to pressure plate measurements for inorganic amendments at water potentials of −1.5 MPa, demonstrating more precision with dewpoint hygrometry than with the pressure plates. Fields et al. [15] used dewpoint hygrometry to describe inaccuracies in measuring water retention of highly porous organic soilless substrate components with a pressure plate set at −1.5 MPa.

As improved water management continues to be an imperative focus for horticultural crop production, refining the characterization of water storage and availability in horticultural substrates is critical. Therefore, the objective of this research was to utilize dewpoint hygrometry to assess VWC of traditional soilless substrates at water potentials <−1.0 MPa. Additionally, we compared the VWC at substrate water potential values near −1.5 MPa measured through dewpoint hygrometry with values attained through other accepted methodologies in the literature and make inferences upon the viability of utilizing dewpoint hygrometry in soilless substrate science.

## **2. Materials and Methods**

## *Preparation of Substrate Components and Composites*

Substrate components tested were coconut coir pith (Densu Coir, Toronto, ON, Canada), horticultural grade Sphagnum peat moss (Premier Tech, Riviere-du-Loup, Quebec Canada), aged pine bark (PB; Pacific Organics, Henderson, NC, USA), pine wood chips (PWC), and shredded pine wood (SPW) with examples shown in Figure 1. The coir was hydrated from compressed bricks with tap water and then fluffed by hand to reconstitute the material. The peat was removed from the compressed bale, fluffed by hand, hydrated, and screened by hand through a 1.25 cm screen to prevent any larger aggregates (foreign debris) from being included in the sampling. Pine bark derived from harvested loblolly pine (*Pinus taeda* L.) trees was processed in a hammer mill through a 16 mm screen, windrowed, and allowed to age for nine months.

**Figure 1.** Examples of the base materials used in this research, including (**A**) Sphagnum peat moss; (**B**) coconut coir pith; (**C**) shredded pine wood; (**D**) pine wood chips; and (**E**) aged pine bark.

For the pine wood materials, 12-year old loblolly pine trees were harvested at ground level, de-limbed, and processed through either a wood chipper or a wood shredder with bark intact. The pine trees used to create PWC were harvested on 9 December 2011 and processed through a DR Chipper (18 HP DR Power Equipment, Model 356447; Vergennes, VT, USA) on 3 January 2012 to produce the coarse wood chips that were then hammer-milled (Meadows Mills, North Wilkesboro, NC, USA) through a 6.35 mm screen on 5 January 2012 yielding the final PWC product (Figure 2). The pine trees used to create the SPW were harvested on 12 December 2011, shredded in a Wood Hog shredder (Morbark; Winn, MI, USA) on 9 January 2012 to create the coarsely shredded wood that was then hammer-milled through a 6.35 mm screen on 10 January 2012 yielding the final SPW product (Figure 2).

**Figure 2.** Pre and post hammer mill processing on the shredded pine wood (SPW) and pine wood chips (PWC). Shredded wood (**A**) passed through the hammer mill to produce SPW (**B**). Chipped pine wood (**C**) is passed through a hammer mill to produce PWC (**D**).

No additional screening was needed for the coir, PB, PWC, or the SPW. After acquiring, preconditioning, or creating the substrate materials all were placed in 60 L plastic bags, sealed, and stored in a controlled environment laboratory until experiment initiation.

On the day of sampling, bags were carefully turned upside down and mixed to ensure uniformity of the contents/materials, after which a representative sample of 14 L was collected. Moisture content (MC = mass of water/total mass) was measured for each material and adjusted to 55% by weight using procedures described by Fonteno and Harden [24]. Two substrate composites also tested in this experiment included a commercially available growing mix comprised of "Canadian sphagnum peat moss, pine bark, perlite, and vermiculite" (Fafard 4P; Sungro, Anderson, SC, USA) and an 80:20 (by vol.) peat: perlite substrate derived from peat (Berger Tourbe de Shaigne Blonde Golden; BP-P; Quebec Canada) that was taken from a compressed bale, loosened/fluffed by hand, and hydrated to 55% MC before being amended with horticultural grade perlite (Carolina Perlite Company, Gold Hill, NC, USA).

Measurements and Analysis. An initial test was conducted using a dewpoint potentiometer (WP4C, Decagon; Pullman, WA, USA) to determine water potentials for each substrate component and composite materials as they air dried. Based on these results, samples were prepared at target MC for each component and composite that fell within the water potential ranges of −1.0 to −2.0 MPa, and allowed to equilibrate for 24 h. Fifteen samples for each substrate component and twelve samples for each composite were evaluated. Five stainless steel sampling dishes (1.1 cm tall × 3.7 cm i.d.; Decagon; Pullman, WA, USA) were loosely filled to approximately half full (0.5 cm depth) from random locations in the prepared samples at each of the predetermined (through the initial test) MCs for each substrate. The dishes were immediately sealed with plastic lids and Parafilm® (American Can Co.; Greenwich, CT, USA) to prevent evaporative water loss. The samples were then individually analyzed for substrate water potential utilizing the dewpoint potentiometer in precision mode. Only samples that resulted in measures between −1.0 and −2.0 MPa were utilized, resulting in six to twelve utilized

measures (12, 9, 8, 7, and 6 measurements for coir, peat, shredded wood, pine bark, and wood chips, respectively) for the components (see example; Figure 3) and four and five samples for the two composites. The reduction in measures within the −1.0 to −2.0 MPa range in the composites is due to the heterogeneous nature of these materials resulting in less uniformity in drying.

**Table 1.** Estimated substrate-bound water contents, container capacity and available water for substrate components and composites determined via dewpoint hygrometry measured between −1.0 and −2.0 MPa.


<sup>z</sup> Mean substrate-bound water content across substrate water potential between −1.0 and −2.0 MPa ± standard deviation (SD). <sup>y</sup> CC = container capacity values from NCSU porometer test. <sup>x</sup> AW = available water content calculated as difference between mean CC and SBW content. <sup>w</sup> Statistics preformed down columns using Tukey's HSD. Means with the same letter are not statistically different. <sup>v</sup> SPW = shredded pine wood made from loblolly pine (*Pinus taeda*) logs that were shredded prior to processing in a hammer mill through a 6.35 mm screen. <sup>u</sup> PWC = pine wood chips made from loblolly pine logs that were chipped prior to processing in a hammer mill through a 6.35 mm screen. <sup>t</sup> Mix 1 = Composite of Peat:perlite 80:20 (by vol.) <sup>s</sup> Mix 2 = Fafard 4P (Sungro,/Anderson, SC, USA).

**Figure 3.** Example of individual sampling measurements of coconut coir. Variation of volumetric water content (VWC) was <1%. Data used to calculate values presented in Table 1.

Subsequently, mass wetness (MW = mass of water/mass of solid) was determined by placing the samples in a drying oven at 105 ◦C for 48 h to attain dry weights. Mass wetness for the samples was transformed to volumetric water content (VWC = volume of water/volume total) through: MW × Db of the material/density of water (1 g/cm3) = VWC. Since both VWC and water potential were measured (i.e., neither were precisely controlled), values for both are presented and discussed according to their range and average. Bulk density and container capacity (CC) values were obtained using the NCSU porometer analysis following procedures of Fonteno and Harden [24] on three samples for each substrate. The values for VWC corresponding to the water potential range of −1.0 to −2.0 MPa were used as an estimate for soil-bound water and subtracted from the CC obtained from porometer analysis to obtain an estimate of available water holding capacity.

#### **3. Results and Discussion**

## *3.1. Estimating Water Availability at Low Water Potentials*

Values for substrate-bound water obtained for all substrate components tested across the range of −1.0 MPa to −2.0 MPa were generally between 3% and 5% (Table 1), with PB providing the highest VWC at 7 to 8%. The PB had a higher VWC within the −1.0 to −2.0 MPa substrate water potential range likely resulting from reduced uniformity in the pore size distribution, as well as increased intraparticle porosity. As substrate water potential decreased, increased quantities of water became trapped within bark particles, thus limiting the water loss from the material. Furthermore, the majority of the accessible water present in the substrate at water potentials <−1.0 MPa exists primarily as hygroscopic water (water that is bound to particle surfaces). Previous research has demonstrated that at much higher substrate water potentials (i.e., −10 to −300 hPa or −0.0001 to −0.03 MPa) the VWC of PB is much lower than the other materials in this study [25]. This is likely from dual-porosity that is more evident in the PB than the other materials. Large pores created by irregular and large particles in PB readily drain at higher substrate water potentials, with smaller pores being either inaccessible or held more tightly at substrate water potentials between −1.0 and −2.0 MPa. The two composite substrates had similar VWCs (5 to 11%) at substrate water potentials in the range of −1.0 to −2.0 MPa (Table 1). These were only slightly larger than the other substrates, which indicates that the primary components in these substrates (i.e., peat or pine bark) dominate the hydraulic characteristics of the composites.

As expected, there was a large range in CC among the substrate materials and composites (Table 1). While many factors influence the CC of the substrate, the similarity between PB and PWC is likely a result of particle size and shape. Peat, coir, and SPW are more fibrous in structure, while PB and PWC had "plate-like" and "blockular" structure, respectively. The difference in particle size and shape can influence pore distribution and connectivity which has a great influence on water retention and CC, due to changes in the ratio of gravitational to capillary pores. Similar CC between the coir and Mix 2, as well as between SPW and Mix 1 highlight the similarities between fibrous materials and fiber-dominated mixtures (Table 1).

The estimated AW storage capacities for both peat and coir were >70% by volume. SPW, PWC, and PB had much lower AW (approx. 48%, 37%, and 35%, respectively). The differences in AW were primarily due to differences in CC, as there was little (<4%) difference observed in SBW among components. Moreover, by calculating the proportion of the water at CC that is AW (i.e., AW/CC from Table 1) coir, peat, SPW, and PWC are similar 94.1%, 94.9%, 90.9%, and 88.9%, respectively). However, the proportion of CC that is AW in PB is much lower at 82.7%. From data presented herein, PWC would appear to have similar properties to more traditionally used greenhouse substrate aggregates, such as PB and perlite. The SPW possessed a similar VWC at low substrate water potentials, yet a significantly greater CC, yielding increased AW (Table 1), which allows it to be incorporated into a substrate to increase drainage, while still retaining moisture needed for plant growth.

A review of the literature was performed and selected references associated with measuring soilless substrate VWC at substrate water potentials of −1.5 MPa were included in Table 2. The current accepted normal range of SBW is 23 to 35% by volume [26]. In fact, current best management practices for nursery growers recommend substrate SBW between 25 and 35% by vol. [27]. This acceptability range is further evidenced as much of the previous research utilizing ceramic pressure plates identifies commonly used substrates as having SBW within these ranges (Table 2). For example, Wright and Browder [28] reported SBW of PB as 26.6% and pine tree substrate at 23.6% (by vol.), respectively. This may be a significant overestimation of SBW (26.6% and 23.6% by vol. as compared to 7.5% and

4.8% by vol.), and therefore a large underestimation of AW (~20% by vol.) for these substrate materials. Water measurements (at −1.5 MPa) as high as 39.0% by volume have been reported in PB substrates using ceramic pressure plates [29]. With previous reports of miscalculations of soilless substrate SBW through ceramic pressure plate analysis at tensions <−1.0 MPa [12–15], it is entirely possible that many values within the literature are overestimating SBW.


**Table 2.** Survey of soilless substrate VWC at substrate water potentials of −1.5 MPa as measured through pressure plate extractors.

Previous research from the authors of this publication involved utilizing dewpoint hygrometry to assess the water potential of substrate components and field soils that had been squeezed to −1.5 MPa on ceramic plates [15]. The authors found that the water in the mineral soil samples did equilibrate at ~−1.5 MPa; however, the hydraulic connection between the coarse substrate components (bark, peat, and perlite) was broken at approx. −0.3 MPa, preventing additional water loss from the samples (Figure 4). Further investigation in that research showed that when peat and pine bark samples were squeezed at −0.1 and −0.3 MPa, the assessed water potential was close to the applied pressure. This leads the authors to believe that highly coarse substrate materials are not coming to equilibrium with pressures exceeding 0.3 MPa in traditional pressure plate extractors. This information further supports the hypothesis that the pressure plate analysis is overestimating water in samples at very low substrate water potentials.

**Figure 4.** Substrate water potential of individual substrate and soil components assessed via dewpoint hygrometry, after being squeezed at −1.5 MPa on ceramic pressure plates in a volumetric pressure plate extractor. Data utilized from Fields et al., 2013 [15].

#### *3.2. Gravimetric vs. Volumetric Water Contents*

The authors also suggest a paradigm shift in discussing moisture contents that evolved during this work. The term used to describe the amount of moisture in a substrate had two forms: MC (expressed on a weight basis) and VWC (expressed on a volume basis). Initial moisture content for substrates is usually expressed as MC; however, almost all discussion of water content as a result of irrigation is expressed in terms of VWC. The moisture contents in these experiments were considered using both forms. In this case, MC was converted to VWC for comparisons in the table and figures (using yet another measure, MW). For example, coir at −1.5 MPa water potential has a resulting VWC of 7.62% (Table 1), which is equivalent to MC of 50% by weight. During the initial potting of most greenhouse crops, it is important to have adequate moisture in the substrate [27], and generally speaking, many growers tend to use ~50% MC substrates for planting. These results suggest that at this MC, coir is already at a water potential <−1.0 MPa, within the currently accepted range of plant unavailable water. Kiehl et al. [41] showed water stress symptoms occurring in plants at −16 kPa, much less negative than the −1.0 MPa of coir at 50% MC. These high (50%) MC values convert to much lower VWC values due to the very low bulk density of organic components. Moisture contents of 50% are considered to be heavy for transportation, in fact, coir is normally dried, compressed, and formed into blocks for shipping [42]. Peat is normally compressed two to three times and bailed at a MC of about 20–25% (personal observation) for shipping purposes, which is significantly lower than at substrate water potential of −1.5 MPa (37% MC). This establishes that not only is proper hydration

of substrates important for potting/planting, but previously accepted MC levels are essential in the plant unavailable range. This also implies that recently potted plants should not be allowed to "sit" for prolonged periods of time before initial hydration (i.e., water) is applied.

## **4. Conclusions**

The use of dewpoint hygrometry allowed estimates of soil-bound water in the water potential range (1.0 to 2.0 MPa) typically considered plant unavailable. These estimates are much lower than values previously reported for similar substrate components using pressure plates. The authors agree that problematic measures <−1.0 MPa can potentially overestimate SBW due to reported issues associated with highly porous organic materials in pressure plate analysis. As such, it is important that more efforts are utilized to investigate SBW from a substrate standpoint and identify more precise methods of analysis to truly identify substrate water relations at low water potentials. The use of dewpoint hygrometry has the potential to improve the estimation of SBW for substrate analysis. If further investigations find that dewpoint hygrometry measures are in fact accurate, best management practices and acceptable ranges should be updated.

**Author Contributions:** Conceptualization, J.S.F., W.C.F., B.E.J., J.L.H., J.S.O.J.; methodology, J.S.F., W.C.F., B.E.J., J.L.H., J.S.O.J.; formal analysis; J.S.F., W.C.F.; investigation, J.S.F.; writing—original draft preparation, J.S.F.; writing—review and editing, J.S.F., W.C.F., B.E.J., J.L.H., J.S.O.J.; supervision, W.C.F. and B.E.J. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article*

## **Comparison of Water Capture E**ffi**ciency through Two Irrigation Techniques of Three Common Greenhouse Soilless Substrate Components**

**Brian A. Schulker 1,\*, Brian E. Jackson 1,\*, William C. Fonteno 1, Joshua L. Heitman <sup>2</sup> and Joseph P. Albano <sup>3</sup>**


Received: 24 July 2020; Accepted: 10 September 2020; Published: 14 September 2020

**Abstract:** Substrate wettability is an important factor in determining effective and efficient irrigation techniques for container-grown crops. Reduced substrate wettability can lead to lower substrate water capture, excessive leaching and poor plant growth. This research examined substrate water capture using surface and subirrigation under three initial moisture contents (IMC). Sphagnum peat moss, coconut coir, and pine bark were tested at IMCs of 67% 50%, and 33%. Substrate water capture was influenced by both IMC and irrigation technique. Surface irrigation increased the water capture of coir and peat, regardless of IMC, whereas IMC influenced pine bark water capture more than irrigation method. Surface-irrigated coir at or above 50% IMC provided the greatest water capture across all treatments. The first irrigation had the highest capture rate compared to all other events combined. Container capacities of pine bark and coir were unaffected by IMC and irrigation type, but the CC of peat was less by ~ 40% volumetrically under low IMC conditions. Coir, had the greatest ability to capture water, followed by pine bark and peat, respectively. Moisture content, irrigation type and component selection all influence the water capture efficiency of a container substrate.

**Keywords:** irrigation; soilless substrates; water; coconut coir; initial moisture; mass wetness; peatmoss; pine bark; wettability; capillary rise; container capacity; capture rate

## **1. Introduction**

Water use efficiency of horticultural soilless substrates represents one of the biggest variables in container plant production. With nearly 21,500 acres of land devoted to greenhouse operations in the U.S., representing a 148% increase since 1998, growers specializing in container plant production need to be able to understand how irrigation specifics impact water use efficiency of soilless substrates [1]. Greenhouse production uses less water and fewer nutrients than many agricultural resources [2], and decrease crop water requirements by up to 40% compared to open field cultivation [3–5]. As water quality, conservation, scarcity and operational costs increase, plant producers must adopt new strategies to improve the sustainable use of water to confront water-climate policies [6–8]. In order for growers to meet these increasing regulations in water use, we need to increase the overall understanding of irrigation techniques.

Two parameters affecting water efficiency in substrates are container capacity (CC) [9–12] and wettability [13–17]. Both wettability and CC are vital to the wetting of a substrate, however neither

completely describes the effectiveness of water capture during irrigation [18]. Container capacity is the maximum amount of water a substrate can hold after wetting and drainage. Wettability includes a liquid's ability to spread laterally at and below the surface of a material [19]. In substrates, proper wettability helps to provide a uniform distribution of water throughout the rooting environment [20]. Moisture content in substrates affects both wettability and CC of a substrate. Fields et al., [21] showed the variability in CC based on substrate and initial moisture content (IMC), with coir and pine bark being less variable than peat. Initial moisture content in this context references the moisture content prior to an irrigation event. At low moisture conditions, peat can have a ~30% lower CC than at high moisture conditions [11]. In mineral soils, hydrophobicity is caused by organic residues coating the mineral materials from the breakdown of organic matter. In substrates, most components are composed almost entirely of organic materials which complicates the nature of hydrophobicity. As organic materials naturally break down, the intensity of hydrophobicity can change, which then alters the substrate's behavior during rewetting [22,23]. Hydrophobicity issues also arise as organic materials dry, and materials such as peat and pine bark begin to see reductions in water capture based on repellency [13,24]. Adequate substrate particle structure, stability, density, and CC are needed to allow water movement through the containers [25,26].

Most irrigation is applied to the top of the soil column. However, in containers, the irrigation delivery direction can be reversed and delivered from the bottom (subirrigation). Irrigating from below can require a finer textured, micro-pore abundant substrate to take up water mainly through capillary action [27]. Conversely, greater air space (AS) and pore size diversity favor surface irrigation methods. Capillary action, the movement of water from a saturated zone upward into an unsaturated zone through surface tension and soil matric potential, provides water and nutrients to the plant root [28]. Subirrigation is a combination of flooding from a perched water table and capillary rise. Ebb and flow subirrigation, was found to reduce water use by ~40% compared to hand watering [29,30]. The confluence of these factors combine to play a pivotal role in the effectiveness of water uptake in specific combinations of irrigation method and substrate components. Water transport research in mineral soils [31] provides the basis to understand soilless substrate systems, but the substantial differences in physical properties and their accompanying calculated values between soil and soilless substrates requires substrate-specific research.

Research has identified the impacts hydrophobicity can have on the wettability of some horticultural substrates [13,19,20]. However, few have studied the influence of irrigation delivery method on the ability of substrates to capture water or rehydrate. In soilless substrates, water distribution in the container can largely change due to a substrate's hydrophobicity, physical characteristics (texture/particle size), as well as the irrigation method used. The objective of this study was to characterize the water capture and retention of three substrate components based on irrigation technique and IMC.

#### **2. Materials and Methods**

#### *2.1. Preparation of Substrates*

On 11 April, 2019, sphagnum peat (Premier Pro-Moss, Quakertown, PA) was hydrated and fluffed to an initial IMC of 70% (by weight; ~2.5 *w*). To do that, peat was removed from the compressed bale and placed into a large tub, water was then added in 3 L increments after which peat was mixed/agitated by hand to allow water absorption. Moisture levels were then tested using an Ohaus MB27 soil moisture determination balance (Ohaus Corp., Parsippany, NJ) to determine further water additions needed to bring peat to an initial moisture of 70%. On 15 April, 2019, three compressed bricks of coconut coir (Densu Coir, Ontario, Canada) were hydrated individually by adding 14 L of water (in 1 L increments), by hand, until the coir was completely broken apart. Moisture levels were then tested using the soil moisture balance to determine further water additions needed to bring coir to an initial moisture of 70%. On 16 April, 2019, loblolly (*Pinus taeda* L.) pine bark (Pacific Organics, Henderson, NC) which

had been aged in outdoor windrows for four months and specifically engineered (hammer milled and screened) to have a CC of 55% volumetric water content (VWC). The volume of pine bark was measured out, initial moisture level was tested and recorded before the bark was further hydrated to a moisture level of 70%.

Each substrate component was tested under three IMC treatments. The most common and recommended IMC for potting soils has a mass wetness of 1.0 g g−1. To test effects of IMC, each component was also brought to half (0.5 g g<sup>−</sup>1) and double (2.0 g g−1) this normal level, which resulted in percent IMCs of 33%, 50%, and 67% by weight. To do this the wet weights and dry weights were determined by taking 500 g subsamples of each substrate, weighing, drying, and then reweighing. Substrate samples were wet to an initial IMC of 70% IM, before being air-dried down to initial IMC's of 67%, 50%, and 33% IMC. Initial IMC and total weight of each sample were used to calculate how much water needed to be lost to reach initial IMC's of 67% IM, 50% IM, and 33% IM. Using a 160 cm × 49 cm × 69 cm four-tier PVC-enclosed dehumidifying chamber, substrates were allowed to air dry to desired wetness. Once the target initial IMC was reached, samples were transferred to plastic bags and sealed to prevent further water loss, while allowing the substrate to reach moisture equilibrium.

#### *2.2. Particle Size*

Particle size distribution (Table 1) was performed on three 50 g oven dried samples of each substrate with 7 sieves. The sieve sizes used were 6.3 mm, 2.0 mm, 0.71 mm, 0.5 mm, 0.25 mm, 0.11 mm, and the bottom pan to collect fine particulates. The 7 sieves (6 sieves and the pan) were stacked together and substrate samples were poured into the top sieve, and placed into the RX-29 Ro-Tap sieve shaker (278 oscillations/min, 150 taps/min; W.S, Tyler, Mentor, OH). The sieves and pan were shaken for five min and the particle fractions retained on each sieve and the amount collected in the bottom pan (representing the smallest particle fractions) were weighed.


**Table 1.** Particle size distribution of three traditional greenhouse substrate components.

<sup>z</sup> Particle size distribution calculated on a dry weight scale using means of three oven-dried samples. <sup>y</sup> Coarse = particles that are greater than 2.0 mm in diameter. <sup>x</sup> Medium = particles that are less than 2.0 mm but greater than 0.5 mm in diameter. <sup>w</sup> Fines = particles that measure less than 0.5 mm in diameter. <sup>v</sup> Values are means of percentages of the total sample. <sup>u</sup> Statistics are determined down columns (denoted by an uppercase letter) and across rows (denoted by a lowercase letter) using Tukey's honestly significant difference to determine similarities and differences across all components.

#### *2.3. Surface Irrigation*

In order to determine the effects of IMC with surface applied irrigation, this experiment followed the procedures described by Fonteno et al. [18] and Fields et al. [20]. The equipment consisted of a transparent cylinder, 5 cm i.d. <sup>×</sup> 15 cm h–1, with a mesh screen (mesh size 18 <sup>×</sup> 16; New York Wire, York, PA, USA) (Figure 1A), attached to one end, using rubber pressure plate rings (Soil moisture Equipment Corp., Santa Barbara, CA, USA); a 250-mL beaker; a 250-mL funnel; as well as a 10 mL plastic vial (4-cm diameter) with five evenly spaced 2.33 mm diameter holes in the base to act as a diffuser displayed in

Figure 1. This allowed the water dripping through the funnel to be evenly dispersed through the five holes onto the surface of the substrate in the cylinder.

**Figure 1.** Surface irrigation apparatus. (**A**) Funnel, sparatory funnel with stopcock. (**B**) Water diffuser with O-ring above cylinder. (**C**) Sample cylinder with 200 mL of substrate.

The transparent cylinders were packed with each substrate component to have a weight within 5% of other samples of the same component. To achieve this, cylinders were filled (by weight) with substrate then raised 12 cm off a flat surface, then tapped four times to bring the top of all 4 replications to 10 cm from the base of the cylinder, representing 200 mL of substrate and providing similar Db across all replications. With three substrates, at three IMCs, and four replications there were a total of 36 experimental samples. After the cylinders were packed, each was fitted with a diffuser and placed in the clamps held up by a ring stand, just under the separatory funnel (Figure 1). Two hundred mls of water was poured into the separatory funnels and allowed to drip onto the surface of the substrate at an average rate of 40 mL min−1, using the stopcock valve to control the flow. Water was applied in 10 individual hydration events. Substrates with an IMC of 33% required the rate of flow to be less to prevent ponding of water on the substrate surface which would have created a hydraulic head greater than 0.5 cm and alter the influence of any native hydrophobicity in the sample. Water was passed from the separatory funnel, through the diffuser, and onto the surface of the substrate. With the help of gravity, water was able to penetrate the surface of the substrate and percolate through the 10 cm depth. Some of the water volume was absorbed as it moved through the substrate, the rest was collected out of the bottom by a 250 mL beaker. After ~5 min, water flow ceased; the substrate was then held at equilibrium for two min. The effluent volume was measured and recorded while water retained was calculated by subtracting the amount of water applied (200 mL) from the amount of effluent captured. With the total event lasting ~7 min, 5 min time intervals were measured out in between events to keep treatments even. This procedure was repeated for each of the 10 hydration events.

## *2.4. Subirrigation*

In order to understand how IMC influences substrate water capture through subirrigation, this experiment was conducted using materials and modified procedures described by Fonteno et al., [12]. The same transparent cylinders as described in surface irrigation above were prepared in the same way (Figure 2), The subirrigation method used an ebb and flood irrigation system (Hawthorn Hydroponics, Vancouver WA) 60.96 cm wide by 121.92 cm in length (Figure 2). Water was introduced into this system via a faucet and controlled through a series of gate valves connected to the system (Figure 3B). Water was maintained at a continuous height with a flow rate of ~21 L min<sup>−</sup>1. To be able to control the height of the water while also having a steady flow into the bench, a standing copper pipe was cut to allow water to be held at 2.5 cm at a steady state (Figure 2C).

**Figure 2.** Subirrigation system. (Left, right, bottom) (**A**) Cylinder (15 cm × 5 cm) with rubber pressure plate ring at base with mesh screen with DIA representing the cylinder diameter. (**B**) Ebb and flood subirrigation system. (**C**) Separated full system (from left to right) with central weight, three copper leveling pipes, large steel ring (for raising wire screen off surface), black wire mesh screen, fully assembled system.

**Figure 3.** Ebb and flood subirrigation system. (**A**) Container capacity testing with 2 kg aluminum weights. (**B**) Partially constructed ebb and flood unit with aluminum rings with mesh screen. (**C**) Fully constructed system complete with packed substrate cylinders.

The transparent cylinders were packed in identical manner as the samples used in the surface irrigation system. Cylinders were then placed on an elevated mesh screen to optimize the lower surface area exposure to water. The unit was then filled with water. It took approximately one minute for the water to reach the bottom of the cylinders and another minute for the water to reach 2.5 cm above the base. At that time, water flow input equaled output, allowing constant flow of water without a change in water level. The substrate samples were kept in the unit for five minutes for each of the hydration events. Once an event was finished, water drained from the unit for one minute before each cylinder was weighed. The weights were used to calculate the amount of water captured by the substrate by subtracting it from the initial weight of the packed cylinders. This procedure was repeated 10 times (10 hydration events), with a total time of hydration equaling 50 min.

## *2.5. Container Capacity*

After the final hydration event was complete and final weights were taken, CC was then determined for each cylinder. The cylinders were returned to the ebb and flood unit (Figure 2), and CC was

determined using a modified version of the NC State University Porometer Method [32]. After placed in the subirrigation unit, an aluminum weight of approximately 2 kg was placed on the top of each cylinder to prevent tipping and buoyancy (Figure 3A). The samples were then saturated from below by allowing water to flow into the unit until it reached 1/3 of the height of the sample (three cm from the base of the sample). After two minutes, additional water was allowed to enter the unit until reaching 2/3 of the height of the cylinder, or six cm from the base of the cylinder. After an additional two minutes, the water was applied until reaching the top of the sample within the cylinder (Figure 3A), 10 cm from the base of the cylinder. After saturating in the system for an additional 30 min, the water was drained and samples were reweighed to record changes in weight (water captured and retained). Samples were then placed into a forced-air drying oven at 105 ◦C for 48 h to dry, after which each sample was weighed and the dry weights were used to determine total water retained and IMC.

#### *2.6. Water Capture Rate*

Water CR was calculated for subirrigated substrates using a modified version of the flow rate formula to account for variables in this experiment, the equation was written as:

$$\text{CR} = \frac{\text{C}\_{\text{i}} - \text{C}\_{\text{p}}}{\text{t}} \tag{1}$$

where CR is the mL/min of water captured by the substrate after one irrigation event, Ci [water capture (g) in the initial irrigation event] is the weight of the substrate after the present irrigation event (minus the weight of the cylinder), Cp (previous water capture in grams) is the weight of the substrate (minus the cylinder) taken after the previous irrigation (for the first irrigation event, Cp is equal to the pack weight of the cylinder (minus the weight of the cylinder), t is the amount of time per irrigation (in minutes). For surface irrigated samples that have a defined volume of water passing through the substrate, the equation was written as

$$\text{CR} = \frac{\text{A}\_{\text{w}} - \text{E}}{\text{t}} \tag{2}$$

where CR is the amount of water captured by the substrate after one irrigation per unit time (in mL min<sup>−</sup>1), Aw is the amount of water applied to the substrate per irrigation event (in this case, 200 mL), E is the effluent captured after the individual irrigation event (in mL), t is the amount of time per irrigation event (minutes).

#### *2.7. Water Capture Curves*

The IMCs of 33%, 50%, and 67% were all determined by weight. Wettability curves were determined by VWC to describe the amount of water contained within the substrate. These curves show a VWC reading at event zero, and represent the percent VWC at the IMC. Therefore, an IMC of 50% (by weight) was actually 12% to 15% *v v*−<sup>1</sup> (moisture) for peat. For coir, an IMC of 50% ranged from 9% to 11% *v v*<sup>−</sup>1, and for pine bark (at 50% IMC) they were 16% to 18% *v v*<sup>−</sup>1.

#### *2.8. Capture E*ffi*ciency Values*

In order to provide both statistical and numeric comparisons, water capture efficiency of the substrates was described in three ways: (1) first hydration, (2) final hydration and (3) CC. First hydration was the amount of water absorbed by the substrate after one irrigation event, and compared across all substrates and moisture levels. Final hydration was the amount of water absorbed by the substrate after the tenth irrigation event. Container capacity was the maximum water content the sample could hold after saturation and drainage. Physical properties of the substrates, including CC, AS, total porosity (TP), and bulk density (Db) were derived using the NC State University Porometer method [20] with three representative samples of each substrate, and CC is reported in Table 2.


**Table 2.** First hydration (H1), container capacity (CC), and final hydration (H10) of three substrate components analyzed at three different initial moisture contents (IMC) irrigated by either subirrigation or surface application.

<sup>z</sup> H1 = the amount (by volume) of water that is absorbed by the substrate after one irrigation event. <sup>y</sup> H10 = the amount (by volume) of water that is absorbed by the substrate after the final hydration event. <sup>x</sup> CC = maximum volumetric moisture content attained by sample. <sup>w</sup> Significance: Linear (L) and Quadratic (Q) regression significance test, NS = nonsignificant, \*\*\* *p* ≤ 0.001, \*\* *p* ≤ 0.01 \* *p* ≤ 0.05 down all columns for peat, coir, and pine bark. S \*: Linear (L) and Quadratic (Q) regression significance test, NS = nonsignificant, \*\*\* *p* ≤ 0.001, \*\* *p* ≤ 0.01 \* *p* ≤ 0.05 across rows for individual substrates, moisture contents, and irrigation techniques. <sup>v</sup> Statistics using Tukey's honestly significant difference with alpha = 0.05 are given down individual columns at a given initial moisture content. Means with the same letter are not statistically different.

Statistics were determined using SAS v. 9.4 (SAS Institute; Cary, NC, USA). A Tukey's HSD test with alpha = 0.05 was used to identify differences and similarities between substrates at individual IMCs and irrigation events. This test also determined the similarities and differences of CC, first hydration, and final hydration across substrates, IMCs, and irrigation techniques. Both linear and quadratic regression was performed and significance was determined using *p* values with significance ranging from > 0.001 to 0.05. An analysis of variance test was conducted to test the effects of initial IMC and irrigation technique on CC, first hydration and final hydration within individual substrate components.

#### **3. Results**

#### *3.1. Particle Size*

Substrate particle size analysis was performed on all three substrates, with the results displayed in Table 1. Coir represented the substrate with the highest percentage of particles smaller than 2.0 mm, representing 93.8% of all particles tested while pine bark showed the highest percentage of coarse particles with a value of 53.6%. Peat occupied a middle ground between coir and pine bark with 13% more coarse particles than coir, but still 34% less than that of pine bark.

#### *3.2. Coir Water Capture*

The VWC curves for coir (Figure 4A–C) indicated a pattern directly related to IMC. Regardless of IMC, the first hydration event had the most water absorbed by the substrate compared to all other irrigation events. The IMC affected the amount of water absorbed in the first hydration, and increased as the IMC increased. For surface irrigation, coir at 33% IMC (Figure 4A) needed four irrigation events to reach maximum absorption through irrigation. Coir at 50% IMC (Figure 4B) needed two events to reach maximum absorption and at 67% IMC (Figure 4C) needed just one. For subirrigation, IMC contributed to the ability of coir to absorb water across all 10 events, however, coir never reached a steady state or maximum absorption at any initial moisture level with subirrigation. At 50% IMC, coir reached a final hydration of 73.5% VWC through surface irrigation and 51.6% VWC with subirrigation (Table 2). However, at 33% IMC, coir was ~20% VWC below the CC after the final hydration in both irrigation techniques. Coir samples at 67% IMC reached near CC in one irrigation using the surface application technique, with a final hydration value < 1% below the CC.

**Figure 4.** Substrate water capture volumetric water content curves for peat, coir, and pine bark over ten irrigation events at three moisture contents and two irrigation techniques. With (**A**) coir at 33% IMC, (**B**) coir at 50% IMC, (**C**) coir at 67% IMC, (**D**) representing peat at 33% IMC, (**E**) peat at 50% IMC, (**F**) peat at 67% IMC, (**G**) pine bark at 33% IMC, (**H**) pine bark at 50% IMC, and (**I**) pine bark at 67% IMC.

The capture rate of coir was affected by irrigation method. Capture rates for coir were greatest at 50% IMC (Figure 5), where 50% IMC captured ~60% VWC of water through one irrigation event whereas 67% IMC captured ~50% VWC of water in surface irrigation. However, with subirrigation the differences were smaller. Initial moisture contents of 33%, 50%, and 67% were within 8% of total water captured (volumetrically) between each increase in moisture. Also, as IMC increased, the difference between events one and 10 were smaller. With surface irrigation, as IMC increased, fewer events were needed to reach maximum capture. In subirrigation, the effect was similar, although 33% IMC and 50% IMC showed minimal differences in water capture.

**Figure 5.** Water capture rate (CR) for peat, coir, and pine bark over ten irrigation events at three moisture contents and two irrigation techniques. With (**A**) coir at 33% IMC, (**B**) coir at 50% IMC, (**C**) coir at 67% IMC, (**D**) representing peat at 33% IMC, (**E**) peat at 50% IMC, (**F**) peat at 67% IMC, (**G**) pine bark at 33% IMC, (**H**) pine bark at 50% IMC, and (**I**) pine bark at 67% IMC.

#### *3.3. Peat Water Capture*

The VWC curves (Figure 4D–F) for peat, similar to coir, identified a pattern related to IMC. With surface irrigation, 33% IMC required six irrigation events to reach maximum absorption through irrigation while 50% IMC needed just five irrigation events (Figure 4D,E), showing very little difference between the two IMC levels. At 67% IMC, maximum absorption was reached in two surface irrigation events. Comparing the surface and subirrigated VWC curves (Figure 4D–F), there was ~20% less water taken up by capillary rise than from gravitational flow. Peat contained 81% of particles < 2 mm (coir had 92%) and nearly 17% of particles between 6.3 mm and 2 mm (Table 1). At both 50% IMC and 33% IMC, the final hydration values were less than 20% VWC, with subirrigated peat at 33% IMC reaching less than 20% VWC after 10 irrigations (Table 2). At 67% IMC, the first hydration of subirrigated peat increased by 30% VWC compared to that of 50% IMC.

The water capture of peat was affected by both irrigation method and IMC. Water CRs for peat show the effects of a low moisture condition on hydration with surface irrigation at 33% IMC reaching ~5 mL min−<sup>1</sup> and 50% IMC reaching ~4 mL min−<sup>1</sup> (Figure 5E). Subirrigated peat at 33% IMC and 50% IMC had much lower CRs than surface irrigation, with CRs at or below 1 mL min−1. Peat did not begin to show a change in hydration until IMCs of 67% in both irrigation methods, with very minimal differences between 33% IMC and 50% IMC. Surface CR at 67% IMC, while more variable than subirrigation, reached 16 mL/min while subirrigation peaked ~8 mL min−<sup>1</sup> (Figure 5). The strong interaction between water capture and IMC was clearly evident in peat.

#### *3.4. Pine Bark Water Capture*

Pine bark had a more consistent increase in VWC over the 10 irrigation events than either peat or coir. Of the three substrates, bark contained the highest percentage of coarse particles (Table 1), while also having a similar portion of medium (2.0–6.3 mm) sized particles compared to peat. The VWC curves (Figure 4G-I) identify a degree of consistency between irrigation techniques, regardless of initial moisture level or irrigation method. At 50% IMC and 67% IMC, subirrigation produced maximum irrigation absorption after one irrigation event, with less than 2% difference between first and final hydration. At all IMCs, surface irrigation had higher VWC after the final irrigation event compared to subirrigation, but the difference between surface irrigation and subirrigation after the final hydration was less than 10% VWC (Table 2). Similar to coir, pine bark achieved maximum capture within the first two to three irrigations at 50% IMC and 67% IMC in both irrigation methods. At 50% IMC, the difference in first hydration between surface and subirrigation was < 5% VWC. Similar to coir, the CC was not influenced by IMC or irrigation method. For bark, water capture differences are evident between 33% and 50% IMC. First hydrations increased from ~25% VWC to ~42% VWC (Figure 4G–I). Unlike coir and peat, pine bark showed less variability by irrigation method, with all first and final hydrations between 50% IMC and 67% IMC less than 8% VWC gain between all ten irrigation events. Pine bark water capture was the most consistent of the three substrate materials.

Water CRs also showed similarities across irrigation methods, with subirrigation having the higher CR at 33% IMC (Figure 5G). In just one hydration event, pine bark at 67% IMC reached 0.90 (90%) of its CC by surface irrigation and 0.81 (81%) of its CC by subirrigation. Water CRs for pine bark at 50% IMC and 67% IMC are within 2 mL/min of each other at the first hydration, with surface irrigation representing the maximum CR across all pine bark treatments. The low variability in water capture and high percentage of coarse and medium sized particles (Table 1) allowed pine bark to have a high capture efficiency, regardless of irrigation technique.

#### *3.5. Capture Rate*

Capture rates of surface-irrigated samples were highest in coir, regardless of IMC. At 50% IMC, coir CR was ~23 mL min−<sup>1</sup> before falling to ~3 mL/min by the second irrigation event (Figure 5E). The steep drop in CR was attributed to the substrate's ability to absorb water at such a high rate during the first irrigation, nearly reaching the maximum it could absorb (through surface application) after one irrigation event. With first hydrations capturing 67.6% VWC for 50% IMC and 75.2% VWC for 67% IMC, coir had the highest absorption rate of all three substrates. The lower initial moisture conditions in peat at 33% IMC and 50% IMC impacted CR more so than irrigation method (Figure 5A,B), further reducing the wettability of low-moisture peat. Conversely, the similar responses of pine bark between IMC and irrigation method could be attributed to increasing particle size which might have decreased variability in uptake. The water volumes absorbed by pine bark were less than coir, however the CC of pine bark was ~20% lower volumetrically than that of coir, giving pine bark physically less capacity to hold water.

#### **4. Discussion**

From the data in Figures 4 and 5, it appears that initial moisture content prior to the first irrigation event had the overall greatest effect on the water capture and retention of peat, coir, and pine bark across both irrigation techniques. The container capacity of pine bark and coir were less affected by irrigation technique than peat. Surface irrigation provided the highest water capture in the first hydration across nearly all substrates and IMCs. Peat had higher initial and final hydration values with surface irrigation compared to subirrigation over all IMCs.

At all initial moisture levels, coir was able to take up water. However, IMC altered the intensity of imbibition. With surface irrigation, coir at 33% IMC needed four irrigation events to reach its maximum of 60% VWC. At 50% IMC, coir needed two irrigation events, and at 67% IMC it needed just one for water capture to equal CC. Coir is known to be very hydrophilic, having a sponge-like ability to capture and hold water [33]. Surface irrigation has the additional potential of gravity to draw water through the substrate, allowing droplets to travel a path of least resistance. This allows water to move through macro and mesopores to hydrate the substrate. Conversely, with subirrigation, water must travel via capillary action and against gravity, along particle surfaces and through mostly micro-pores [34]. Initial moisture content did not have an effect on coir CC. With 92% of coir particles ranging from 2 mm or less (Table 1), water retention was very high.

For peat, IMC had the greatest influence on the substrate's ability to capture water with surface irrigation. As is well documented, intensity of hydrophobicity of peat increases at lower substrate moisture contents. These hydrophobic intensities can influence rewetting and impair the physical properties of the substrate [35,36]. At 33% IMC and 50% IMC, peat had difficulty hydrating through the first five irrigation events (Figure 4A). In the case of peat at 33% IMC and irrigated from the surface, water delivery from the separatory funnel had to be slowed to reduce ponding of water on the surface and increasing the hydraulic head at the substrate surface. For perspective, the first hydration at 33% IMC and 50% IMC through surface irrigation for peat was 17.8% and 21.1% (Table 2) respectively, while the first hydration of coir at the same moisture levels reached 35.4% and 67.6% respectively (Table 2). It wasn't until 67% IMC that peat began to capture and retain water during surface irrigation. The large proportion of coarse particles may relate to a greater portion of macro-pores in peat than coir (Table 1). These larger pores allowed surface irrigated water to preferentially flow through peat, even at lower initial moisture levels. Water moved through the large pores with less wetting of the substrate matrix due to increased intensity of hydrophobicity of the peat at both 33% IMC and 50% IMC Conversely, with subirrigation, at 33% IMC, peat was unable to eclipse 13% VWC after 10 irrigation events, representing the lowest first hydration of all treatments. At 50% IMC, peat reached 23% VWC with a final hydration of 37.1% VWC and a CC of 58.0 (Table 2). Compared to coir, irrigation method and IMC both impacted the CC of peat. However, at 67% IMC, the substrate absorbed water in the first irrigation event. The capture potential, or total volumetric water captured, of peat was nearly 40% less than that of coir.

In pine bark, an increase in fine (greater than 0.5 mm) particles has been shown to greatly influence the physical properties (AS and CC), while larger particle sizes had a minor influence on physical properties [9]. Larger particles, for surface irrigation, may provide water with more channels to move through the container, better hydrating the bark as pine bark just doesn't have as much surface area/microporosity for absorption. However, micro and meso-pores have higher abilities to capture and retain water. Larger pore sizes tend to drain more easily under gravitational potentials than smaller pores [37]. Pine bark can have variable properties based on processing, and it can have more AS, lower TP and easily available water than both peat and coir [38]. Pine bark had the most similar water capture and retention across all IMC and, aside from 33% IMC, reached their maximum VWC in the first two irrigation events.

#### **5. Conclusions**

Comparing first hydration, final hydration, and capture rate across all treatments (Table 2), there were varied effects among irrigation methods and IMCs. Peat was highly affected by IMC, the intensity of hydrophobicity was altered by IMC, and irrigation delivery. Coir and peat, through every IMC, captured less water through subirrigation than surface irrigation. Most notably, the higher the initial moisture content in the substrate prior to irrigation, the greater the overall water capture. These three substrate components demonstrated markedly different responses to water capture and retention in response to irrigation method and IMC.

**Author Contributions:** Conceptualization, B.A.S., B.E.J., W.C.F., J.L.H.; methodology, B.A.S., B.E.J., W.C.F., J.L.H., J.P.A.; formal analysis, B.A.S., B.E.J., W.C.F.; investigation, B.A.S.; writing-original draft preparation, B.A.S.; writing-review and editing, B.A.S., B.E.J., W.C.F., J.L.H.; supervision, B.E.J., W.C.F. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

MDPI St. Alban-Anlage 66 4052 Basel Switzerland www.mdpi.com

*Agronomy* Editorial Office E-mail: agronomy@mdpi.com www.mdpi.com/journal/agronomy

Disclaimer/Publisher's Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Academic Open Access Publishing

mdpi.com ISBN 978-3-0365-8761-5