An Environmentally Sustainable Approach for Raw Whey Treatment through Sequential Cultivation of Macrophytes and Microalgae

Abstract

The cheese industry produces substantial amounts of raw cheese whey wastewater (RW), which requires effective treatment prior to environmental disposal. This study presents an innovative sequential batch system that combines macrophyte and microalgal cultivation for RW remediation. The efficacy of Lemna minor MO23 in first-line photobioreactors (PBR-1) and Chlorella sp. MC18 (CH) or Scenedesmus sp. MJ23-R (SC) in second-line photobioreactors (PBR-2) for pollutant removal was evaluated. The nutrient removal capacity of L. minor, CH, and SC was assessed at optimal tolerance concentrations, alongside nutrient recovery from treated RW (TRW) by PBR-1 for microalgae biomass production. The results demonstrate that all three species effectively purified the cheese whey wastewater. L. minor efficiently removed COD, nitrate, phosphate, and sulfate from RW, producing TRW effluent suitable for microalgal growth. CH and SC further purified TRW, enhancing biomass production. CH outperformed SC with a 4.79% higher maximum specific growth rate and 20.95% higher biomass yield. Biochemical analyses revealed the potential of CH and SC biomass for applications such as biofuels and aquaculture. After treatment, the physicochemical parameters of the effluent were within the regulatory limits. This demonstrates that the PBR-1 and PBR-2 series-coupled system effectively purifies and recovers dairy effluents while complying with discharge standards.

1. Introduction

As global concerns about sustainable waste management intensify, industries are increasingly looking for innovative solutions to address environmental challenges. In the dairy sector, the management of raw whey—a by-product generated in abundance during the production of cheese and yogurt—is a major environmental challenge [1]. Raw whey, characterized by its high organic load and nutrient content, poses significant challenges for disposal and treatment [2]. Traditional methods of dealing with raw whey, such as land application and direct discharge to water, have been associated with negative environmental impacts, including soil and water pollution, eutrophication of water bodies, and the release of greenhouse gases [3].

According to sustainable indicators within the dairy industry, the valorization of whey is an excellent alternative that offers both economic and environmental benefits [4,5]. Valorizing whey is therefore essential to meet the requirements of the 2030 Agenda, in particular with regard to SDGs 6, 9, and 12 (drinking water and sanitation; industry, innovation, and infrastructure; responsible consumption and production) [5].

The current raw whey treatment landscape includes a variety of methods, such as anaerobic or aerobic digestion, coagulation–flocculation, oxidation processes, and direct land application [6,7]. While these approaches have shown promise in addressing certain aspects of raw whey management, they often face challenges in efficiently managing the high nutrient content and organic load. As a result, residual environmental impacts and operational inefficiencies persist, highlighting the need for novel, environmentally friendly solutions that can mitigate the adverse effects of raw whey on ecosystems [6,8]. There is an urgent need to explore novel, environmentally friendly solutions that effectively mitigate the negative impact of raw whey on ecosystems.

This research seeks to bridge this gap by proposing an environmentally sustainable approach to raw whey treatment. Specifically, it is hypothesized that a novel method involving the sequential cultivation of macrophytes and microalgae can effectively reduce the nutrient load and organic pollutants in raw whey. This hypothesis is based on the unique properties of these organisms, which offer potential for developing a cost-effective, efficient, and environmentally friendly treatment method.

The coexistence of macrophytes and microalgae in a unified treatment system typically poses challenges for microalgae. In addition to competing with microalgae for nutrients, densely growing floating macrophytes on the surface of the wetlands or tanks drastically reduces light penetration into the water [9], thereby limiting microalgal activity and growth [10]. Concomitant treatment with macrophyte could, therefore, lead to the eradication of the microalgal population from the effluent. To address these challenges, the efficacy of a two-stage photobioreactor system was explored, where Lemna minor L. MO23, a macrophyte, is used in the initial step, followed by microalgae (Chlorella sp. MC18 and Scenedesmus sp. MJ23-R) in a second stage. Lemna minor presents attributes that make it a promising candidate for treating raw whey [11]: (i) high nutrient uptake efficiency, particularly for nitrogen and phosphorus, along with strong phytoremediation capabilities that aid in degrading organic compounds in wastewater; (ii) rapid growth and low maintenance, which make it both efficient and cost-effective; (iii) genetic diversity, allowing for selection of optimal strains. Similarly, microalgae from the genera Chlorella and Scenedesmus have also proven effective in wastewater remediation [12].

To test our hypothesis, batch culture experiments in flasks were conducted to assess the individual tolerance of the microalgae and the macrophyte to raw whey. Next, a sequencing batch photobioreactor system, utilizing Lemna minor in the initial treatment step and microalgae in the subsequent polishing stage, was operated. The effectiveness of the system in removing major pollutants was evaluated.

2. Materials and Methods

2.1. Macrophytes and Microalgae

The study used two species of microalgae and one macrophyte (see Supplementary Material), all isolated from different aquatic environments in southwestern Peru. The microalgae, Chlorella sp. MC18 and Scenedesmus sp. MJ23-R, were isolated from different environments contaminated by industrial effluents and have been described in detail elsewhere [13]. The macrophyte Lemna minor L. MO23 was isolated from the freshwater body ‘Ojo del Milagro’ (Characato, Arequipa, Peru; southern latitude 16°28′8″, northern longitude 71°27′27″). Lemna minor L. MO23 was identified at the Center Herbarium Areqvipense of the National University of Saint Augustin (Arequipa, Peru) through morphological and taxonomic analysis at the species level previously described [14,15,16]. Briefly, morphological observations of Lemna minor (also known as duckweed) were conducted using optical stereoscopy (Motic SMZ-171, Hong Kong, China) equipped with a digital camera (Motic^®, MGT101, Hong Kong, China) and software MoticamBTU 3.0. Stereoscopic visions revealed dark green fronds with an oval and asymmetrical shape, featuring a rounded frond apex, a hyaline frond edge at the base, and an aerenchyma with an apex extended upwards. The morphometric analysis indicated the following values for the frond: 3.50 mm length, 1.95 mm width, 0.50 mm thickness, three contiguous fronds, 21.50 mm² area, and 1.80 length to width ratio; a distance between the base of the frond and the root insertion of 6.15 mm and two roots. These species are deposited in the private culture collection of the Biochemistry and Molecular Biology Laboratory, located within the facilities at the National University of Saint Augustin.

Microalgal inocula were maintained in aerated flasks at an aeration rate of 0.25 vvmin. Lemna minor was maintained in 4 L parallelepiped-shaped photobioreactors (PBR), which have been described elsewhere for the culture of both microalgae [13], at the same temperature. Illumination was provided by an overhead array of 18 W cool daylight LED tube lights (Philips Lighting, Eindhoven, The Netherlands), resulting in an average irradiance of 60 μmol photons m⁻² s⁻¹ at the surface of the culture flasks and PBR, following a 12:12 h light–dark cycle at a temperature of 24 ± 2 °C. The growth medium for both microalgae consisted of BG11 medium [17], while half-strength Hutner’s (HSH) medium [18] was used for Lemna minor. These media were prepared using distilled water and sterilized by autoclaving at 121 °C for 15 min.

2.2. Characterization of the Raw Whey Wastewater

In this study, raw wastewater derived from the dairy product manufacturing process was employed, specifically focusing on the whey generated during cheese production, hereinafter referred to as raw cheese whey (RW). The RW was sourced from an artisanal factory located in the Polobaya district of the Arequipa province, Peru. Figure 1 provides a concise description of the process that generated the effluent, from which the RW sample used in the study was obtained; it also details the proposed RW treatment process, which integrates macrophytes and microalgae.

Briefly, the process begins with the pasteurization of cow’s milk, heated to a temperature of 70 °C in industrial galvanized aluminum containers, each with a capacity of 155 L. After pasteurization, the milk is cooled to a temperature range of 37–38 °C. At this stage, a rennet (CHY-MAX^® M 1000, Hørsholm, Denmark), which promotes curd formation, is added. The coagulation process typically takes about 1 h for a proper reaction. Following curd formation, it is cut into small fragments and allowed to rest for 10 min. Subsequently, the curd is subjected to gentle agitation to facilitate the release of raw cheese whey (RW), followed by the addition of sea salt. The mixture is then filtered through a 2 mm membrane and compressed through a 1 mm pore size to eliminate residual RW and attain the desired consistency. Finally, the resulting mass is poured into circular molds, each with a production capacity of 1 kg, to shape the cheese. The pressed cheeses are then subjected to a controlled maturation process before being released on the market. Approximately 150 L of milk yields around 15 cheeses, each weighing 1 kg.

The treatment of RW is essential to ensure compliance with local environmental regulations. There is a lack of regulations in Peru to monitor and control raw whey spills, particularly at the local level. As a result, many small and medium-sized dairies do not prioritize the treatment of raw whey before discharging it into sewers or the environment, such as rivers or soil. In this specific case, and in the absence of treatment, the manufacturer would have to dilute the effluent approximately 70-fold to meet Peruvian regulatory standards in terms of COD (chemical oxygen demand). Diluting RW to comply with environmental regulations poses challenges to achieving the Sustainable Development Goals (SDGs). While it may temporarily address compliance issues, dilution fails to promote sustainable water management, biodiversity conservation, and responsible consumption and production. Effective RW management aligns with the SDGs by prioritizing pollution prevention, resource recovery, and circular economy principles to ensure environmental sustainability and human well-being. In the present study, a sample of RW was collected directly from the wastewater pipeline and stored at −20 °C until needed. The physicochemical composition of the RW is shown in

Table 1. Physical–chemical characteristics of raw cheese whey (RW) from the Peruvian industry.

Parameters	Unit	RW
pH	−log [H+]	5.82
Temperature	°C	30.7
Electrolytic conductivity (EC)	μS cm−1	6823
True color	CU	4122
Optical density (OD) at 570 nm	AU	2.40
Turbidity	NTU	529.33
Total suspended solids (TSS)	mg L−1	4091
Biochemical oxygen demand (BOD5)	mg L−1	42,188
Chemical oxygen demand (COD)	mg L−1	69,531
Oil and grease (OG)	mg L−1	1472
Total Kjeldahl Nitrogen (TKN)	mg L−1	98.0
Nitrate (NO3−)	mg L−1	313
Nitrite (NO2−)	mg L−1	3.761
Ammoniacal nitrogen (N–NH3)	mg L−1	14.4
Phosphate (PO43−)	mg L−1	1186
Sulfate (SO42−)	mg L−1	344.8
Potassium (K+)	mg L−1	1599

. For the treatments described below, the raw cheese whey was filtered through a 0.5 mm sieve, followed by a dilution with distilled water, similar to that available by the manufacturer in its facilities, in varying proportions.

2.3. Flask Culture Experiments

Initially, the tolerance of duckweed and microalgae to RW was studied. Treatment cultures consisted of mixing RW with distilled water (DW) in different proportions, specifically 10%, 20%, 30%, 40%, and 50% (v/v) (duckweed and microalgae did not survive above 50% and 30%, respectively). The control culture medium consisted of BG11 medium for microalgae, while HSH medium was utilized for Lemna minor. Treatment cultures with RW were not supplemented with nutrients from BG11 and HSH. The cultivation conditions matched those described in Section 2.1.

In the case of duckweed, 400 mL circular glass containers with a working volume of 200 mL were utilized. The factor RW to DW ratio was investigated for three initial population surface densities of Lemna minor: 2.50, 3.75, and 5.00 g dm⁻². The effect on growth was assessed based on the final wet biomass yield. Briefly, at the end of each culture experiment, all duckweed plants from each treatment were harvested. The harvested plants were washed with distilled water to remove surface impurities and then subjected to forced ventilation with cold air to remove excess surface water. Subsequently, the biological material was weighed using a high-precision analytical balance (A&D GH-200, Tokyo, Japan). The biomass yield measurements were expressed in terms of wet or fresh weight (FW).

The total chlorophyll content (Chl_total), which comprises chlorophylls a and b, was determined following a modified protocol based on Ceschin et al. [19]. Briefly, wet duckweed fronds were prepared as described above for the whole plants. Subsequently, 1 g wet weight of this biological material was weighed using a high-precision analytical balance, followed by crushing in a mortar with 10 mL of 99.9% absolute anhydrous ethanol (denatured SDA-40B, Merck, Darmstadt, Germany). The resulting extract was filtered through a 0.45 μm PTFE-L membrane filter (Cobetter, Nantong, China) and transferred to a 15 mL Falcon™ tube, where it was left to stand for 48 h at 4 °C in the absence of light. The suspension was then centrifuged at 23,478 RCF-max for 10 min at 4 °C (Rotanta 460R, Hettich Zentrifugen, Tuttlingen, Germany). Subsequently, the absorbance of the supernatant was measured using a UV–visible spectrophotometer (UH-5300, Hitachi, Japan) at the wavelengths of 649 and 664 nm, as recommended by Lichtenthaler [20], with a 99.9% ethanol solution used as the blank. Chlorophyll concentrations were calculated following the equations proposed by Lichtenthaler [20] (see Section 2.6), and the results were expressed in mg of Chl_total per gram of wet weight of frond tissue (mg g⁻¹ FW).

The assays were conducted using duckweed plants acclimated to each culture medium (for at least one month with several passages of fresh medium) to ensure the adaptability of the cultivated species to the RW-based culture medium. For the microalgae, the cultures were carried out in 300 mL Erlenmeyer glass flasks with a 250 mL working volume, as described elsewhere [13], following the same environmental conditions as described in Section 2.1. The influence of different culture medium formulations on growth and biomass yield was studied. The assays were carried out using cells acclimated to each culture medium (at least two months with several medium passages) to ensure the long-term adaptability of the cultivated species to the wastewater-based culture medium. The media were inoculated using exponentially growing cells. The experiments started at a mean initial cell concentration of 1.0 × 10⁶ cells mL⁻¹. The initial pH for all cultures (Lemna minor and microalgae) was adjusted to 7.2 using a 2 M NaOH solution. The pH was allowed to evolve freely in all cultures. The experiments were carried out in triplicate.

2.4. Sequential Batch Photobioreactor Experimentation

The sequential batch culture system consisted of a photobioreactor (PBR-1) containing duckweed connected in series to a second photobioreactor (PBR-2) with a microalga, either Chlorella sp. MC18 or Scenedesmus sp. MJ23-R. The chosen RW to distilled water (DW) proportions (10, 20, and 30%, v/v) and initial population surface density of duckweed (5.00 g dm⁻²) from results provided by Section 2.3 were used in this experimentation. The hydraulic residence time in each PBR was 0.268 ± 0.039 h. At the start of the experiment, PBR-2 containing 3.8 L of RW medium was inoculated with 0.2 L of exponentially growing cells (5% v/v) of one of the two microalgae as reported earlier [13], where 5.00 g dm⁻² of duckweed biomass was added in PBR-1 containing culture medium based on RW to DW proportions (10%, 20%, and 30%, v/v). The experiment was run for 96 days, equivalent to 3 cycles, under the conditions described in Section 2.1, a sufficient number of cycles to ensure acclimation of duckweed and microalgae to the cultivation system. The bath culture times in PBR-1 and PBR-2 were 6 and 7 days, respectively, equivalent to hydraulic residence times of 0.280 ± 0.041 h and 0.236 ± 0.036 h, respectively. Upon completion of each experimental cycle in PBR-1, 3.8 L of treated cheese raw whey (TRW) were transferred to PBR-2 for further treatment, while an equal volume was removed from PBR-2. Only the last cycle was monitored. Thus, at the start and at the end of this last experimental cycle, the chlorophyll concentration in Lemna minor was measured in PBR-1, while cell counting of the culture was also measured in PBR-2 for the estimation of microalgal growth.

Both PBR-1 and PBR-2 were identical to those used in a previous study [13]. Inspired by innovative indoor duckweed farming techniques [11], our PBR system utilizes multi-tiered cultivation systems and low-cost LED strip lighting commonly used in horticulture. Commercially available stacking flow-through systems hold promise for large-scale microalgae cultivation, particularly for optimizing duckweed cultivation within horizontally stacked photobioreactors. Recent reports also highlight the integration of stacked horizontal planar waveguide modules for attached cultivation photobioreactors [21]. Briefly, PBRs were flat-panel structures, each comprising a 4 L parallelepiped-shaped glass container (17 cm wide, 35 cm long, 10 cm high). To ensure uniform light exposure and mitigate self-shading, the top-illuminated PBRs maintained a shallow culture depth of 8.5 cm, allowing for potential stacking in scale-up scenarios. Culture medium circulation relied on a submerged centrifugal pump positioned near one PBR wall. Air sparging, at a rate of 2 L min⁻¹ (0.125 vvmin) for duckweed and 8 L min⁻¹ (i.e., 0.5 vvmin) for microalgae, utilized compressed air filtered from a compressor. A 12 mm nozzle sparger near the pump outlet facilitated efficient air bubble dispersion, while regularly spaced baffles along the axial axis enhanced culture mixing. Employing a 12 h/12 h light/dark (L/D) cycle, the culture temperature was controlled at 25 ± 2 °C. PBR sterilization and cleaning procedures, detailed elsewhere [13], were executed before use. The initial pH was set at 7.2 and allowed to fluctuate naturally. COD, NO₃⁻, PO₄³⁻, and SO₄²⁻ in the supernatants were monitored regularly.

After attaining the stationary phase in PBR-2 within 7 days, the microalgal broths underwent centrifugation at 23,478 RCF-max for 10 min at 4 °C. The resulting pellets were delicately rinsed with distilled water. Subsequently, it underwent two additional pelleting cycles, with careful decantation of the supernatants each time. The biomass pellets were then dried in an oven at 100 °C for 12 h. The dried microalgal biomass samples were utilized for determining carbohydrates, lipids, proteins, and photosynthetic pigments as described below.

In order to assess the tolerance of the microalgae to the diluted RW effluent treated by duckweed in PBR-1, culture experiments were conducted using flasks similar to those described in Section 2.3. However, in this case, the effluent served as the primary culture medium. The rationale behind this experiment was to evaluate whether the effluent from the pretreatment of RW in PBR-1 (coded as TRW–10%, TRW–20%, and TRW–30%), based on duckweed-mediated bioremediation, improved the microalgal response compared to the untreated diluted effluent (RW) used in Section 2.3. The experiments were carried out in triplicate.

2.5. Growth Kinetic Data and Parameters

The monitoring of microalgal growth involved assessing cell concentration in the cultures (cells mL⁻¹). Cell counting was performed with a Neubauer chamber, using an improved method compared to the conventional approach [22]. In this method, all border cells were counted and then divided by two, whereas the conventional method only counts border cells on two boundaries (top and left). Sample preparation followed the supplier’s recommendations (https://mural.uv.es/basgaros/Cell-counting-Neubauer-chamber.pdf) (accessed on 24 August 2024). Briefly, the concentration range for cell counting with the Neubauer chamber must be between 250,000 cells mL⁻¹ and 2.5 10⁶ cells mL⁻¹. It is recommended that the dilution concentration be approximately 10⁶ cells mL⁻¹, with appropriate dilutions applied as needed. The Neubauer chamber was positioned on an optical microscope (Carl Zeiss, Primo Star, Göttingen, Germany) with a 40× objective lens. Sampling involved withdrawing 1.5 mL samples from each culture at specified time intervals, and measurements were conducted in triplicate. The biomass yield (Y_B), biomass productivity (P_B), and maximum specific growth rate (µ_max) of each microalga in each treatment were calculated as follows:

Y_B = X_max − X₀   (cells mL⁻¹ or mg L⁻¹)

(1)

P_B = (X_max − X₀)/Δt   (cells mL⁻¹ d⁻¹ or mg L⁻¹ d⁻¹)

(2)

µ_max = (ln X − ln X₀)/t   (day⁻¹)

(3)

In the above equations, X signifies the cell concentration at a specific given culture time (t), with X_max representing the maximum cell concentration and X₀ denoting the initial cell concentration throughout the cultivation period. The time interval (Δt) refers to the duration (in days) between the instances X_max and X₀. The µ_max value was derived from the exponential growth phase, which was identified in growth kinetics by the interval where a linear correlation exists between ln (X/X₀) and time. The coefficients of determination (r²) were consistently above 0.94 in all cases.

2.6. Other Analytical Measurements

The physicochemical characterization of RW and TRW, including analyses of color, total suspended solids (TSS), BOD₅, COD, oil and grease (OG), organic nitrogen (TKN), ammoniacal nitrogen (N–NH₃), and potassium (K⁺), were carried out in a laboratory accredited by the National Quality Institute, Lima, Peru (http://www.cerper.com) (accessed on 24 August 2024) and certified by international standards ISO 9001:2015 [23], ISO 14001:2015 [24], ISO 45001:2018 [25], and ISO 37001:2016 [26], which legitimizes the control of equipment, calibrations, validations, measurements, and authenticity of the analyses. These methods followed the standard methodology as prescribed in SMEWW-APHA-AWWA-WEF 2017 [27]. The method of each analysis is described in detail in a previous study [13]: color (Part 2120 C); TSS (Part 2540 D); BOD₅ (Part 5210 B); COD (Part 5220 D); OG (Part 5520 B); TKN (Part 4500-N_org B); N–NH₃ (Part 4500-NH₃ D); and K⁺ (Part 3111 B).

The quantification of inorganic anions—NO₃⁻, NO₂⁻, PO₄³⁻, and SO₄²⁻—was performed using ion chromatography (Sykam, S–150 IC Systems, Eresing, Germany). The detailed procedure for analyzing each analyte using ion exchange chromatography is described in a previous study [13]. Nutrient removal efficiencies (RE) were assessed as follows:

RE = (C_i0 − C_if)/C_i0 × 100  (%)

(4)

where C_i0 and C_if represent the initial and final nutrient (i) concentrations in the supernatants of the cultures.

The analysis of the biochemical biomass composition followed established protocols as previously described [13]. Briefly, total carbohydrate content was determined by the spectrophotometric method [28]. Total protein content was calculated by multiplying the Kjeldahl organic nitrogen value by 4.78 [29]. Lipid content was determined using EPA Method 1664—Review B [30]. The contents of total chlorophyll (Chl_total) and carotenoids (Car) (expressed in mg L⁻¹) were quantified using the following equations described elsewhere [20]:

Chl_total = 5.24 × OD₆₆₄ − 22.24 × OD₆₄₉

(5)

Car = (1000 × OD₄₇₀ − 2.13 × Chl_a − 97.64 × Chl_b)/209

(6)

where OD₆₆₄, OD₆₄₉, and OD₄₇₀ are optical densities at wavelengths of 649, 664, and 470 nm, respectively, corresponding to extracts from microalgal pellets obtained as indicated by Lichtenthaler [20].

2.7. Statistical Analysis

Each experiment setup was performed in triplicate with one measurement in each (n = 3) unless otherwise specified. One-way and two-way analysis of variance (ANOVA) and multivariate analysis of variance (MANOVA) tests were used for significant difference analysis. Statistically significant differences in the mean response between factors were fixed at a 5.0% significance level threshold (p < 0.05). Post hoc comparisons were performed using Tukey’s honestly significant difference (HSD) test at p < 0.05. Statistical data analyses were performed using the IBM SPSS Statistics 29.0.2.0 software (SPSS Inc., Chicago, IL, USA).

3. Results and Discussion

3.1. Tolerance of Duckweed to Raw Whey (RW) Wastewater

A two-way ANOVA was performed to assess the influence of the proportion of RW in the culture medium (RW%) and surface density of the wet biomass of the duckweed inoculum (DEN) on Chl_total. The results indicated that both RW% and DEN had a statistically significant effect on Chl_total (F (5, 36) = 1191, p < 0.001; F (2, 36) = 421, p < 0.001, respectively). Additionally, there was a significant interaction between RW% and DEN (F (10, 36) = 42, p < 0.001), suggesting that the combined effects of these factors were not independent. The overall variability in Chl_total was partitioned into components attributed to the main effect of each factor and their interaction. The contribution of each factor and interaction to the variation in Chl_total was estimated as the percentage of their respective F-ratios relative to the sum of all F-ratios. Consequently, the contribution of the RW%-DEN interaction (2.5%) to the global variation in Chl_total was substantially lower than that of RW% (72%) and DEN (25%).

Figure 2 presents the outcomes of the two-way ANOVA. As depicted in Figure 2A, the presence of RW resulted in a reduction in the mean Chl_total of duckweed compared to the control group (without RW). Conversely, Figure 2B illustrates that the mean Chl_total exhibited an upward trend with a higher DEN (Figure 2B). Duckweed demonstrated limited tolerance to the culture media containing the highest RW proportions (RW–40% and RW–50%), as evidenced by unhealthy visual appearances characterized by chlorophyll bleaching, which is indicative of stress. Consequently, these RW concentrations were excluded from the sequential culture tests in the photobioreactor (PBR) (see Section 3.3). Additionally, Figure 2C highlights that the effect of the RW proportion on Chl_total differed between the DEN levels of 5.00 and 2.50 or 3.75 g dm⁻², respectively. The decline in Chl_total at 5.00 g dm⁻² DEN from the control group (2.15 mg g⁻¹) to RW–40% (0.71 mg g⁻¹) was less pronounced compared to 2.50 and 3.75 g dm⁻².

Given the high concentration of organic compounds present in RW (see COD and BOD₅ in Table 1), the results preliminarily suggest that duckweed had to be nourished mixotrophically to survive in water supplemented with RW. This is in line with other recent studies [31,32], in which mixotrophic growth of duckweed species was observed to lead to a decrease in chlorophyll content compared to photoautotrophic growth. Mixotrophic duckweed cultures grew faster than phototrophic cultures but exhibited reduced chlorophyll levels. The addition of sucrose or glucose in mixotrophic conditions influenced the chlorophyll content of Lemna species. Findings from Klamann et al. [31] suggested that the presence of an additional carbon source in mixotrophic conditions may impact the chlorophyll levels in Lemna species compared to their growth solely under photoautotrophic conditions.

While the main compound of RW is lactose, duckweed cannot assimilate lactose directly [33,34]; however, it can utilize lactic acid as a carbon source [35]. Cheese RW often contains lactic acid bacteria (LAB) that convert lactose into lactic acid [36,37]. The pasteurization process typically involves heating whey to a temperature around 70 °C for a specific duration to reduce the number of harmful bacteria, including LAB. Although pasteurization effectively reduces the bacterial load, it does not necessarily eliminate all microorganisms, including LAB [38]. Some LAB may survive pasteurization, but their numbers should be significantly reduced, making them less likely to cause spoilage or other issues. The presence of live LAB after pasteurization depends on various factors such as the initial bacterial load, the effectiveness of the pasteurization process, and post-pasteurization handling and storage conditions. Although pasteurization is effective at reducing the microbial load, including LAB, to safe levels, there may be certain strains that exhibit varying degrees of resistance to heat. Factors such as specific strains of LAB, its genetic makeup, and environmental conditions can influence its heat resistance. In fact, the high value of BOD₅ measured for RW (see Table 1) suggests the presence of LAB (a sterile organic wastewater typically has a low initial BOD₅ value, often close to zero, because it lacks active microbial populations that would contribute to organic matter decomposition). Hence, it is plausible that heat-resistant LAB continues to convert lactose into lactic acid during duckweed cultivation, potentially allowing the duckweed to assimilate this lactic acid. This hypothesis is consistent with the observed drop in culture pH below 5 within the initial three days of culture in both flask and PBR-1 cultures (Supplementary Material), suggesting LAB’s acidifying activity through lactic acid excretion into the medium [37]. Given these considerations, it is plausible that the conversion of lactose to lactic acid observed in this study could be partially attributed to the residual LAB present in the pasteurized RW. However, this remains a hypothesis since the presence and activity of LAB—or other microorganisms capable of metabolizing lactose—have not been definitively confirmed. Further studies, including genetic sequencing, would be necessary to validate this mechanism and to explore the potential involvement of other microbial species in the process.

Extracellular and intracellular proteolytic and lipolytic activities of LAB are extensively documented [37,39]. Proteolytic enzymes can break down high-molecular-weight and complex proteins into peptides and amino acids, along with generating aromatic compounds such as aldehydes, alcohols, ketones, esters, or among others. Given that triacylglycerides constitute a significant portion of the lipids present in whey, lipolytic enzymes play a crucial role in hydrolyzing triglycerides to produce free fatty acids, glycerol, and intermediate products such as mono- and diglycerides. These metabolites, whether released into the medium by LAB or formed through extracellular proteolytic and lipolytic transformations, hold potential for assimilation by duckweed or microalgae. Furthermore, duckweed and microalgae cultures in RW-based media could leverage the antibacterial properties of lactic acid and bacteriocin peptides, which are released into the culture medium as by-products of LAB metabolism [37,39]. This mechanism has the potential to inhibit the proliferation of other bacterial contaminants that may be present in the cultures.

The decline in Chl_total at 5.00 g dm⁻² DEN from the control group (2.15 mg g⁻¹) to RW–40% (0.71 mg g⁻¹) was more pronounced compared to 3.75 g dm⁻² and similar to 2.50 g dm⁻² (Figure 2C). This effect could be attributed to self-shading among duckweed plants, high turbidity (529.33 NTU; see Table 1), or the color of the RW medium (4122 CU; see Table 1), all of which hinder efficient photosynthesis. Walsh et al. [40] found that variations in the photosynthetic quantum yield (expressed in F_v/F_m, Y(II), Y(NPQ), and Y(NO)) of Lemna minor grown at high surface density cultures in synthetic dairy wastewater occurred because the plants used light energy less efficiently compared to those grown at lower surface density. Duckweed surface density may affect aspects of plant metabolism (e.g., nutrient assimilation or metabolism) rather than having a direct effect on photosystem II activity (i.e., F_v/F_m).

3.2. Tolerance of the Microalgae to Raw Whey (RW) Wastewater

The tolerance of both Chlorella sp. and Scenedesmus sp. to RW diluted with distilled water in different proportions (v/v) was explored in batch flask cultures (Section 2.3), coded as follows: RW–10%, RW–20%, RW–30%, RW–40%, and RW–50%. Figure 3 shows semilog plots illustrating the normalized cell concentration for the surviving microalgae treatments (RW–10%, 20%, and 30%): Figure 3A for Chlorella sp. and Figure 3B for Scenedesmus sp. In addition, results from the effluents generated during the pretreatment of the different RW (coded as TRW–10%, TRW–20%, and TRW–30% in Section 2.4) in PBR-1 for the sequential batch photobioreactor experimentation, which were also tested for tolerance in flask cultures, are included in Section 3.3.2 for later comparison. No apparent lag phase was observed for any of them, regardless of the cultured species. The onset of the exponential growth phase occurred at the beginning of cultivation and extended until around the third day for both microalgae. Throughout this phase, the maximum specific growth rates were determined through regression analysis using Equation (3). Except for the BG11 control, the RW-based cultures did not have a linear phase, and therefore their exponential phase was followed by a stationary phase.

Based on the growth kinetics depicted in Figure 3A,B, µ_max, Y_B, and P_B were calculated as detailed in Section 2.5. A two-way ANOVA was performed to assess the influence of RW% and microalgal species (ALGA) on each of the three response variables. The results revealed that both RW% and ALGA had a statistically significant effect (p < 0.001) on: (i) µ_max (F (3, 16) = 30.4), F (1, 16) = 28.3, respectively); (ii) Y_B (F (3, 16) = 101.5), F (1, 16) = 28.1, respectively); and (iii) P_B (F (3, 16) = 26.9), F (1, 16) = 24.2, respectively). Moreover, a significant interaction (p < 0.001) between RW% and ALGA was observed (µ_max, F (3, 16) = 9.5; Y_B, F (3, 16) = 7.6; P_B, F (3, 16) = 8.75), indicating that the combined effects of these factors were not independent. According to the F values, the percentage of the variance was distributed from highest to lowest in the following order: RW%, ALGA, and interaction.

Figure 3D–L presents the outcomes of the two-way ANOVA. Only RW–10% significantly improved the mean µ_max (0.835 d⁻¹) compared to the control group (BG11 medium) (Figure 3D). Although no statistically significant differences were observed between the RW–20% and RW–30% groups relative to the control group, their values were lower than RW–10%, indicating an inhibitory effect due to excess substrate; the higher the RW%, the lower the µ_max. The group of Chlorella sp. MC18 (CH) cultures exhibited a mean µ_max value, 0.760 d⁻¹, 1.12-fold higher than that of Scenedesmus sp. MJ23-R (SC) (Figure 3E). Figure 3F highlights that the effect of the RW% on µ_max differed within the group ALGAE. The increase in Chlorella µ_max at RW–10% relative to the control was more pronounced compared to Scenedesmus. The RW% and ALGAE factors affected in a similar way to P_B (Figure 3G–I). Conversely, the mean Y_B exhibited a downward trend with a higher RW% relative to the control group (Figure 3J), suggesting potential nutrient imbalance, toxicity, stress responses, or metabolic burden. While these hypotheses warrant further experimental validation, they offer plausible explanations for the observed trend. Future research efforts could focus on elucidating these mechanisms to enhance our understanding of microalgal growth dynamics in RW-based cultures.

The mean Y_B of the Chlorella culture group was higher than that of Scenedesmus as observed for µ_max and P_B (Figure 3K). However, the interaction between RW and ALGA was manifested for the RW–20% and RW–30% groups, for which Y_B was higher in Chlorella than in Scenedesmus (Figure 3L). Results demonstrated that Chlorella and Scenedesmus thrive in culture media containing moderate RW proportions (≤RW–30%). Consequently, these RW concentrations were included in the sequential culture tests in the photobioreactor (PBR) (see Section 3.3). Results from Figure 3 seem to indicate that Chlorella sp. possesses a slight competitive advantage compared to Scenedesmus sp.

In essence, these results underscore the effectiveness of using 10%, 20%, and 30% RW media in providing sufficient nutrients for the growth and biomass production of both microalgae compared to the control. This approach offers at least two key advantages: (i) significantly reduces the need for extensive dilution of raw RW to comply with environmental regulations, thereby minimizing the water footprint associated with microalgae-based RW treatment; (ii) eliminates the requirement for additional chemically defined nutrients, thereby reducing the environmental footprint associated with their manufacturing.

However, it is important to note that microalgae must assimilate organic material present in RW. While some microalgae species possess the metabolic machinery to utilize lactose as a carbon source, others may lack efficient mechanisms for metabolizing this sugar compound, particularly those considered obligate phototrophs relying solely on photosynthesis for energy production and growth [41,42,43]. For example, studies have demonstrated that species such as Chlorella vulgaris and Chlorella protothecoides did not show growth when cultured in lactose-supplemented medium, indicating their limited ability to metabolize lactose under mixotrophic conditions, while Scenedesmus obliquus does [44]. In contrast, experiments conducted in this study with Chlorella sp. MC18 and Scenedesmus sp. MJ23-R strains in raw cheese whey have revealed their ability to thrive in lactose-rich environments. This suggests that these native strains possess the necessary metabolic machinery to utilize lactose as a carbon source. Furthermore, previous research has shown that certain microalgae, including Dunaliella tertiolecta, Chlorella minutissima, and Nannochloropsis oculate [42], can metabolize lactose under mixotrophic conditions, underscoring inter- and intra-species variability in lactose utilization among microalgae. Effective lactose metabolism is associated with key traits such as the expression of β-galactosidase (an enzyme that hydrolyzes lactose into its constituent monosaccharides, glucose and galactose, simple sugars assimilable by microalgae), metabolic flexibility for switching between different carbon sources, and efficient sugar uptake mechanisms. Genetic and metabolic diversity within the Chlorella and Scenedesmus genera is likely instrumental in enabling them to metabolize lactose and other carbon sources, as evidenced by their performance in cheese-whey-based cultures.

Building upon the premise established in duckweed cultures, it is plausible that the introduction of LAB into the medium used for microalgae cultures resulted in the formation of lactose-derived lactic acid. Notably, previous studies have highlighted the metabolic capacity of Chlorella and Scenedesmus species to assimilate lactic acid [45,46]. Therefore, it is reasonable to infer that a comparable phenomenon of lactic acid assimilation could have taken place in the microalgae cultures, thus providing additional support for the hypothesis. Despite the non-axenic nature of the cultures, microscopic analysis revealed a notable absence of thriving contaminating microorganisms such as bacteria, fungi, or yeasts.

The genera Chlorella and Scenedesmus have also proven effective in recovering nutrients from various types of wastewaters rich in compounds containing essential elements such as nitrogen, phosphorus, or sulfur [12]. This is the case of RW, which has concentrations of nitrate, phosphate, and sulfate of 0.3 times, 54.4 times, and 11.7 times those contained in the BG11 medium (see Table 1).

Overall, both duckweed and microalgae, as demonstrated in flask culture experiments, exhibit the ability to remediate contaminants present in cheese whey wastewater when diluted up to 30%, eliminating the need for additional nutrients found in the BG11 medium (control). Consequently, both biological systems can effectively treat the effluent independently. The selection between microalgae and duckweed biomass depends on the specific application and economic viability for the intended purpose. Each exhibits unique characteristics that render them valuable biomass sources in distinct contexts [47]. Microalgae, rich in protein, essential fatty acids, vitamins, and antioxidants, are valuable for food, feed, and nutraceutical applications. Similarly, duckweed, also rich in protein, serves as a rapid plant source for animal feed, rapidly doubling its biomass. Cultivation costs vary depending on the system used, with economic assessments showcasing diverse applications for both microalgae and duckweed, including biofuels, animal feed, supplements, bioplastics, and wastewater treatment. The ultimate biomass value depends on market demand.

3.3. Sequential Batch Photobioreactor Experimentation with Duckweed and Microalgae

According to the findings from the preceding sections, both duckweed and Chlorella and Scenedesmus have the capacity to remediate diluted RW at proportions below 40%. However, if a microalgae compound-targeted approach is sought for RW bioremediation, it is essential to consider that the biomass yield of microalgae in the flask cultures decreased with higher proportions of raw whey (RW%) in the culture medium. Employing a sequential batch photobioreactor approach, where RW% is pretreated with duckweed before its use by microalgae, may offer a potential solution to mitigate this inhibitory effect and enhance microalgal biomass production. To assess long-term growth reproducibility in RW-based medium during scale-up from flasks to PBRs, the described photobioreactor from Section 2.4 was used, labeled as PBR-1 for duckweed and PBR-2 for microalgae.

3.3.1. Bioremediation Potential of Duckweed in RW-Based Media

PBR-1 pretreated the culture media RW–10%, RW–20%, and RW–30% at the three initial duckweed biomass surface densities (DEN) of 2.50, 3.75, and 5.00 g dm⁻². To facilitate comparison, the percentage production of duckweed biomass (Y_PBR-1) was calculated relative to DEN in each PBR-1. The results of a two-way ANOVA for the factors RW% and DEN are depicted in Figure 4. Both RW% and DEN showed statistically significant effects (p < 0.001) on Y_PBR-1 (F (2, 18) = 1801, F (2, 18) = 269, respectively). Although the contribution of the RW%-DEN interaction to the variance was also significant (p < 0.001; F (4, 18) = 60), it was much lower than that of the main factors. In PBR-1, biomass increased when RW–10% was used but declined with higher RW% (Figure 4A). In contrast, Y_PBR-1 increased with DEN (Figure 4B). Notably, the decline in Y_PBR-1 at 5.00 g dm⁻² DEN from RW–10% to RW–30% was less pronounced compared to 2.50 and 3.75 g dm⁻², respectively (Figure 4C). The above trends were consistent with those observed for Chl_total in flask cultures (Figure 2), suggesting that Chl_total served as a proxy for Y_PBR-1. Additionally, as shown in Figure 4C, the initial surface density of 5.00 g dm⁻² yielded the highest Y_PBR-1 values across all RW% conditions, with the maximum value (22.9% ± 5.5%) attained at RW–10%, corresponding to a productivity of 0.194 ± 0.053 g dm⁻² d⁻¹. Therefore, the effluent from PBR-1 was used as influent in PBR-2 for further treatment with microalgae at this surface density.

The bioremediation effectiveness of duckweed at RW–10%, RW–20%, and RW–30% in PBR-1 is depicted in Figure 5 for an initial inoculum surface density of 5.00 g dm⁻². Observably, COD, nitrates, phosphate, and sulfate concentrations decreased in all three culture media, with the most significant reduction observed for nitrates, which were completely depleted in RW–10% and RW–20%. A one-way MANOVA test was conducted to assess the impact of RW% on removal efficiencies (RE%) of the four pollutants (Figure 5E). Multivariate tests revealed a highly significant effect of RW% on the combined RE% values (Wilk’s Lambda = 0.001, F (8, 6) = 25.9, p < 0.001), indicating an overall influence. Univariate tests further demonstrated significant effects of RW% on RE%-COD (F (2, 6) = 28.4, p < 0.001), RE%-nitrate (F (2, 6) = 31.5, p < 0.001), RE%-phosphate (F (2, 6) = 29.3, p < 0.001), and RE%-sulfate (F (2, 6) = 7.9, p < 0.021). These findings underscore the influential role of RW% in shaping duckweed performance in RE%.

According to Figure 5A, COD removal was notably high and unexpected, probably due to several factors. First, during the treatments with duckweed at each RW concentration (10%, 20%, or 30%, v/v), there was significant removal of TSS, particularly a cream or fine white particles present in each %RW (see Supplementary Material). This cream adhered to the roots and internal surfaces of each macrophyte seedling. Whey cream (WC), a component of cheese whey, consists mainly of total solids, lactose, protein, fat, phospholipids, and total nitrogen [48,49], all of which contribute to the organic matter in %RW media. A recent study by [50] suggests that the roots of duckweed act as a filter, trapping suspended particles in the water, which reduces high levels of TSS and COD when Lemna minor is grown in dairy effluent. From a resource recovery perspective, traditionally expensive methods (e.g., centrifugal force) are needed to separate whey cream from cheese whey [49]. However, cultivating Lemna minor in PBR-1 not only remedies RW but also allows the recovery of WC and macrophyte biomass, both of which are high-value bioproducts worth exploring in the future.

Additionally, the consumption and removal efficiency of COD, nitrate, phosphate, and sulfate (Figure 5A–D) can be attributed to the duckweed metabolism, which assimilates these nutrients during growth. Regarding nitrogen removal, it is important to note that the initial high concentration of nitrates observed in the RW (see Table 1) is not due to nitrification of ammonium nitrogen but rather reflects the native composition of the RW used in this study. Nitrates can be present in dairy products from both endogenous and exogenous sources [51]. In traditional wastewater treatment processes, nitrate removal can be challenging, often requiring specific conditions for denitrification. However, the reduction in the nitrate concentration observed in Figure 5 is primarily due to the high uptake by Lemna minor, which efficiently assimilates nitrates and other nutrients, incorporating them into its growth. Obviously, the organic nitrogen is assimilated alongside the organic carbon sources that contain it. Hemalatha and Mohan [52] observed a high uptake of organic carbon, nitrates, and phosphates by duckweed in dairy wastewater, with simultaneous biomass growth. Duckweed absorbs nitrogen through its leaves and roots, incorporating it into its biomass, and stores absorbed phosphorus endogenously, as it is essential for energy metabolism and nucleic acids [53].

Over the six-day pretreatment period, duckweed significantly reduced the concentration of pollutants in the culture medium. Interestingly, although the highest pollutant concentrations were observed in the medium with the greatest proportion of RW (RW–30%), it exhibited the best removal efficiencies (Figure 5E): COD 60.69%, nitrate 90.93%, phosphate 40.52%, and sulfate 22.58%. It can be seen in Figure 5 that the efficiency of pollutant removal could have been increased with more days of duckweed cultivation, but it is unlikely that complete bioremediation could have been achieved as nitrate was the limiting nutrient, as it was either exhausted in the RW–10% and RW–20% cultures, or its concentration was very low in the RW–30% culture.

3.3.2. Tolerance Test of the Microalgae to RW-Based Media Pre-Treated with Duckweed

The aim of this duckweed pretreatment was not to completely bioremediate the pollutants but to reduce their concentration in each RW% to less inhibitory levels for the microalgae. To test this hypothesis, the supernatants of each pretreatment in PBR-1, coded as TRW–10%, TRW–20%, and TRW–30% (see Section 2.4), were used to cultivate both microalgae in flasks (Figure 6A,B), following the same procedure as described in Section 2.3 with RW%. The results of the two-way ANOVA conducted to examine the effects of factor MEDIUM and factor ALGA on the dependent variables µ_max, P_B, and Y_B are presented in Figure 6. The analysis revealed a significant main effect of MEDIUM (F (6, 28) = 92.7, p < 0.001) and ALGA (F (1, 28) = 46.2, p < 0.001) on the three response variables. Additionally, there was a significant, but weak, interaction effect between both factors (F (6, 28) = 5.7, p < 0.001). Post hoc comparisons using Tukey’s HSD test for MEDIUM showed that µ_max, P_B, and Y_B for media pre-treated with duckweed (TRW%) were significantly higher (p < 0.05) than their counterparts (RW%) (Figure 6D,G,J). For factor ALGA, µ_max, P_B, and Y_B were also significantly higher for Chlorella than Scenedesmus (p < 0.01) (Figure 6E,H,K) as observed for the RW% in Section 3.2. Further examination of the interaction effect (Figure 6F,I,L) revealed that the relationship between both factors and µ_max, P_B, and Y_B varied weakly depending on the level of the factor ALGA. Specifically, µ_max and P_B (Figure 6F,I) at Chlorella increased more rapidly than Scenedesmus as MEDIUM changed from BG11 to RW–10%; above RW–10%, the trends of the three variables for both microalgae were parallel. These findings suggest that both MEDIUM and ALGA independently influence µ_max, P_B, and Y_B, while their interaction hardly further modulates the effect on them.

The above outcomes anticipated promising results regarding the tolerance of microalgae to raw whey (RW)-based media pre-treated with duckweed. By assessing the growth parameters of microalgae in both treated (TRW%) and untreated (RW%) media, significant improvements in growth performance were observed in TRW% media. Furthermore, the selection of microalgal species significantly influenced their tolerance to RW-based media, with Chlorella demonstrating superior performance compared to Scenedesmus. These findings underscore the potential of duckweed pretreatment to improve RW bioremediation and highlight the importance of microalgal species selection in optimizing remediation strategies.

This investigation of the tolerance of microalgae to raw whey (RW)-based media pretreated with duckweed anticipated promising results. Assessment of growth parameters in both treated (TRW%) and untreated (RW%) media reveals significant enhancements in growth performance within the TRW% media, suggesting that duckweed pretreatment effectively reduces the inhibitory effects of RW on microalgae growth, leading to improved bioremediation potential. Moreover, the choice of microalgal species notably influences their tolerance to RW-based media, with Chlorella exhibiting superior performance when compared to Scenedesmus cultured in flasks.

3.3.3. Microalgae Bioremediation in Duckweed-Treated RW Media

The three supernatants obtained from PBR-1 (TRW 10% to 30%) served as culture media in the PBR-2; consequently, the initial concentrations of each of the pollutants in PBR-2 are the same as the final concentrations in PBR-1. Figure 7 shows the temporal changes in the concentrations of (A) COD, (B) nitrates, (C) phosphate, and (D) sulfate in the supernatants for both Chlorella and Scenedesmus cultures. At first sight, both microalgae significantly reduced the concentration of the four pollutants in each of the three culture media over seven days of cultivation in PBR-2. The reduction rate was highest for pollutants initially present at higher concentrations in the TRW% medium. Nitrates were completely depleted in all TRW% for both microalgae, whereas phosphates and sulfates were only depleted in TRW–10% with Scenedesmus.

As the consumption kinetics of Figure 7 suggested that both MEDIUM and ALGA factors could influence the pollutant removal efficiency (pollutant-RE%), a multivariate analysis of variance (MANOVA) was carried out, and results are represented in Figure S4 (Supplementary Material). It revealed a significant main effect of the MEDIUM factor on the combined pollutant-RE%, Wilk’s Lambda = 0.018, F (8, 18) = 14.74, p < 0.001 (Figure S4A,D,G,J). Subsequent univariate tests showed that MEDIUM significantly influenced three pollutant-RE%: COD-RE% (F = 136.5, p < 0.001), NO₃⁻-RE% (F = 17.7, p < 0.001), and PO₄³⁻-RE% (F = 5.4, p < 0.021). This observation confirms the importance of adjusting TRW% to effectively modulate pollutant removal rates, offering a means to enhance bioremediation outcomes by tailoring medium composition to specific environmental conditions.

Similarly, there was a significant main effect of the ALGA factor on the combined pollutant-RE%, Wilk’s Lambda = 0.015, F (4, 9) = 147.15, p < 0.001 (Figure S4B,E,H,K), particularly affecting COD-RE% (F = 251.3, p < 0.001), PO₄³⁻-RE% (F = 15.5, p < 0.002), and SO₄²⁻-RE% (F = 136.0, p < 0.001), but not in NO₃⁻-RE%. Chlorella or Scenedesmus exhibit varying efficiencies in pollutant removal, emphasizing the need for careful consideration when choosing species for bioremediation applications. By selecting microalgae with optimal pollutant removal capabilities, researchers can maximize remediation efficiency and improve overall treatment outcomes.

Interestingly, while NO₃⁻-RE% appears unaffected by variations in MEDIUM (i.e., TRW%) and ALGA (Figure S4), other pollutants exhibit differential responses, suggesting complex interactions between culture medium composition and microalgal species. For instance, the decreasing trend of COD-RE% with increasing RW% concentration highlights a potential inhibitory effect of higher RW% levels on COD removal, being more pronounced for Scenedesmus. The trend of PO₄³⁻-RE% for both microalgae was opposite up to TRW–20%; above this medium level, it remained statistically similar in both microalgae. Differences in SO₄²⁻-RE% between both microalgae were observed for medium levels below TRW–30%, with SO₄²⁻-RE% being higher for Scenedesmus. The above suggests that pollutant removal efficiency is influenced by nuanced interactions between TRW% concentration and microalgal species, as evidenced by MANOVA (Wilk’s Lambda = 0.049, F (8, 20) = 7.9, p < 0.001) (Figure S4C,F,I,L). This emphasizes the significance of simultaneously considering both factors in the design of bioremediation systems for raw whey. By comprehending the interaction between culture medium composition and microalgal species selection, researchers can devise more precise and efficient bioremediation strategies tailored to the specific environmental remediation requirements of raw whey.

3.3.4. Effect on Biomass Production and the Proximate Chemical Composition of Microalgae

Figure 8 shows the results from a two-way multivariate ANOVA (MANOVA) of main effects MEDIUM (i.e., treated raw whey, TRW%) and ALGA, and their interaction (MEDIUM–ALGA) on growth kinetic parameters and concentrations of the elemental biochemical compounds in the PBR-2 cultures after a 7-day cultivation period, coinciding with the entry of the cultures into the stationary phase.

Regarding growth, the analysis revealed a significant main effect of the MEDIUM factor on the combined µ_max, P_B, and Y_B, Wilk’s Lambda = 0.000, F (14, 12) = 1102.7, p < 0.001 (Figure 8A,D,G). Subsequent multivariate tests showed that MEDIUM significantly influenced the three growth kinetic parameters: µ_max (F = 88.2, p < 0.001), P_B (F = 726.1, p < 0.001), and Y_B (F = 726.1, p < 0.001). The mean maximum values were obtained at RW–20%. This observation confirms the importance of adjusting the RW% level to effectively modulate microalgal growth. Similarly, there was a significant main effect of the ALGA factor on the combined µ_max, P_B, and Y_B, Wilk’s Lambda = 0.001, F (7, 6) = 1087.7, p < 0.001 (Figure 8B,E,H), affecting the three parameters: µ_max (F = 41.3, p < 0.001), P_B (F = 364.6, p < 0.001), and Y_B (F = 364.6, p < 0.001). There was an effect of the interaction of factors (Figure 8C,F,I).

As evident from Figure 8A–I, both strains demonstrated the ability to produce biomass from RW (10% to 30%) treated with duckweed on the PBR scale. Chlorella sp. consistently outperformed Scenedesmus sp. with 4.79%, 20.95%, and 20.95% higher µ_max, Y_B, and P_B (p < 0.05), respectively. This suggests that Chlorella sp. may be more efficient in converting the available nutrients in TRW% into biomass. Similarly, the higher μ_max of Chlorella sp. indicates its superior ability to thrive in the TRW% medium. This faster growth rate can be beneficial in the context of bioremediation, where the aim is to rapidly remove pollutants.

As far as we know, there are no studies reporting the sequential cultivation of duckweed and microalgae to take advantage of the nutrients contained in raw dairy wastewater.

Table 2. Dairy wastewater conditions and biomass productivity comparison. Maximum removal efficiency (%) and biomass productivity values (mg L⁻¹ d⁻¹) of Chlorella sp. MC18 and Scenedesmus sp. MJ23-R obtained in this study and results reported in the literature.

Strain	Wastewater Classification	Treatment	Concentration	Removal Efficiency (RE, %)			Biomass Productivity	Reference
				COD	NO3−	PO43−	(mg L−1 d−1)
Chlorella sp.	Dairy manure	AED and filtration	20% (v/v)	34.3	82.5 II	78.3	~81.4	[54]
Neochloris oleoabundans	Dairy manure	AED and sterilization	2% (v/v)	na	90−95	na	88.3 ± 7.9	[55]
Clorococo sp.	Dairy effluent	Sterilization	100% (v/v)	93	na	na	53.3	[56]
Chlorella sp.	Dairy farm WW	Filtration and centrifugation	10% (v/v)	87.85	93.01 III	90.94 V	80	[57]
Chlorella sp.	Dairy farm WW	Filtration and centrifugation	20% (v/v)	89.7	83.20 III	91.97 V	95	[57]
Scenedesmus sp.	Cheese whey	No treatment specified	6.25% (w/w) a	64.9 I	na	na	~10.58	[58]
Chlorella sorokiniana	Cheese whey	AED and centrifugation	10% (v/v)	41%	84 III	71 VI	60 ± 10	[59]
Chlorella vulgaris	Dairy effluent	Sterilization 1	20% (v/v)	Minimum RE	na	na	~16.25	[60]
Chlorella sp.	Dairy effluent	Sterilization 1	50% (v/v)	Partial RE	na	na	~85.63	[60]
Chlorella pyrenoidosa	Dairy WW	No treatment specified	75% (v/v) b	87.5	88.91	79.02	28.44 ± 8.02	[61]
Scenedesmus abundans	Dairy WW	No treatment specified	75% (v/v) b	62.5	84.72	86.51	18.72 ± 2.06	[61]
Scenedesmus quadricauda	Dairy WW	Treatment systems 2	100% (v/v)	69.1 ± 4.6 I	86.7 ± 2.3 IV	71.2 ± 8.4	32.5	[62]
Tetraselmis suecica	Dairy WW	Treatment systems 2	100% (v/v)	40.2 ± 1.9 I	66.8 ± 5.1 IV	42.2 ± 0.8	45	[62]
Ascochloris sp.	Raw dairy WW	Filtration 3	100% (v/v)	95.1	79.7	98.1	102 ± 3	[63]
Tetradesmus obliquus	Cheese whey permeate	Filtration and sterilization	100% (w/w)	na	16.7	0.32	94	[64]
Scenedesmus acuminatus	Milk whey processing WW	Aerobic digestion 4	100% (v/v)	93	88 IV	90 V	56	[65]
Oscillatoria sp.	Cheese whey water	Sterilization	25% (v/v) c	na	na	na	32	[66]
Chlorella vulgaris	Dairy WW	Sedimentation and filtration	25% (v/v)	na	29.27 ± 0.01	30.64 ± 0.01	~1.72 × 105 i	[67]
Chlorella vulgaris	Dairy effluent	Sedimentation and filtration	25% (v/v)	na	56.62 ± 0.01	51.84 ± 0.01	~3.15 × 105 i	[67]
Chlorella sp. MC18	Raw cheese whey	Pretreatment with duckweed	20% (v/v)	76.58 ± 1.35	98.36 ± 0.64	84.48 ± 4.60	91.10 ± 1.50	This study
Scenedesmus sp. MJ23-R	Raw cheese whey	Pretreatment with duckweed	20% (v/v)	47.83 ± 3.10	99.39 ± 0.35	87.89 ± 4.19	108.67 ± 1.57	This study

Wastewater (WW). Anaerobic digestion (AED). ¹ WW from the stabilization pond; ² WW from the treatment plant; ³ WW from the storage tank; ⁴ aerobic biological treatments based in a membrane bioreactor. ^a 40 g lactose from whey diluted to 2.5 g lactose; ^b Three parts of DWW and one part of BG11 medium; ^c whey water to BG11 media in the ratios of 1:3 (v/v). ^I TOC (total organic carbon); ^II (total Kjeldahl nitrogen); ^III NH₄-N (ammonia); ^IV TN (total nitrogen); ^V TP (total phosphorus); ^VI (soluble phosphorous). na (data not available). ⁱ Unit (cells mL⁻¹ d⁻¹).

summarizes results reported in previous literature on other treatments of dairy wastewater using microalgae. The results of the study presented here show higher biomass productivity. Despite the variations in conditions and treatment types between the studies, the comparison in Table 2 consistently showed the superior ability of Chlorella and Scenedesmus to thrive in culture mediums based on dairy wastewater.

The findings suggest that microalgae benefit from duckweed-treated raw whey in several ways. Duckweed acts as a natural filter, removing excess nutrients and creating a more suitable medium for microalgae growth. This balanced nutritional profile supports healthier conditions, leading to improved growth rates and biomass yield. Additionally, the use of treated raw whey reduces the need for costly synthetic nutrient supplements, making the cultivation process more cost-effective. Moreover, it promotes environmental sustainability by recycling organic waste and reducing the risks of water pollution.

MANOVA revealed a significant and diverse main effect of the factors (MEDIUM and ALGA) and their interaction on the combined biomass chemical compositions (see Figure 8). The observed patterns in the biochemical composition of both microalgae in response to different concentrations of TRW% in the culture medium can be interpreted in the context of nutrient availability and its impact on metabolic processes and growth parameters. A first pattern, characterized by higher cellular contents of carbohydrates and lipids in TRW–20% compared to TRW–10% (Figure 8J,M), suggests that microalgae allocate more resources towards energy storage and structural components when exposed to moderate percentages of TRW. This could be attributed to an increased availability of carbon and other essential nutrients, promoting the synthesis and accumulation of carbohydrates and lipids as energy reserves. Conversely, the lower protein content in TRW–20% (Figure 8O) indicates a potential trade-off between energy storage and protein synthesis, with microalgae prioritizing energy storage over protein production under these conditions. The consistent carotenoid content across TRW–10% and TRW–20% (Figure 8U) suggests that carotenoid biosynthesis may be significantly influenced by changes in nutrient availability within this range. Additionally, the stable total chlorophyll content (Figure 8R) indicates that photosynthetic efficiency remains unaffected by variations in RW concentration within this range.

A second pattern, observed between TRW–20% and TRW–30%, reveals contrasting responses in carbohydrate and lipid content (Figure 8J,M), with a decrease in both parameters as the TRW concentration increases. This suggests that microalgae may reallocate resources away from energy storage towards other metabolic pathways or biomass production when exposed to higher concentrations of TRW. The decrease in protein and chlorophyll content further supports this interpretation, indicating a shift in metabolic priorities toward biomass synthesis and growth under nutrient-rich conditions. The consistent carotenoid content suggests that carotenoid biosynthesis remains unaffected by changes in RW concentration within this range, similar to the first pattern.

The depletion of nitrates observed at all TRW concentrations and the depletion of phosphates and sulfates only at TRW10% for Chlorella (Figure 7) indicate efficient nutrient uptake and utilization by both microalgae species. This efficient nutrient assimilation likely contributes to the observed growth parameters, with maximum specific growth rates, biomass productivities, and biomass yields being highest at TRW–20% for both microalgae. The superior performance of Chlorella in terms of growth parameters suggests its greater adaptability to this type of nutrient-rich wastewater compared to Scenedesmus. This contrasts with a previous study reported with the same species, where Scenedesmus consistently outperformed Chlorella when explosives production effluents were bioremediate [13]. This is likely because Chlorella is more resistant than Scenedesmus to wastewater with high organic carbon loads and assimilates COD more efficiently [68].

Overall, the observed changes in the biochemical composition of microalgae reflect their metabolic responses to varying nutrient availability in the TRW-based culture medium. These responses are intricately linked to growth parameters, with nutritional availability influencing metabolic processes and biomass production. Understanding this interplay between biochemical composition of microalgal biomass, nutrient availability and types in TRW, and growth parameters is crucial for optimizing sequential batch photobioreactor strategies with duckweed and microalgae to adjust the biomass produced to the standards required for different applications such as biofuel production, bioremediation, nutraceuticals, aquaculture, and animal feed. In fact, Chlorella and Scenedesmus grown in TRW yield proximate compositions that are within the ranges previously reported for these genera with other culture media [69]. This information could be valuable for scaling up microalgae-based RW bioremediation processes and using the resulting biomass for various applications.

3.3.5. Final RW% Composition after Each Treatment with Duckweed and Microalgae

Table 3. Physicochemical composition of raw cheese whey at concentrations of 10% (v/v) and 20% (v/v), and after sequential treatment with Lemna minor L. MO23 (macrophyte), Chlorella sp. MC18, and Scenedesmus sp. MJ23-R (microalgae).

Parameters	Environmental Regulations					Raw Whey	Treated Whey (TRW)			Raw Whey	Treated Whey (TRW)
	Peru	Italy	Argentina	Brazil	Swiss		Macrophyte	Microalgae			Macrophyte	Microalgae
	MAV 1	ILD 2	LRL 3	CSR 4	SLV 5	RW–10%	L. minor	Chlorella	Scenedesmus	RW–20%	L. minor	Chlorella	Scenedesmus
pH	6–9	5.5–9.5		5–9	6.5–8.5	7.18	7.27	8.47	8.75	7.13	6.47	8.62	8.33
Temperature	<35			≤40		24.7	27.3	22.5	21.6	24.1	27.0	21.1	21.4
OD at 570 nm						0.284	0.015	0.059	0.028	0.549	0.032	0.079	0.051
EC					1400	1107	1203	1223	1327	2030	2183	2203	2290
Turbidity				≤100	25	64.86	6.85	14.27	9.62	128.55	10.67	22.16	13.58
True color						35.6	75.1	95.8	99.2	62.1	43.0	122	123
TSS	500		50			249	10.3	12.1	9.8	358	13.8	14.9	12.7
BOD5	500		125		<5	3505	1146	194.6	217.9	9750	3620	485	1188
COD	1000	160	190		<5	6096.67	2039	321	364	14,885	4190	977	2176
OG	100					180	<0.5	<0.5	<0.5	300	4	<0.5	<0.5
NO3−		20			30	25.4	0.051	<0.005	<0.005	60.4	0.50	0.02	<0.005
NO2−		0.6				3.13	0.011	ND	0.007	3.33	0.067	nd	nd
N–NH3	80		25	≤20		1.95	0.973	<0.02	0.035	3.79	0.299	0.042	0.044
TKN			15			22.7	1.29	0.24	0.27	48.9	3.66	0.96	2.13
PO43−		10 a	2 a	≤0.15 a		104	54.1	10.10	1.64	238	141	22.04	17.22
SO42−					200	29.4	18.8	5.47	0.03	61.4	44.9	11.24	0.16
K+					12	167.4	135.8	115.26	119.674	326.8	265.7	218	209

MAV: maximum admissible value; ILD: limits for discharge in surface water; LRL: law regulatory limits; CSR: conditions and standards regulated (CONAMA 430/11-357/05); SLV: standard limit value (World Health Organization). ¹ MVCS [70]; ² Marazzi et al. [65]; ³ Schierano et al. [71]; ⁴ Schwantes et al. [72]; ⁵ Mekonnen et al. [73]. ^a Regulation for total phosphorus. nd: not detected (<0.005 ppm). Units of all the parameters analyzed are the same as the raw whey (Table 1).

shows the physicochemical composition of RW (10% and 20%, v/v) before and after cultivation with duckweed and subsequently after cultivation with each microalga analyzed. The findings of this study demonstrate the feasibility of the novel sequential culture process with Lemna minor L. MO23 and either Chlorella sp. MC18 or Scenedesmus sp. MJ23-R to produce an effluent that meets environmental regulations for most parameters (see Table 3). This allows the safe discharge of the treated effluent into the environment without causing harm, emphasizing environmental sustainability.

However, the final effluents of RW–10% and RW–20% are not in compliance with potassium (K⁺) regulations. This issue can be addressed by reusing the effluent for agricultural irrigation. Additionally, in the final effluent of RW-20%, some parameters—EC, BOD₅, COD, and PO₄³⁻—are slightly above regulatory limits, but a simple ~50% dilution can solve this problem. Overall, for the referenced states/countries (see Table 3), most of the %TRW media comply with environmental standards, indicating that the sequence of PBR-1 and PBR-2 is feasible for real-world applications.

3.3.6. Scalability Considerations

In addition to demonstrating the effectiveness of PBR-1 and PBR-2 in treating raw whey wastewater, it is important to consider their scalability for industrial applications. The photobioreactor system used in this study was designed for indoor operation, using low-cost LED lighting and multi-tiered cultivation systems inspired by innovative indoor farming techniques [10,11]. Recent advancements in stackable horizontal planar waveguide modules further enhance the scalability of our approach by optimizing space usage and light distribution—critical factors in industrial operations [21]. This controlled environment enables consistent operation regardless of external seasonal variations, ensuring stable light availability and temperature, which are critical for process efficiency and scalability.

However, for outdoor applications where the light source is solar, the stacked photobioreactor system used in this study would be inefficient because of limited light penetration. In such scenarios, open-pond systems or alternative designs that optimize sunlight exposure would be more appropriate for duckweed cultivation. It is important to note that in outdoor systems, seasonal variations would indeed impact treatment efficiency due to fluctuations in irradiance and temperature. These variations could affect the growth rates of both duckweed and microalgae unless additional measures, such as a thermostat system, are implemented.

Overall, while this study serves as proof of concept, the design and technology used are translatable to larger scales. This does not exclude other designs reported in the literature for the cultivation of duckweed and microalgae that could also be effectively coupled with sequential culture systems [21,74]. Future research should focus on pilot-scale experiments to validate these findings and investigate operational efficiency in an industrial context. If outdoor conditions are preferable, the design and performance of alternative systems should be explored in varying seasonal conditions to assess their viability for large-scale implementation.

4. Conclusions

This study demonstrated that conducting sequential batch experiments with duckweed in the PBR-1 and microalgae in the PBR-2 fed by the effluent from the PBR-1 offers interesting advantages. Thus, duckweed effectively removes nutrients from the RW-based culture medium in PBR-1, resulting in an effluent that is less nutrient-/pollutant-rich and more suitable for microalgae growth in the PBR-2. This is, by separating duckweed and microalgae into sequential PBRs, it allows for the customization of growth conditions tailored to each organism’s specific requirements, potentially resulting in enhanced overall productivity of both duckweed and microalgae. Sequential batching minimizes direct competition between duckweed and microalgae for nutrients and light, enabling each organism to thrive without interference from the other. Sequential batch systems offer flexibility in managing different stages of the process, allowing adjustments to optimize conditions for microalgae growth and overall system performance. The sequential approach enhances economic feasibility by maximizing the utility of each biological system, potentially improving the overall economic viability of the process. Both duckweed and microalgae offer potential for resource recovery, such as biofuel production, animal feed, and high-value compounds, allowing for the extraction of diverse resources from the effluent.