Effects of Sugar Beet Pulp Pretreatment Methods on Hydrogen Production by Dark Fermentation

Abstract

Methane and hydrogen generated from waste and biomass are renewable resources, which may successfully replace traditional fossil fuels. This paper investigates the enhancement effect of lignocellulosic biomass pretreatment on dark fermentative hydrogen production from sugar beet pulp (SBP). The results showed that sugar beet pulp after pretreatment contained significant amounts of unfermented sugars (mainly glucose, arabinose, galactose, and raffinose), and, therefore, represented an attractive substrate for methane and hydrogen production. The greatest methane yield (495 dm³ CH₄/kg VS) was achieved from sugar beet pulp after alkaline pretreatment. High methane production of up to 445 dm³ CH₄/kg VS was also obtained using acidic and enzymatic hydrolysis as a preliminary treatment of the pulp. All the pretreatment methods also resulted in the enhancement of hydrogen yield with the highest value of 229 dm³ H₂/kg VS achieved using acid hydrolysis compared with 17 dm³ H₂/kg VS for raw material subjected to digestion.

1. Introduction

The growing demand for energy leads to an increased use of fossil fuels, the amount of which is constantly decreasing. Furthermore, conventional energy sources such as oil, coal, and natural gas have been proven to be damaging to the environment mainly due to the emission of greenhouse gases into the atmosphere. On the other hand, the potential of renewable energy sources is enormous as they can, in principle, meet many times the world’s energy demand [1]. In this context, lignocellulosic biomass is becoming more attractive thanks to its low cost, zero greenhouse gas emission, and wide availability. It is mainly composed of carbohydrates (cellulose and hemicellulose) and lignin as well as other minor contents, including pectins, proteins, and inorganic materials. The source of lignocellulosic biomass for the production of biofuels can be inedible raw materials or residual agriculture, forest, paper waste, or municipal solid waste. Lignocellulosic biomass has been successfully applied to the production of second-generation bioethanol [2]. The energetic potential of lignocellulosic biomass is related to the abundance of sugars built into the structures of cellulose and hemicellulose. Hence, the idea is to transform industrially useless lignocellulosic waste into fermentable sugars, which, in turn, can be converted into valuable products [3]. However, to liberate sugars from the lignocellulosic structure, it is necessary to break down the lignin coating that surrounds the cellulose and hemicellulose fibers and limits its microbial degradation [4]. Pretreatment methods are categorized into physical, chemical, physicochemical, and biological ones. Physical methods are used to grind biomass into smaller pieces before further chemical and biological processing [5]. In turn, chemical pretreatment is aimed at breaking down lignocellulosic structures of biomass by concentrated acids or is based on higher temperatures. However, these conditions favor the generation of by-products, including furfural, hydroxymethylfurfural, vanillin, and levulinic acid, which inhibit fermentation processes in the subsequent biological step [6]. Considering this, the use of diluted acids or bases and lower processing temperatures reduce the potential for the formation of inhibitory compounds, but biomass decomposition is less efficient. The selection of the pretreatment method should include these aspects as well as technical-economic viability and sustainability [7].

Sugar beet pulp (SBP) is processing waste received after sugar production in large quantities. According to the Central Statistical Office and Institute of Agricultural Economics and Food Economy data, around 2.15 million tons of sugar was produced in Poland in the period from September 2021 to January 2022. To produce 150 kg of sugar, 1000 kg of sugar beet is used, which results in the generation of 250 kg of wet pulp. This year only, in Poland, 7.88 million tons of sugar beet pulp was received. Sugar beet pulp belongs to lignocellulosic biomass and consists mostly of carbohydrates (cellulose 20–25%, hemicellulose 20–30%, lignin 5–6%) and polymeric saccharides (pectin 15–25% of the dry matter) [8]. Processing waste contains large amounts of water and has a high biodegradable organic load, which makes it susceptible to fermentation and a cause of environmental pollution (producing fetor and generating harmful substances). Huge heaps of SBP violate the pollution control norms and industrial safety issues. Moreover, the costs of traditional treatment and sugar beet pulp waste management are high. Methane or hydrogen production is a great alternative that allows for waste disposal and energy production in the same process [9]. Biohydrogen can be generated via several different metabolic pathways, including direct water biophotolysis, biophotolysis, photofermentation, and dark fermentation (DF). A good direction for using this waste is the production of hydrogen through dark fermentation. This process could accommodate various biodegradable organic wastes with high carbohydrate content to produce gas rich in hydrogen. Moreover, H₂ is a clean energy carrier and a promising alternative to fossil fuels, one that is less energy-intensive and more environmentally friendly in terms of its potential to contribute to the reduction of global carbon dioxide emissions. Hydrogen has the highest energy yield among known fuels (122 kJ/g), which is more than two times that of methane (50.1 kJ/g) [10].

Therefore, the main aim of the present study was to improve the efficiency of hydrogen and methane production from sugar beet pulp under mesophilic conditions (35 °C), using various pretreatment methods. The SBP biomass was first preliminary treated by selected enzymes, NaOH and H₂SO₄, and then subjected to anaerobic treatment in the subsequent step. Furthermore, the impact of various types of hydrolyzing agents under mild temperature and pressure conditions was examined.

2. Materials and Methods

2.1. Substrate

The characteristics of substrates and inoculum are depicted in

Table 1. Characteristics of inoculum and sugar beet pulp.

Indicator	Unit	Inoculum	Raw SBP	Enzymatic Pretreatment	Alkaline Pretreatment	Alkaline and Enzymatic Pretreatment	Acidic Pretreatment
TS	g/kg	32.61 ± 0.61	212.33 ± 0.25	47.56 ± 0.62	66.98 ± 1.05	51.79 ± 0.78	54.87 ± 0.58
VS	g/kg	21.21 ± 0.52	197.45 ± 0.42	40.14 ± 0.48	53.92 ± 0.79	38.00 ± 0.82	39.62 ± 0.45
COD	gO2/kg	3.25 ± 0.02	255.13 ± 0.05	121.19 ± 0.02	198.47 ± 0.08	164.24 ± 0.04	171.71 ± 0.10
Carbon	% TS	59.60 ± 0.61	69.25 ± 0.15	-	-	-	-
Nitrogen	% TS	3.30 ± 0.15	0.14 ± 0.06	-	-	-	-
Phosphorus	% TS	1.97 ± 0.08	0.02 ± 0.01	-	-	-	-
Hydrogen	% TS	5.95 ± 0.12	7.58 0.48	-	-	-	-
Sulfur	% TS	0.93 ± 0.18	0.02 0.01	-	-	-	-
C/N	-	18.72 ± 0.96	361.64 ± 1.65	-	-	-	-
TAN	gN/m3	2337.54 ± 1.08	5.48 ± 0.45	70.05 ± 1.15	35.45 ± 0.02	72.02 ± 0.42	49.58 ± 0.56
P-PO43−	gP/m3	307.57 ± 1.45	0.28 ± 0.08	9.11 ± 0.08	2.12 ± 0.05	9.65 ± 0.12	6.25 ± 0.14
pH	-	7.46 ± 0.02	5.87 ± 0.09	3.74 ± 0.02	13.05 ± 0.08	4.24 ± 0.05	1.87 ± 0.12
Glucose	g/dm3	-	6.14 ± 0.65	21.71 ± 1.11	6.25 ± 0.01	26.27 ± 0.29	20.62 ± 0.31
Fructose	g/dm3	-	3.30 ± 0.80	6.97 ± 0.35	1.16 ± 0.01	6.16 ± 0.05	4.62 ± 0.15
Mannose	g/dm3	-	1.87 ± 0.75	5.61 ± 0.12	0.45 ± 0.01	4.41 ± 0.56	3.94 ± 0.08
Arabinose	g/dm3	-	0.62 ± 0.02	8.04 ± 0.22	0.85 ± 0.06	10.38 ± 079	5.64 ± 0.25
Galactose	g/dm3	-	0.54 ± 0.08	9.56 ± 0.55	0.91 ± 0.01	16.47 ± 1.09	7.56 ± 0.17
Raffinose	g/dm3	-	0.86 ± 0.03	5.89 ± 0.68	0.39 ± 0.04	0.20 ± 0.02	4.86 ± 0.44
Rhamnose	g/dm3	-	1.07 ± 0.04	1.75 ± 0.18	2.71 ± 0.21	0.55 ± 0.05	0.87 ± 0.06
Xylose	g/dm3	-	0.25 ± 0.07	0.48 ± 0.12	1.04 ± 0.53	6.77 ± 0.54	0.31 ± 0.02
Galacturonic acid	g/dm3	-	0.22 ± 0.01	3.98 ± 0.36	0.00 ± 0.00	8.47 ± 0.48	1.02 ± 0.08
Total sugar content	g/dm3	-	14.87 ± 0.56	78.86 ± 1.59	13.76 ± 0.24	79.68 ± 1.36	49.44 ± 1.12
Cellulose	%	-	19.57 ± 0.25	22.15 ± 0.84	19.42 ± 1.12	21.98 ± 1.08	20.11 ± 1.44
Hemicellulose	%	-	27.96 ± 0.14	13.87 ± 1.56	14.06 ± 1.22	11.12 ± 1.05	16.56 ± 1.62
Lignin	%	-	2.22 ± 0.02	1.18 ± 0.04	2.08 ± 0.12	1.17 ± 0.08	1.45 ± 0.06

± Standard deviation.

. Anaerobic sludge was used as mesophilic inoculum for the batch experiments and was obtained from the anaerobic mesophilic digester at the municipal wastewater treatment plant in Lodz, Poland. The inoculum had total solid (TS) and volatile solid (VS) concentrations of 33 gTS/kg and 21 gVS/kg, respectively (Table 1). The SBP originated from the sugar factory in Dobrzelin (Poland) and was stored at −18 °C until its use for the experiments.

2.2. Pretreatment of SBP

2.2.1. Enzymatic Pretreatment

The pretreatment experiments were performed with two enzymatic preparations Ultraflo Max (main enzyme activity: cellulase (endo-1,4-β-glucanase) and Viscozyme L (main enzyme activity: beta-glucanase (endo-1,3(4)-β-glucanase) (Novozymes, Bagsvӕrd, Denmark). The SBP biomass was subjected to enzymatic hydrolysis for 10 h at 45 °C and pH = 5.0 [11].

2.2.2. Acidic Pretreatment

Acid pretreatment is more effective in solubilizing hemicellulose and pectin and generally utilizes inorganic acids. Acidic SBP hydrolysates were obtained by pretreatment with a 5% sulfuric acid solution (by volume). The mixture was placed in an Erlenmeyer flask (1 dm³), covered with aluminum foil, and stored in a water bath at 90 °C for 4 h. These conditions were adopted to break down lignocellulosic structures of SBP in order to release easily degradable sugars for methane and hydrogen production.

2.2.3. Alkaline Pretreatment

Alkaline pretreatment is focused on the solubilization of lignin and some hemicelluloses. Alkaline SBP hydrolysates were obtained by using a 20% solution (w/v) of NaOH (Stanlab, Lublin, Poland) at pH 13.0. The mixture was placed in an Erlenmeyer flask (1 dm³), covered with aluminum foil, and stored in a water bath at 90 °C for 4 h.

2.2.4. Alkaline and Enzymatic Pretreatment

The main goal of alkaline pretreatment used before the enzymatic hydrolysis is to improve the digestibility of the lignocellulosic biomass by changing its structure and making cellulose more accessible to enzymes. The hydrolysis process was performed by mixing enzymatically pretreated SBP with 20% H₂SO₄ (Chempur, Piekary Śląskie, Poland) in an Erlenmeyer flask (1 dm³) to achieve pH = 5.0. The flask was then incubated at 45 °C for 10 h. Each experiment was carried out in triplicate.

2.3. Experimental Setup for Anaerobic Digestion Experiments

The biochemical hydrogen and methane potentials of the hydrolyzed SBP were examined in glass reactors. Briefly, each reactor had a total volume of 1000 cm³ and a working volume of 500 cm³ (Figure 1). The inoculum-to-substrate ratio was 2:1, based on VS. During the experiment, the biogas produced from each bottle was collected into the glass tank for analysis and measurement by the water displacement method as described earlier [12]. After the substrate and inoculum addition, the reactors were flushed with nitrogen gas to remove oxygen, then closed with plastic caps, and placed in an incubator in mesophilic conditions (35 ± 1 °C). In order to homogenize the working volume and avoid the formation of dead zones, the reactors were manually shaken twice a day. The experiments were conducted in triplicates.

2.4. Preparation of Microorganism and Inoculum

Inoculum contains microflora that produces hydrogen and, also, archeones responsible for methane generation. To avoid the activity of methanogens, the inoculum was thermally pretreated at 80 °C for 1.5 h before being placed in the reactors. Moreover, during the experiments, pH in the reactors was maintained at approximately 5.5 using a 20% NaOH solution.

2.5. Analytical Methods

2.5.1. Elemental Analysis

Total solids (TS), volatile solids (VS), pH [10], ortophosphates (PhosVer^® 3, test no. 8048), ammonium nitrogen (TAN, modified Nessler method, test no. 8038), and chemical oxygen demand (COD, modified Nessler method, test no. 8000, LCK 514) were measured according to the methodology described in previous research [12]. Determination of trace elements (C, N, H, P, S) was performed via Flash Elemental Analyzer (Thermo Finnigan, Monza, Italy), following the manufacturer’s procedures. The C/N ratio was calculated by dividing total carbon by total nitrogen.

2.5.2. Biogas Composition

Biogas yield was monitored on a daily basis by the water displacement method as described anywhere in the literature. The hydrogen and methane volumes were calculated to NmL (normal milliliter, the volume conversion at STP condition), which is comparable with other studies. Hydrogen and methane concentrations in batch reactors were measured using a portable gas analyzer GA-21plus (Madur, Zgierz, Poland), equipped with electrochemical sensors, which determined the concentrations of O₂, H₂, CO₂, CH_4, and H₂S. The bottles were not vented during the whole experiment.

2.5.3. Polysaccharide Content

Cellulose, hemicellulose, and lignin were determined under optimal conditions according to the laboratory analytical procedures of the National Renewable Energy Laboratory (NREL) [13].

2.5.4. Sugar Content

The monosaccharide profile in the stillage was determined by UV-spectrophotometry (Thermo Scientific Multiskan GO; Thermo Fisher Scientific, Munich, Germany) and using Megazyme Kits (Megazyme Ltd., Bray, Ireland), following the manufacturer’s procedure. Samples for monosaccharide contents were prepared by centrifuging the digestate for 10 min at 4000 rpm.

2.5.5. Scanning Electron Microscopy

Scanning electron microscopy (SEM) (JSM-6610LV, JEOL Ltd., Tokio, Japan) was used for the microstructural analysis of the raw and pretreated SBP [14]. All images were taken at an accelerating voltage of 30 kV. The physical changes in the material are shown in representative images.

2.5.6. Crystallinity Index

The crystallinity of sugar beet pulp was analyzed using X-ray diffraction (XRD). Each sample was ground to powder to obtain uniform X-ray exposure. Diffractograms were received using an Empyrean powder diffractometer 7602 EA Almeo (Malvern Panatytical, Malvern, United Kingdom) with Cu Kα radiation source of λ = 1.79 °A. Diffracted intensities were recorded in continuous scan mode at room temperature over the angular range of 10° to 40°, under a step size of 0.1° and a time per step 4 s using an Xe proportional counter with mounted 0.18° parallel plate collimator.

The crystallinity index (IC) was determined using the Segal equation [15]:

$$CI={\displaystyle \frac{I-I_{am}}{I_{002}}}\times 100$$

(1)

where I₀₀₂ is the intensity of the peak at 2θ ≈ 26, representing crystalline material and I_am is the counter reading at 2θ ≈ 21, representing amorphous material in the samples.

2.5.7. Individual Volatile Fatty Acids

Individual volatile fatty acid concentrations (acetic, formic/propionic, i-butyric, n-butyric, i-valeric, n-valeric, i-caproic, n-caproic, and heptanoic acid) were analyzed using the gas chromatograph (GC) (Agilent 7890A, Santa Clara, CL, USA) equipped with a flame ionization detector (FID) and a high-polarity, nitroterephthalic-acid-modified polyethylene glycol (PEG) column (DB FFAP, 30 m length × 0.32 mm ID, 0.25 µm df, Agilent Technologies, Santa Clara, CL, USA). Helium was used as a carrier gas at a flow rate of 1.2 mL/min. Approximately 2 µL of a sample was injected to GC. The initial temperature of a GC column was 100 °C (1.5 min hold), which was increased at a rate of 8 °C/min to the final temperature of 200 °C (hold 4.0 min). The injector temperature was set at 220 °C, while that of FID was set at 220 °C, split 1:10, H₂ 30 mL/min, air 400 mL/min, makeup He 20 mL/min.

2.6. Statistical Analysis

All the analytical tests were performed in triplicates. The mean and standard deviation values and error bars were calculated in Microsoft Excel 2010 and in R version 3.5.0.

3. Results

3.1. Pretreatment of Sugar Beet Pulp

Generally, pentoses and hexoses are the best carbon sources for dark fermentation. However, hydrogen can also be produced efficiently and economically using lignocellulosic substrates, such as sugar beet pulp. The dry basis of SBP is composed mainly of carbohydrates (20–25% cellulose, 20–30% hemicellulose, 15–25% pectin) with small amounts of fats, proteins, ash, and lignin (1–2%, 10–15%, 3–4%, and 5–6%, respectively) [16]. In contrast to the lignocellulosic materials, SBP has a low lignin content (approximately 2%) and, therefore, does not need severe pretreatment to liberate simple sugars [16]. Effects of different pretreatments on the SBP solubilization were evaluated based on TS and VS changes as well as considering individual sugars. The composition of carbohydrates in sugar beet pulp showed the dominating amounts of glucose (derived from cellulose and hemicellulose), xylose, glucose, mannose, galactose, arabinose (from hemicelluloses), galacturonic acid, galactose, and rhamnose (from pectin) [16,17]. Hydrolysis of lignocellulosic materials is a crucial step for hydrogen production in the subsequent dark fermentation. Carbohydrates constituted the main part of SBP before and after pretreatment. As shown in Table 1, the sugar composition of SBP was extensively modified during the hydrolysis. In agreement with the increase in cellulose, the glucose content showed a parallel increase to become the dominant constituent of the SBP hydrolysate. Regarding the release of sugars, the most effective treatment turned out to be a combination of alkaline and enzymatic processing of SBP, for which as much as 79.68 g/dm³ of sugars, including 26.27 g/dm³ of glucose, was obtained in the hydrolysate. The enzymatic pretreatment had only a slightly lower efficiency since nearly 79 g/dm³ of sugars were released, including 21.71 g/dm³ of glucose. These results demonstrate that chemical pretreatment does not significantly increase the efficiency of the subsequent enzymatic hydrolysis but only generates additional costs for the entire process. Enzymatic hydrolysis is widely used to degrade cellulose, whereas alkali pretreatment is applied to increase the porosity of lignocellulose biomass by breaking the bonds in between lignin and hemicellulose [17]. The goals for dilute acid pretreatment of SBP were to solubilize hemicellulose and make residual cellulose more degradable by microorganisms [18]. Both temperature and strongly acidic conditions had significantly positive effects on hemicellulose decomposition from solids. However, as mentioned above, the combination of the acid and enzyme treatment did not significantly increase the release of sugars compared with the enzymatic treatment alone, which is in agreement with the findings reported in our previous study [18]. One of the possible explanations is that sugars might have degraded to inhibitory by-products. The combined pretreatment resulted in a strong color change and the appearance of a caramel-like aroma, which indicated maillardation under the conditions used. Biomass pretreatment carries the risk of forming undesired degradation products that could negatively influence further use. Previous studies showed the formation of formic acid, acetic acid, lactic acid, furfural, and hydroxymethylfurfural as decomposition products [18].

A relatively high crystallinity index (CI) of raw pulp (56.7%) may have resulted from the processing of sugar beets in a sugar factory. Beet pulp does not have a uniform structure, which probably affects the presence of crystalline and amorphous regions in it. They are formed during mechanical and physical processing, which probably also affects their internal structure. Physical treatment reduces the crystallinity of cellulose, which may also be a possible reason. The CI value of beet pulp after acid hydrolysis (5% sulfuric acid) was 41.85%. The decrease in this value may have resulted from the decomposition of cellulose by sulfuric acid or by swelling of the cellulose [19]. During the decomposition of rye straw with 5% and 10% sulfuric acid, a similar correlation was observed when the CI value decreased by approx. 15% and 45%, respectively [14]. The crystallinity index of pulp treated with sodium hydroxide was 51%. However, after the pretreatment with sodium hydroxide and enzyme, CI increased to 56%. Alkaline pretreatment results in simultaneous decomposition of crystalline and amorphous regions but with less negative effect on crystalline regions, especially at lower temperatures [20]. Li and colleagues [20], conducting the pretreatment of beet pulp with aqueous ammonia (10% wt, 80 °C, 6 h), observed an increase in the CI value by 88%. Enzymatic treatment increased the CI level to 76%. This increase was probably due to the removal of amorphous regions from hemicellulose, lignin, and, probably, cellulose as reported previously [21].

The SEM images of raw beet pulp show the changes that occurred during its generation in the sugar factory. The surface is heterogeneous and jagged. Internal structures are visible in places where the beet pulp was cut (Figure 2A,B). The impact of 5% sulfuric acid led to the solidification of the beet pulp surface and the formation of acid crystals or their salts on the surface of the pulp (Figure 3I,J). The effect of the alkaline pretreatment resulted in the smoothing and rounding of both the surface and structures of the beet pulp (Figure 2E,F).

3.2. Batch Tests

The influence of pretreatment with NaOH, H₂SO₄, and enzymes on dark hydrogen fermentation was investigated in batch experiments under mesophilic conditions.

Table 2. Parameters of the anaerobic digestion tests.

Parameter	Unit	Raw SBP	Enzymatic Pretreatment	Alkaline Pretreatment	Alkaline and Enzymatic Pretreatment	Acidic Pretreatment
Mass of substrate	g	18	133	98	140	134
Mass of inoculum	g	500	500	500	500	500
Duration time	days	9	14	9	9	9
pHinitial	-	7.12	7.08	6.98	6.99	7.10
pHend	-	6.89	6.51	6.45	6.50	6.35
Specific gas production (SGP)	dm3/kg VS	239.05 ± 1.25	1302.21 ± 5.02	1240.03 ± 2.10	1166.90 ± 1.08	1293.04 ± 1.10
Specific methane production (SMP)	dm3CH4/kg VS	120.87 ± 0.15	411.23 ± 0.36	494.81 ± 0.88	406.35 ± 0.82	444.91 ± 0.75
CH4 content	%	5.68 ± 0.10	55.96 ± 0.22	54.88 ± 0.18	57.46 ± 0.29	29.58 ± 0.16
Specific hydrogen production (SHP)	dm3H2/kg VS	15.54 ± 0.52	102.15 ± 0.42	92.65 ± 0.35	82.43 ± 0.61	95.15 ± 0.74
H2 content	%	1.68 ± 0.08	14.51 ± 0.11	11.32 ± 0.10	6.86 ± 0.05	1.44 ± 0.02
CO2 content	%	32.15 ± 0.14	11.18 ± 0.12	12.26 ± 0.12	11.02 ± 0.09	15.65 ± 0.08
Acetic acid	g/dm3	0.258 ± 0.00	0.524 ± 0.00	0.464 ± 0.00	0.525 ± 0.00	0.575 ± 0.00
Propionic acid	g/dm3	0.114 ± 0.00	0.445 ± 0.00	0.399 ± 0.00	0.377 ± 0.00	0.413 ± 0.00
i-Butyric acid	g/dm3	0.021 ± 0.00	0.187 ± 0.00	0.165 ± 0.00	0.097 ± 0.00	0.131 ± 0.00
n-Butyric acid	g/dm3	0.018 ± 0.00	0.045 ± 0.00	0.066 ± 0.00	0.019 ± 0.00	0.031 ± 0.00
i-Valeric acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.012 ± 0.00
n-Valeric acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00
i-Caproic acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00
n-Caproic acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00
i-Heptanoic acid	g/dm3	0.001 ± 0.00	0.010 ± 0.00	0.006 ± 0.00	0.004 ± 0.00	0.007 ± 0.00
Total VFA	g/dm3	0.408 ± 0.00	1.211 ± 0.00	1.100 ± 0.00	1.022 ± 0.00	1.169 ± 0.00

± Standard deviation.

and

Table 3. Parameters of the dark fermentation experiments.

Parameter	Unit	Raw SBP	Enzymatic Pretreatment	Alkaline Pretreatment	Alkaline and Enzymatic Pretreatment	Acidic Pretreatment
Mass of substrate	g	18	133	98	140	134
Mass of inoculum	g	500	500	500	500	500
Duration time	days	9	14	9	9	9
pHinitial	-	5.48	5.52	5.45	5.55	5.52
pHend	-	5.41	5.28	5.36	5.32	5.26
Specific gas production (SGP)	dm3/kg VS	26.89 ± 0.05	375.12 ± 0.25	161.65 ± 0.11	370.45 ± 0.28	498.45 ± 0.50
Specific methane production (SMP)	dm3CH4/kg VS	1.00 ± 0.02	4.28 ± 0.05	14.10 ± 0.18	4.69 ± 0.22	17.99 ± 0.45
CH4 content	%	0.10 ± 0.01	0.29 ± 0.01	3.91 ± 0.24	0.36 ± 0.01	0.72 ± 0.02
Specific hydrogen production (SHP)	dm3H2/kg VS	17.14 ± 0.08	220.59 ± 1.28	50.46 ± 0.68	136.52 ± 0.86	229.24 ± 0.88
H2 content	%	2.40 ± 0.02	42.01 ± 0.87	13.89 ± 0.05	9.67 ± 0.25	9.20 ± 0.45
CO2 content	%	22.05 ± 0.12	15.21 ± 0.09	20.78 ± 0.14	18.05 ± 0.12	13.54 ± 0.21
Acetic acid	g/dm3	0.221 ± 0.00	0.825 ± 0.00	0.457 ± 0.00	0.871 ± 0.00	0.751 ± 0.00
Propionic acid	g/dm3	0.202 ± 0.00	0.172 ± 0.00	0.115 ± 0.00	0.198 ± 0.00	0.272 ± 0.00
i-Butyric acid	g/dm3	0.031 ± 0.00	0.141 ± 0.00	0.086 ± 0.00	0.145 ± 0.00	0.141 ± 0.00
n-Butyric acid	g/dm3	0.015 ± 0.00	0.023 ± 0.00	0.019 ± 0.00	0.020 ± 0.00	0.083 ± 0.00
i-Valeric acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.004 ± 0.00	0.068 ± 0.00
n-Valeric acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00
i-Caproic acid	g/dm3	0.000 ± 0.00	0.000 ± 0.00	0.000 ± 0.00	0.011 ± 0.00	0.043 ± 0.00
n-Caproic acid	g/dm3	0.004 ± 0.00	0.001 ± 0.00	0.002 ± 0.00	0.002 ± 0.00	0.005 ± 0.00
i-Heptanoic acid	g/dm3	0.001 ± 0.00	0.007 ± 0.00	0.002 ± 0.00	0.001 ± 0.00	0.000 ± 0.00
Total VFA	g/dm3	0.472 ± 0.00	1.149 ± 0.00	0.681 ± 0.00	1.252 ± 0.00	1.363 ± 0.00

± Standard deviation.

summarize the operating parameters and performances of the anaerobic digestion and dark fermentation experiments. A batch reactor type was applied because it allows for establishing biohydrogen and biomethane potential. After 9 days, no considerable gas was detected for almost all samples. Therefore, the experiments were stopped, and the samples were analyzed for volatile fatty acid (VFA) concentrations. Moreover, their final pH was measured and reported (Table 2 and Table 3). The cumulative biohydrogen, biomethane, and biogas yields of the experiments are presented in Figure 3.

3.2.1. Methane Production

As shown in Figure 3, biogas production within 9 days for untreated material reached 239 dm³/kg VS. However, at the same time, enzymatic pretreatment allowed for an increase in biogas yield to 1302 dm³/kg VS. The corresponding methane yield was 411 dm³/kg VS.

The SBP demonstrated significant biogas production from day 2 of the experimental run. These yields are generally comparable to the values reported in the literature for similar substrates. Ziemiński et al. [22] investigated the mixture of sugar beet pulp silage and vinase and reported nearly 500 dm³ CH₄/kg VS for the materials after alkaline pretreatment, which was 4-fold greater than methane yield from raw substrates.

Also, in our previous study [23], the application of pretreatment resulted in a significant increase in methane production from 125 dm³ CH₄/kg VS for the control run to 390 dm³ CH₄/kg VS for pretreated material. The methane contents of biogas from all experiments were almost in the same range, varying between 55% and 57%, except for the control run and run with SBP after acidic pretreatment, with the methane content of 6% and 29%, respectively. The highest methane content of 57% was obtained in the run with combined pretreatment (Table 2). Slight amounts of hydrogen were noted in all batch tests (2–15%). In the present study, no toxic effects of chemicals on anaerobic digestion were observed. This can be explained by the fact that anaerobic microorganisms are fairly resistant to Na⁺ ions, which is confirmed by the results presented by Bala et al. [24].

3.2.2. Hydrogen Production

Specific gas production for pretreated SBP was higher (162–498 dm³/kg VS) compared with the raw SBP subjected to digestion (27 dm³/kg VS). The biogas is mainly composed of hydrogen, carbon dioxide, and trace amounts of methane (0.1–3.9%). The percentages of hydrogen and carbon dioxide in gas were 2.4–42% and 15–22%, respectively. Different results were received by Ozkan et al. [9]; they reached cumulative hydrogen yields in the range of 112–149 cm³ from the sets with pretreatment substrate, whereas the samples with no pretreatment produced only 90.1 cm³ of hydrogen. The current research has also demonstrated that hydrogen could hardly be produced from raw sugar beet pulp (only 17.14 dm³ H₂/kg VS; 105 cm³); this value is similar to the one obtained by Ozkan et al. [9] in their research. However, when SBP was subjected to preliminary treatment, much greater volumes of this gas were obtained. Ozkan et al. [9] studied the effect of different pretreatment methods such as alkaline, thermal, microwave, thermal-alkaline, and microwave-alkaline, and showed that alkaline-pretreated SBP exhibited the highest H₂ yield (148.5 cm³, 115.6 dm³/kg COD) and achieved 10 % higher hydrogen production compared with the single pretreatment method (alkaline, 134 cm³ cumulative H₂). The current publication showed that cumulative H₂ yields were within the range of 120–745 cm³, and the highest hydrogen yield was obtained for enzymatic pretreatment (102.15 dm³ H₂/kg VS; 745 cm³). Discrepancies between the results from those sets and the literature may be due to the type of substrate, enzyme, or inoculum used in sets. As shown in Figure 3, hydrogen yield significantly increased with enzymatic hydrolysis applied. The hydrogen production from enzymatically pretreated SBP was, therefore, 13-times greater than from raw biomass and reached 220.59 dm³ H₂/kg VS. The maximum H₂ yield observed was 229.24 dm³ H₂/kg VS for SBP after acidic pretreatment. In contrast, the H₂ yield was only 50.46 dm³ H₂/kg VS when SBP was treated with sodium hydroxide. Furthermore, the amount of H2 produced corresponded to the concentrations of sugars released during the treatment. Hence, the lowest hydrogen yield was observed for the corresponding low sugar concentrations in the experiments with raw SBP and after alkaline pretreatment (Table 1). Likewise, in our previous research, the highest hydrogen production was noted for enzymatically treated SBP than that obtained for acidic hydrolysates [18]. This could be linked to the formation of specific by-products, including furfural and hydroxymethylfurfural (both arising from the dehydration of sugar) [25] and vanillin (originated from lignin depolymerization) [26], which might have inhibited the activity of hydrogen-producing bacteria. Hydroxymethylfurfural and furfural are formed during acid hydrolysis of lignocellulosic biomass, mainly from glucose and xylose [27]. Akobi et al. [28] observed that at furfural concentrations higher than 2.0 g/dm³, the hydrogen production may be notably lower. In our previous studies, the concentration of this compound exceeded 2.0 g/dm³ only in the experiments with thermochemical pretreatment [18]. An interesting solution to the problem of inhibitors was proposed by Zoppi et al. [29]. The pre-hydrogenation step to sugar alcohols was evaluated to avoid homogeneous reactions and without side products. This procedure allowed for obtaining a much higher hydrogen yield compared with processes without this stage. In the present research, the highest SHP obtained from acidic hydrolysates could be linked to relatively mild conditions applied (lower temperature and pressure), which could limit the formation of inhibitory compounds. During the experiments, the gas volume and composition were measured each day and the daily values were plotted to obtain the cumulative curves (Figure 3). The highest hydrogen production was observed in the first two days of the experiment reaching a maximum value after 48 h and practically stopping in the following days of DF. In the previous research, hydrogen was generated for 8 days; hence, enzymatic pretreatment could significantly reduce this time [18].

The analysis of the metabolic products provides useful information on the evolution of the AD and DF process and can be used to explain the observed H₂ and CH₄ yields. It is widely recognized that pH is a crucial factor governing both methane and hydrogen fermentation performance as well as enzymatic activity. The optimal pH values for classical anaerobic digestion reported in the literature vary between 6.8 and 8.2 [30]. In contrast, the maximum hydrogen yields are observed at pH values close to 5.5, whereas a pH value lower than 4.0 significantly inhibits hydrogen production.

As shown in Table 2, the final pH values (measured in the digestates) reported in experiments with pretreatment were slightly lower than in the control runs. Specifically, the pH value measured in experiments with pretreatment lay in the range of 6.35–6.51 for AD and 5.26–5.36 for DF. In turn, the corresponding values reported in the control runs (with no pretreatment) were 6.8 and 5.41, respectively. Generally, the pH drop observed during digestion results from the production of organic acids. Ziemiński and Wentel [31] noted a similar pH decrease (from 7.18 to 6.90) during anaerobic digestion of SBP after thermal and pressure treatment. Despite methanogenesis being commonly known to proceed between pH 6.6 and 7.8, the activity of methanogens was also noted when the DF reactors operated at pH 5.5 [32,33]. Previous experiments determined that archeons in the digestate belonged to the genus Methanosphaera [33], which could tolerate acidic conditions and grow at the pH value of 5 [34]. The presence of methanogens could suggest that thermal pretreatment of inoculum did not completely activate these organisms, which was also confirmed by Venkata Mohan et al. [35].

During the digestion processes, substrates are initially transformed to VFA, H₂, and CO₂ by hydrolytic and fermentative bacteria. Volatile fatty acids such as propionic, butyric, and iso-butyric acid are subsequently oxidized by acetogenic bacteria producing acetate, H₂, and CO₂, and finally, these products are converted to CH₄, CO₂, and H₂O by methanogens [36]. In both AD and DF experiments, five organic acids were detected in the digestates, including acetic acid, propionic acid, butyric acid, caproic acid, and heptanoic acid. Among them, acetic and butyric acids are the main metabolic products of the dark fermentation process of carbohydrate-rich materials [18]. Similar VFA composition (acetic acid, butyric acid, and propionic acid) received from sugar beet pulp juice was obtained by Dhar et al. [37] The distribution of key volatile fatty acids produced in the experiments as well as the corresponding final pH are summarized in Table 2 and Table 3. In the experiments with pretreated SBP, VFA concentrations (1.022–1.211 g/dm³) were higher than those in the control reactor (0.408 g/dm³). The concentrations of acetic acid, propionic acid, i-butyric acid, and n-butyric acid in the blank experiment were 0.221 g/dm³, 0.202 g/dm³, 0.031 g/dm³, and 0.015 g/dm³, respectively, and were lower than the corresponding values obtained for the hydrolyzed SBP after digestion. The abundance of organic acids varied depending on the pretreatment method applied and, therefore, the efficiency of the digestion process. Lower VFA concentrations have no significant effect on methanogenesis. However, high concentrations of carboxylic acids in combination with low pH could inhibit methane production. Franke-Whittle et al. [38] found that 0.9 g/dm³ of propionate (or greater than 18% of total VFA) inhibits activity of the anaerobic community, especially methanogens, thereby leading to process failure. In each experiment, the concentration of propionic acid (0.11–0.45 g/dm³) did not exceed this threshold; however, this acid represented 36% of total VFA in the experiment with alkaline pretreatment at which the highest methane production was reported. This suggested no inhibition of methanogenesis. In the dark fermentation experiments, the VFA concentrations varied between 0.475 g/dm³ for the control run to 1.149 g/dm³, 0.681 g/dm³, 1.252 g/dm³, and 1.363 g/dm³ for the experiments with enzymatic, alkaline, enzymatic-alkaline, and acidic pretreatments, respectively. Not surprisingly, higher concentrations of sugars corresponded to greater amounts of VFA produced. The exception was acidic pretreatment where sugar content was almost two times lower than that obtained in the other treatment methods and the concentration of the VFA was the highest (1.363 g/dm³). Regarding the distribution of individual acids, the concentrations of acetic acid were greater in the DF effluents (221–871 g/dm³) than in the AD digestates (258–575 g/dm³) because this acid is easily converted into methane. In turn, caproic and valeric acids showed the highest concentrations in the experiments with acidic pretreatment, which is consistent with our previous findings [16].

4. Conclusions

This study showed that the anaerobic digestion and dark fermentative hydrogen production could be effectively enhanced by the pretreatment of sugar beet pulp. The highest methane production of nearly 500 dm³/kg VS was obtained from SBP previously subjected to alkaline pretreatment, whereas enzymatic and acidic pretreatment greatly improved hydrogen production up to 229 dm³/kg VS. However, during dark fermentative hydrogen production, a noticeable methane yield was also observed despite low pH and thermal pretreatment applied to substrates and inoculum. The proposed pretreatment method does not require large amounts of chemicals, and the temperatures applied are relatively low, thus this method can be an attractive alternative to more rapid techniques applied to plant biomass.