Phosphorus-Containing Catalyst Impact on Furfural and Glucose Production during Consecutive Hydrothermal Pretreatment and Enzymatic Hydrolysis

Abstract

Lignocellulosic biomasses have a very important role as raw materials to produce biobased chemicals. However, a sustainable, efficient, and economically competitive way to convert lignocellulosic biomass into these chemicals has still not been achieved. This study is related to the selective separation and conversion of birch wood C5 carbohydrates into furfural during the H₃PO₄–NaH₂PO₄-catalyzed hydrothermal pretreatment simultaneously preserving cellulose in the lignocellulosic leftover for glucose production by the enzymatic hydrolysis. The ratio of H₃PO₄–NaH₂PO₄ in the catalyst solution was changed (3:0, 2:1, 1:1, and 1:2). Results show that around 64.1 to 75.9% of available C5 carbohydrates were converted into furfural. The results of birch wood lignocellulosic leftover chemical composition analysis show that cellulose losses during the pretreatment stage did not reach more than 10% of the initial amount. Based on the enzymatic hydrolysis screening experiments, a suitable catalyst for pretreatment was selected and an in-depth study was carried out. Enzymatic hydrolysis experiments were organized based on the three-factor central composite face-centered design. The variable parameters were treatment time (24–72 h), enzyme load (10–20 U/g cellulose), and substrate amount in reaction media (10–20%). At optimal conditions, 49.9 ± 0.5% of available cellulose in lignocellulosic leftover was converted into glucose.

1. Introduction

Today, the European Union (EU) recognizes the crucial role of biobased materials in transitioning from a fossil-based to a biobased economy [1]. Therefore, it is vital to foster a circular and decarbonized economy. The world is currently confronting a range of global sustainability challenges, necessitating the development of advanced materials and technologies for biobased products [1,2,3]. Latvia’s economy relies significantly on a knowledge-intensive biobased approach, which aligns with Europe’s emphasis on circular economy. However, there is still untapped potential in leveraging bio-resources for value-added sectors, requiring the exploration of innovative approaches.

A viable and sustainable feedstock for the creation of numerous value-added products has arisen in the form of lignocellulosic biomass [4]. As a renewable resource, lignocellulosic biomass differs from conventional fossil-based feedstock in a number of ways, including quantity, accessibility, and the potential to lower greenhouse gas emissions [5]. However, there is a disadvantage to effectively using such kind of biomass in value-added products: lignocellulosic biomass has a rigid structure because of how its three primary components—cellulose, hemicellulose, and lignin—are arranged and interact [6]. Common sources of lignocellulosic materials widely used as feedstocks for the production of valuable biobased products due to their abundance and renewable nature include agricultural residues (sugarcane bagasse, straw, corn cobs, etc.), grass and bamboo, and wood and its processing residues. These biomass resources have a wide range of cellulose, hemicellulose, and lignin content, composition, and structure, which results in completely different qualities.

In Latvia, birch wood can be considered a promising feedstock for biorefinery concepts among the available lignocellulosic biomass due its high content of carbohydrates [7]. Moreover, according to data from the State Forest Service, an analysis of the past decade reveals a consistent trend—the available volume of standing birch wood in Latvia’s forest has remained relatively stable at approximately 159 million m³ and is widespread in the country. This stability is evidence of the careful and efficient management of this resource. In Latvia, the predominant utilization of birch wood is plywood manufacturing. With a firmly entrenched birch plywood industry, the nation has risen to prominence as a world leader in this field. Notably, this industry demonstrates slightly growth, consistently expanding its production capacity year after year. During the production of birch plywood, almost two-thirds of the original birch wood stem remains. This waste stream is chipped and sold as a waste product with low value. For example, the largest local plywood producer “Latvijas finieris” annually produces around 200,000 m³. Therefore, it can be expected that birch wood chips are locally available, further offering a promising avenue for fostering economic diversity within the local community if the right way to utilize them is discovered.

The major component of birch wood chips, and all lignocellulosic biomass, is cellulose. It is protected by a barrier formed by the tightly packed and tangled arrangement of hemicellulose and lignin, making it less accessible to enzymes or chemicals that may break it down into glucose for further processing [8]. The heterogeneous polymer hemicellulose surrounds and interacts with cellulose microfibrils by containing different sugar units such as xylose, mannose, arabinose, and galactose [9]. Additionally, it has acetyl groups, which are identified as inhibitory components for microbiological processing [8,9]. Hemicellulose serves as a cross-linking matrix, giving the overall structure flexibility and the ability to transport water [8,9,10,11]. Plant cell walls receive structural support and defense from the aromatic polymer lignin. Due to its intricate three-dimensional structure, lignin is extremely resistant to degradation and challenging to degrade using standard techniques [12]. Therefore, access to cellulose and hemicellulose is limited. Cellulose itself, being the main load-bearing component, forms long, linear chains of glucose molecules linked by β-1,4-glycosidic bonds. These chains are highly organized and form a crystalline structure through hydrogen bonds between adjacent cellulose chains. Such an arrangement leads to the formation of cellulose microfibrils, which are stiff, insoluble, and resistant to enzymatic degradation [13]. Effective pretreatment techniques, enzyme cocktails customized to certain biomass types, genetic engineering of microbes or plants, and process optimization are required to overcome these difficulties associated with birch wood breakdown [14].

According to reviews in the literature, pretreatment and enzymatic hydrolysis are crucial processes that need to be improved for the conversion of lignocellulosic biomass into useful biobased compounds [15]. The following are the major objectives of an efficient pretreatment procedure for the development of biobased products from lignocellulosic biomass: (1) the production of glucose either directly or indirectly through hydrolysis, (2) the prevention of loss and/or degradation of glucose produced, (3) the restriction of inhibitory product creation, (4) the reduction of energy requirements, and (5) the minimization of expenses [16,17].

The lignocellulosic biomass conversion rate into bioproducts can be increased by the appropriate pretreatment stage. Nowadays, it has been common practice to degrade lignocellulosic biomass into valuable products using chemicals such as acids, alkalis, organic solvents, and ionic liquids. Chemical pretreatment is commonly used in industry due to its simple operation and high efficiency. However, the production of environmental pollution and unwanted byproducts are common disadvantages of chemical pretreatment. Therefore, from the viewpoint of the lignocellulosic biorefinery concept, the main focus is how to divide efficiently the hemicellulose and cellulose from lignocellulosic biomass while avoiding any loss of carbohydrates or formation of inhibitors.

In this investigation, we used H₃PO₄/NaH₂PO₄ in different ratios during the catalyzed hydrothermal pretreatment. Such a combination offers the potential to achieve a high rate of C5 carbohydrate conversion into furfural, acetyl groups into acetic acid, and cellulose into lignocellulosic residue, according to our prior study [18]. There are various significant benefits and relevance to producing these two biobased compounds as co-products during the processing of lignocellulosic biomass. Furfural and acetic acid production during pretreatment can help reduce the inhibitory effects of these compounds on downstream processes because both of them have been reported to inhibit microbial growth and enzyme activity during subsequent enzyme reactions. This is because furfural and acetic acid are versatile platform chemicals with numerous applications that can enhance overall process economics [19,20,21].

By the selective separation of C5 carbohydrates and acetyl groups during the hydrothermal pretreatment, we suggest that such a pretreatment will enhance enzyme accessibility to the cellulose of lignocellulosic biomass compared to using phosphoric acid as the catalyst alone. The objective of this study was to determine an optimal ratio of H₃PO₄ and NaH₂PO₄ in a catalyst solution that enables the competitive production of furfural and acetic acid compared to today’s conventional processes prior to enzymatic hydrolysis, while simultaneously achieving a high conversion rate of cellulose into glucose. To identify a balance where efficient production of furfural and acetic acid is achieved without compromising the overall conversion of cellulose to glucose, the response surface methodology was used.

2. Materials and Methods

2.1. Preparation of Feedstock

Byproduct cellulose-grade birch wood chips from a local plywood production plant (JSC Latvijas Finieris-Lignums, Riga, Latvia) were used as a raw material in this study. The bark content was less than 2 wt.%. The moisture content of the collected wood chips was 49 wt.%. To avoid unnecessary wood degradation during storage, the wood chips were air-dried to reduce the moisture content below 15 wt.%. After that, the chips were fractionated using the Muototerä classifier MT300, according to the SCAN-CM 40:01 standard, with the following sieve arrangement: oversize chips (45 mm holes), over thick chips (8 mm slots), large accept chips (13 mm holes), small accept chips (7 mm holes), pin chips (3 mm holes), fines. Wood chips were shaken for 10 min ± 10 s. A fraction of Ø13–7 mm was used for further experiments as a feedstock.

2.2. Characterization of Feedstock

To better understand the impact of the used treatment methods’ effectiveness on the carbohydrate’s conversion into bioproducts, the chemical composition of untreated birch wood chips was analyzed. Before the determination of carbohydrates in the untreated birch wood chips, the quantification of extractives was performed according to the TAPPI 204 cm-07 standard. The extraction process was organized in the Knöfler-Böhm extractors. The used solvent was ethanol–benzene in the ratio 1:2. The extraction time was 6 h. The structural carbohydrates (glucose, xylose, galactose, arabinose, mannose) from birch wood samples were obtained using the two-step sulfuric acid hydrolysis method, in accordance with the National Renewable Energy Laboratory (NREL) standard TP-510-42618. The concentrations of saccharides were determined by HPLC using a Shimadzu LC20AD liquid chromatograph equipped with an RI detector (Shimadzu RID 10A) and a Thermo Scientific HyperREZ XP Carbohydrates Pb²⁺. The oven temperature was 70 °C and Milli-Q water was used as a mobile phase under a flow rate of 0.6 mL/min. Analysis time was 35 min. Before performing HPLC analyses, barium carbonate was used to neutralize sulfuric acid. The concentration of hydrolysis process byproducts, such as formic acid, acetic acid, levulinic acid, 5-hydroxymethylfurfural, and furfural were also analyzed by the same HPLC without hydrolysate neutralization. In this case, the Shodex Sugar SH-1821 column was used at 50 °C, with 0.005 M H₂SO₄ as the eluent, and a flow rate of 0.6 mL/min. Analysis time was 60 min. Analytical standards (cellobiose (≥99%), D(+)-glucose (≥99%), D(+)-xylose (≥99%), D(+)-galactose (≥99%), L(+)-arabinose (≥99%), D(+)-mannose (≥99%), barium carbonate (99–101%), formic acid (≥95%), glacial acetic acid (≥99%), levulinic acid (≥98%), 5-hydroxymethylfurfural (≥99%), furfural (≥99%), and sulfuric acid (95–97%)) were purchased from Merck (Darmstadt, Germany) and used without further purification. Acid-insoluble lignin was determined according to the NREL TP-510-42618 standard. Ash content was determined according to the NREL TP-510-42622 standard.

2.3. Hydrothermal Pretreatment System and Experimental Procedure

Before the hydrothermal pretreatment process, the fractionated wood chips were moisturized up to the moisture content of 35 ± 2 wt.%. After that, the catalyst solution was sprayed on the fractionated wood chips surface in the specially constructed paddle mixer. Four different H₃PO₄–NaH₂PO₄ ratios (3:0, 2:1; 1:1 and 1:2) in the catalyst mixture were studied. The total amount of catalyst was 3 wt.%, calculated based on oven-dried wood chips. Prepared wood chips (approximately 2500 g of oven-dried material) were loaded into the reactor of a homemade bench-scale reactor system that allows modulation of the industrial furfural production process. The reactor is equipped with a steam jacket and corresponding automatic control system to maintain a constant temperature throughout the experiment. The reactor was previously preheated to 175 °C. The material was treated in a continuous steam flow for 90 min. The steam that passed through the reactor was cooled down to room temperature and collected. Obtained hydrolysate sample was filtered through a syringe nylon membrane filter (pore size < 0.22 μm) into vials and analyzed by HPLC using a Shimadzu LC20AD liquid chromatograph equipped with an RI detector (Shimadzu RID 10A) and a Shodex Sugar SH-1821 column. HPLC procedure is described in Section 2.2. At the end of the hydrothermal pretreatment process, the treated birch wood chips were discharged from the reactor and weighed. All of the hydrolysis experiments were performed in duplicate under the same conditions, and their average values from the oven-dried mass were reported (variation < 0.05 wt.%, calculated on oven-dried mass).

2.4. Enzymatic Hydrolysis and Experimental Design

The previously obtained birch wood lignocellulosic leftover from hydrothermal pretreatment was dried to an ambient moisture content and ground in a Retsch GmbH SM100 cutting mill for chemical compositional analysis and enzymatic hydrolysis experiments. The used sieve was 0.75 mm. Carbohydrates, acetyl groups and acid-insoluble lignin in the lignocellulosic leftover were analyzed according to the NREL TP-510-42618 standard before enzymatic hydrolysis. HPLC procedures are described in Section 2.2.

Enzymatic hydrolysis was performed in two stages. Cellic^® CTec3 HS (Novozymes A/S, Bagsvaerd, Denmark) enzyme complex was used as a catalyst in all experiments. In the first stage, enzymatic hydrolysis was performed for all obtained biomasses to select the most suitable for further work. Treatment was organized in the Biosan ES-20/80 orbital shaker–incubator at 150 rpm and 50 °C in 0.2 M sodium citrate buffer at initial pH of 5.2. The pH of the medium was controlled and adjusted with a 5 M NaOH solution. The range of pH was kept within 4.9 +/− 0.2. The lignocellulosic leftover load was 15% (w/v). Treatment time was changed from 24 to 96 h. The amount of enzyme complex used was 15 U/g cellulose in the lignocellulosic leftover. Measurement of the enzyme activity was performed according to the method of Denault et al. [22] in combination with the method of Dygert et al. [23]. The calculated enzyme activity was 253 U/mL. The calibration curve for glucose determination by copper (II)–neocuproine method was in the range of 0.01 to 0.1 g/L. Glucose was determined by the Shimadzu LC20AD HPLC using the Shodex Sugar SH1821 column, with the same parameters that were described in Section 2.2. After the selection of appropriate birch wood lignocellulosic leftover optimal enzymatic hydrolysis parameters were detected based on a central composite face-centered design (CCF). The experimental design was prepared by the Design Expert v13 software. The treatment time, enzyme load and solid load/buffer ratio were variable parameters (see

Table 1. The parameter levels used in the CCF experimental design.

Parameter	Symbol	Factor Level
Treatment time (TT)	h	24	48	72
Enzyme load (EL)	U/g cellulose	10	15	20
Solid load/buffer ratio (S/B)	%	10	15	20

). A total of 16 experimental trials were performed. Two of them were center-point replicates.

3. Results and Discussion

3.1. Hydrothermal Pretreatment

To remove the C5 carbohydrates from birch wood chips and convert them into furfural, the phosphoric acid (H₃PO₄) and sodium dihydrogen phosphate (NaH₂PO₄) catalyzed hydrothermal pretreatment process was performed. Four different ratios of H₃PO₄/NaH₂PO₄ catalyst (3:0, 2:1, 1:1, and 1:2) were used to catalyze this process. The total amount of catalyst was 3 wt.%, calculated based on oven-dried birch wood chips. The treatment temperature was 175 °C and the treatment time was 90 min. Another constant parameter in this stage was birch wood initial moisture (35 wt.%). The yield of the main products (furfural and acetic acid) after the hydrothermal pretreatment stage is summarized in Figure 1. Low yields of formic acid (0.51 ± 0.02%, calculated on oven-dried mass), levulinic acid (0.51 ± 0.02%, calculated on oven-dried mass), and 5HMF (0.02 ± 0.00%, calculated on oven-dried mass) were also produced.

As can be observed in Figure 1, the conversion efficiency of C5 carbohydrates into furfural increased with the increase in H₃PO₄ amount in the catalyst mixture. These results indicate that the effectiveness of our used hydrothermal pretreatment is from 64.1 to 75.9% of the theoretically possible amount of furfural. This means that, in practice, during the biomass pretreatment stage, the conversion process of C5 carbohydrates into furfural can achieve a substantial fraction of the maximum potential yield. Obtained results are in accordance with previously reported results [18].

3.2. Effect of Hydrothermal Pretreatment on the Composition of Birch Wood

To understand the impact of the used hydrothermal pretreatment effectiveness on the C5 carbohydrate’s conversion into furfural, the chemical compositions of untreated and pretreated birch wood chips were analyzed. The chemical composition of the birch wood is summarized in

Table 2. Chemical composition changes after hydrothermal pretreatment, % based on oven-dried mass.

Components	Untreated	H3PO4/NaH2PO4 (1:2)	H3PO4/NaH2PO4 (1:1)	H3PO4/NaH2PO4 (2:1)	H3PO4
Extractives *	4.2 ± 0.2	n.d.	n.d.	n.d.	n.d.
Glucan	38.7 ± 0.5	48.0 ± 0.5	48.2 ± 0.3	48.5 ± 0.6	48.2 ± 0.1
Xylan	18.6 ± 0.4	5.8 ± 0.0	4.4 ± 0.2	3.9 ± 0.1	2.9 ± 0.0
Arabinan	0.5 ± 0.0	0.1 ± 0.0	0.5 ± 0.1	0.6 ± 0.0	0.4 ± 0.0
Galactan	1.0 ± 0.1	0.8 ± 0.0	1.1 ± 0.1	1.3 ± 0.1	1.9 ± 0.1
Mannan	0.9 ± 0.1	0.7 ± 0.0	0.7 ± 0.0	0.6 ± 0.0	0.6 ± 0.0
Acetyl groups	4.6 ± 0.0	0.5 ± 0.0	0.4 ± 0.1	0.2 ± 0.0	0.1 ± 0.0
Acid-insoluble residue	19.6 ± 0.4	39.7 ± 0.2	40.3 ± 0.1	40.2 ± 0.3	41.8 ± 0.3

* Ethanol-benzol (1:2) soluble; n.d.—not detected.

.

Chemical composition clearly illustrates that cellulose (expressed as glucan) and lignin (expressed as acid-insoluble residue) comprise the predominant parts in the birch wood lignocellulosic leftovers. A comparison of the chemical composition of untreated and treated biomasses clearly shows that the largest part of xylan and almost all acetyl groups are removed from birch wood during hydrothermal pretreatment. As a result, the proportion of cellulose and lignin increased in the obtained birch wood lignocellulosic residue. The content of cellulose was similar in all obtained samples but lignin slightly increased with the increase in H₃PO₄ proportion in the catalyst mixture. This allows us to conclude that more cellulose was irreversibly degraded. Recalculating it to the initial amount, cellulose degradation does not exceed 4–10%, respectively. On the other hand, the high lignin content (around 40% of total mass) could cause an impact on the cellulose conversion process, as it can cause the formation of phenolic compounds that may reduce the diffusion efficiency of enzymes in the birch wood lignocellulosic leftovers.

Low amounts of other carbohydrates (arabinan, galactan and mannan) were also detected in birch wood lignocellulosic leftovers. The total amount of them was in the range of 1.6–2.9%, calculated on oven-dried mass. Therefore, their changes will not be investigated in further study.

3.3. Preliminary Study of Enzymatic Hydrolysis

The main goal of the hydrothermal pretreatment process was to modify the structure of the birch wood in order to increase the digestibility of cellulose. The effectiveness of the hydrothermal pretreatment was evaluated through preliminary enzymatic hydrolysis tests. The catalyst used in the hydrothermal pretreatment stage was not washed from the substrates. Based on previous research data [18], an enzyme load of 15 U/g cellulose and a substrate load of 15% (g/v) were selected as constant parameters in these tests. Considering that almost all C5 carbohydrates were cleaved during the hydrothermal pretreatment, the main focus of these experiments was on the cellulose conversion rate into glucose. Obtained data of cellulose conversion into glucose are shown in Figure 2.

As can be seen in Figure 2, the used catalyst in the hydrothermal pretreatment stage has an impact on the enzymatic hydrolysis effectiveness. Examining the impact of the pretreatment stage on cellulose conversion into glucose through enzymatic hydrolysis reveals a distinct correlation between the catalyst used and the resulting conversion rate. The lowest cellulose conversion rate into glucose was obtained from the substrate pretreated with H₃PO₄. The yield of glucose slightly increased by replacing H₃PO₄ with the NaH₂PO₄. The highest conversion rate was achieved when the birch wood was treated with H₃PO₄/NaH₂PO₄ in a ratio of 1:2 at all studied enzymatic hydrolysis times. This allows us to conclude that H₃PO₄ has an inhibitory effect on enzyme cocktail activity. Unfortunately, the obtained yield of furfural (Figure 1) at this catalyst mixture was the lowest. Therefore, the substrate obtained by the hydrothermal pretreatment using H₃PO₄/NaH₂PO₄ in a ratio of 1:1 as a catalyst is selected for enzymatic hydrolysis process optimization.

Looking at the results from the viewpoint of treatment time, revealed an interesting finding, indicating that extending the treatment time from 72 to 96 h does not lead to a substantial improvement in glucose yield. This suggests that there is a diminishing return in terms of enzymatic hydrolysis efficiency beyond a certain treatment time. Based on these results, it can be concluded that the optimal treatment time for achieving higher glucose yield is in the range of 48 to 72 h. Based on this finding, enzymatic hydrolysis time of 24 to 72 h will be used as boundaries for process optimization.

3.4. Optimization of Enzymatic Hydrolysis

For the optimization of the enzymatic hydrolysis, an experimental face-centered central composite (CCF) design of the response surface was used, with two points at the center. Three factors were considered—treatment time (24–72 h), enzyme load (10–20 U/g cellulose in lignocellulosic residue), and solid/buffer ratio (10–20% (g/v)). Glucose yield, calculated on the theoretically possible amount (% of the t.p.y.), was chosen as the response. The shaking speed (150 rpm) and treatment temperature (50 °C) were kept constant in all experiments. Design-Expert 13 (Stat-Ease, Minneapolis, MI, USA) was used for the generation of CCF experimental design, optimization, and analysis of the design data. The obtained yield of glucose according to the CCF design experimental runs are shown in

Table 3. Experimental data of glucose yield after enzymatic hydrolysis according to the CCF design.

Run	Treatment Time	Enzyme Load	Solid/Buffer Ratio	Glucose
	(TT)	(EA)	(S/B)
	h	U/g Cellulose	%	% of the t.p.y.
1	24	20	20	42.27
2	72	10	10	32.80
3	72	20	20	43.72
4	48	15	15	39.50
5	24	20	10	50.69
6	48	15	20	33.48
7	24	10	10	23.48
8	48	15	10	45.62
9	72	15	15	38.94
10	48	10	15	23.30
11	24	15	15	35.88
12	48	20	15	51.21
13	24	10	20	23.71
14	72	20	10	60.86
15	72	10	20	21.21
16	48	15	15	38.86

.

The effects of the variable factors and the value of the coefficients estimated in the model were determined from the Analysis of Variance (ANOVA). The empirical relationship between conversion efficiency yields of cellulose into glucose (G) and the three variables in actual values are shown in a quadratic regression equation (Equation (1)) that fitted the data after the removal of the insignificant terms. For the three response variables, the high value of the coefficient of determination (R²adjusted = 0.9862) indicates that the adjusted model assertively predicts the result for the yield of glucose. This means that the relationship between the experimental factors and the glucose yield was adequately described.

$$G\left(\mathrm{\%}oft.p.y.\right)=-23.54+0.54A+4.98B-0.2C+0.005AB-0.02AC-0.07BC-0.002A^2-0.06B^2+0.04C^2$$

(1)

As expected, the cellulose conversion rate into glucose increases with the increase in treatment time and enzyme load, at the same time reducing the amount of previously pretreated biomass in the reaction zone in the working range (Figure 3 and Figure 4). At a substrate amount of more than 15% (wt./v), the effect of treatment time on glucose yield is low. Moreover, at a substrate amount of 20% (wt./v), it slightly decreases with the increase in treatment time to more than 48 h. Similar observations were made in Bittecourt et al. study [24]. In our case, this reduction can likely be attributed to the attainment of the critical lignin mass within the reaction media that acts as a physical barrier to hinder the enzymatic degradation of cellulose. This leads to the fact that the diffusion of enzymes into the lignocellulosic structure to divide glucose molecules from the cellulose structure is much slower. Djajadi et al. [25] showed that enzymes from a commercial cellulolytic mixture can be adsorbed onto lignin-rich residues isolated from hydrothermally pretreated grass biomass. In addition, the pseudo-lignin that can be formed at a high severity level of hydrothermal pretreatment can block the productive binding of cellulase and cellulose [26,27]. The lignin content in our substrate was 41% and can be considered as a lignin-rich biomass. Therefore, this points to the future need to better understand the influence of our hydrothermal pretreatment on lignin degradation, pseudo-lignin formation, and its impact on enzymatic hydrolysis to improve substrate load and glucose yield.

Reducing the substrate amount under 15% (wt./v), specifically at the lowest studied ratio of 10% (wt./v), the conversion rate of cellulose into glucose demonstrates a notable enhancement as the treatment time and enzyme load are increased. In fact, this increase leads to a cellulose conversion rate exceeding 60%. Similar trends can be observed in previously reported studies [18].

The obtained model also demonstrates a consistent and almost linear relationship between enzyme load and solid/buffer ratio within the all working range of treatment time (Figure 4). This indicates that the interaction of these variables results in a proportional increase in glucose yield when the enzyme load is raised and the solid load is reduced. This finding suggests that changing these two factors within the studied ranges can have a predictable and proportional impact on the overall outcome of the process.

From an economic point of view, it is necessary to work with the highest possible amount of substrate at the lowest possible amount of enzyme load in the shortest possible time. Therefore, based on the obtained mathematical model the established optimal conditions for glucose production were treatment time of 70 h, enzyme load of 17 U/g cellulose, and solid/buffer ratio of 14% (wt./v). Under these conditions, the model predicted an overall cellulose conversion rate into glucose of 52.3 ± 1.2%. To determine the validity of the predicted value, new enzymatic hydrolysis experiments (in triplicate) were conducted at the optimal conditions leading to the overall conversion rate of 49.9 ± 0.5%. These data were within the confidence interval of the predicted values. Therefore, the obtained model is considered adequate.

4. Conclusions

Through the selective separation of C5 carbohydrates and acetyl groups during the hydrothermal pretreatment, our findings suggest that this particular pretreatment method can enhance enzyme accessibility to the cellulose within lignocellulosic biomass. In comparison to using phosphoric acid as the sole catalyst, the proposed approach demonstrates improved effectiveness in improving enzyme access to the cellulose of lignocellulosic leftovers. The enzymatic hydrolysis shows the limitations of high solid loads. This points to the need to better understand the influence of our hydrothermal pretreatment on lignin degradation, pseudo-lignin formation, and enzymatic hydrolysis to improve substrate load and glucose yield.