Next Article in Journal
Protein Engineering of Pasteurella multocida α2,3-Sialyltransferase with Reduced α2,3-Sialidase Activity and Application in Synthesis of 3′-Sialyllactose
Next Article in Special Issue
New Trends in the Use of Catalysts for Biofuel and Bioproduct Generation
Previous Article in Journal
Synthetic and DFT Modeling Studies on Suzuki–Miyaura Reactions of 4,5-Dibromo-2-methylpyridazin-3(2H)-one with Ferrocene Boronates, Accompanied by Hydrodebromination and a Novel Bridge-Forming Annulation In Vitro Cytotoxic Activity of the Ferrocenyl–Pyridazinone Products
Previous Article in Special Issue
Single-Atom Catalysts: A Review of Synthesis Strategies and Their Potential for Biofuel Production
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Catalysts
Volume 12
Issue 6
10.3390/catal12060577
catalysts-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Coupled Biohydrogen Production and Bio-Nanocatalysis for Dual Energy from Cellulose: Towards Cellulosic Waste Up-Conversion into Biofuels

by
Jaime Gomez-Bolivar
1,*,†,
Rafael L. Orozco
2,3,†,
Alan J. Stephen
2,3,
Iryna P. Mikheenko
2,
Gary A. Leeke
3,
Mohamed L. Merroun
1 and
Lynne E. Macaskie
2
1
Department of Microbiology, Faculty of Sciences, University of Granada, Campus Fuentenueva, 18071 Granada, Spain
2
Schools of Biosciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
3
Chemical Engineering, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
*
Author to whom correspondence should be addressed.
†
These authors contributed equally to this work.
Catalysts 2022, 12(6), 577; https://doi.org/10.3390/catal12060577
Submission received: 6 April 2022 / Revised: 16 May 2022 / Accepted: 19 May 2022 / Published: 24 May 2022
(This article belongs to the Special Issue New Trends in the Use of Catalysts for Biofuel and Bioproduct Generation)
Browse Figures
Review Reports Versions Notes

Abstract

:
Hydrogen, an emergent alternative energy vector to fossil fuels, can be produced sustainably by fermentation of cellulose following hydrolysis. Fermentation feedstock was produced hydrolytically using hot compressed water. The addition of CO2 enhanced hydrolysis by ~26% between 240 and 260 °C with comparable hydrolysis products as obtained under N2 but at a 10 °C lower temperature. Co-production of inhibitory 5-hydromethyl furfural was mitigated via activated carbon sorption, facilitating fermentative biohydrogen production from the hydrolysate by Escherichia coli. Post-fermentation E. coli cells were recycled to biomanufacture supported Pd/Ru nanocatalyst to up-convert liquid-extracted 5-HMF to 2,5-dimethyl furan, a precursor of ‘drop in’ liquid fuel, in a one-pot reaction. This side stream up-valorisation mitigates against the high ‘parasitic’ energy demand of cellulose bioenergy, potentially increasing process viability via the coupled generation of two biofuels. This is discussed with respect to example data obtained via a hydrogen biotechnology with catalytic side stream up-conversion from cellulose feedstock.

1. Introduction

More than 80% of global energy demand is still met via fossil fuels [1]. Hydrogen, emergent for clean transport [2,3], is also widely used in the bulk and fine chemicals sectors, with between 6 and 15% growth annually [4]. Hydrogen production is currently mainly from fossil resources. For some ‘clean’ applications (e.g., in proton exchange membrane fuel cells (FCs)), petrochemical-H2 requires purification, whereas ‘clean’ H2 produced via fermentation of sugary and waste feedstocks (bio-H2) can supply a fuel cell directly [5].
Cellulose is one of the most abundant polysaccharides on earth. For efficient utilisation as a source of biofuels, biogases and chemicals its breakdown to smaller monomeric units (glucose) is required through hydrolysis. Cellulose comprises β-(1,4)-linked glucosyl units in water-insoluble crystalline structures which are hard to hydrolyse; hence, the enzymatic breakdown of cellulose to fermentable sugars is slow [6]. It can be accelerated by using ‘designer’ enzymes, but this requires a high enzyme dosage while enzymatic hydrolysis remains slow [7,8]. A novel protein acted synergistically with cellulases [9]; however, protein purification is costly.
More traditional approaches, e.g., acid hydrolysis, yield glucose at up to 100%, but drawbacks include the corrosion of equipment, a requirement for H2SO4 and product recovery from acidic residues [10,11]. A clean alternative, ohmic heating (OH), has been used for hydrolysis (see Supplementary Materials); the advantages include rapid heating and instant shutdown, but disadvantages include ‘hot’ and ‘cold’ spots (problematic at scale), with the need to adjust the set-up according to the conductivity of the target liquid.
A more benign approach uses hot compressed water (HCW) at sub-critical temperatures (180 to 300 °C), which enables the solvolysis of cellulose and other complex polysaccharides, such as hemicelluloses and starch [12,13,14]. For example, Minowa et al. [15] used HCW (200 to 350 °C) with short reaction times using Ni2+ or alkali (Na2CO3) catalysts. With the former, cellulose decomposition produced little glucose; moreover, a Ni-contaminated hydrolysate would give problems of final waste disposal. Carbonate-catalysis gave a maximum glucose yield at 260 °C; later work confirmed that CO2 accelerated polysaccharide hydrolysis [16]. Additionally, the diversity of chemical composition of (ligno)cellulosic feedstock itself can have a significant effect on the rate of reaction for other innovative processes, such as aqueous phase (AP) reforming [17]. Modifications in reaction temperatures in AP reforming also favoured hydrogen production when using glucose and xylose as representative compounds of hemicellulose [18]. In the case of HCW, decreasing temperature (to 2 °C/min) increased the glucose yield [13], although toxic degradation by-products were also favoured (e.g., 5-hydroxymethyl furfural: 5-HMF). Despite this, HCW remains the potentially most cost-effective hydrolysis pre-treatment for cellulosic biomass, with the lowest environmental impact [12,13,14,19].
For the onward fermentation of cellulose hydrolysate, 5-HMF, a potent metabolic inhibitor [20], must be removed, e.g., via use of activated carbon (AC). The efficacy of this was shown in the selective removal of degradation products from starch hydrolysate [21,22,23,24]. Hence, AC was initially used in this study to evaluate bio-H2 synthesis from native and ‘cleaned’ cellulose hydrolysates.
However, the AC method may not be convenient for 5-HMF side stream valorisation e.g., to 2,5 dimethylfuran (DMF). Here, an effective segregation of 5-HMF from the ‘raw’ hydrolysate is required as the first step towards biofuel process integration. A liquid/liquid extraction method for 5-HMF extraction was developed (permitting downstream direct addition of catalyst for a one-pot system) taking advantage of the favorable partition coefficient of 5-HMF [25]. Effective solvent extraction of 5-HMF from cellulose hydrolysate into methyl tetrahydrofuran (MTHF; a solvent which can itself be derived from biomass hydrolysate) has been reported previously [26]. MTHF is advantageous due to its low solubility in water while exhibiting similar properties to other widely used alternatives (e.g., tetrahydrofuran; THF) in several organometallic-catalysed reactions [25,26,27].
Working towards a positive energy balance, obtained from the catalytic hydrogenation of 5-HMF, can be used as a ‘drop in’ fuel for conventional engines, containing comparable energy to gasoline [28] (energy contents of DMF and gasoline are 31.5 MJ/L and 35 MJ/L, respectively) and bringing the potential to power a local generator, allowing the recycling of energy into the process.
5-HMF, the source of DMF, arises from the dehydration of fructose (and to a lesser extent glucose) catalysed by mineral acids and metal catalysts [29,30]. Biomass contains a significant amount of iron as well as trace transition metals. Hence, the co-production of 5-HMF during hydrolysis of plant cellulose is expected. This, if managed correctly, could potentially deliver dual carbon-neutral energy (i.e., the fermentative bio-H2 stream (see later) plus the derived DMF component) from cellulosic wastes.
For up-valorisation of the 5-HMF side stream into DMF, a metallic catalyst can be biomanufactured as an alternative to costly commercial catalysts. A one-step conversion (94%) of 5-HMF to DMF using 5 wt% palladium on a carbon commercial catalyst was shown, with 33% yield [31], whereas a comparably active supported Pd-catalyst was made on ‘second life’ bacterial cells [32,33,34] (fermentation bio-waste) [32]. The high cost of catalytic metal was mitigated by metal biorefining from wastes [33,34], which was shown as economically feasible by an example Life Cycle Analysis [35].
With respect to the fermentation as a source of waste bacteria, various microbial systems have been used for bio-H2 production [36]. E. coli, while not necessarily the best for industrial application, has well-defined molecular genetic tools to optimise metabolic pathways. It is important to note that while the disposal of genetically modified organisms is problematic, the metallisation process kills the cells. E. coli cells left over from the biohydrogen fermentation (below) were used (‘second life’) to make a supported ‘bio-Pd’ catalyst, which had comparable activity to commercial comparators [32,37]. Later work showed that a bio-Pd/Ru bimetallic catalyst was active in 5-HMF up-conversion to DMF using Bacillus benzeovorans [38] and Desulfovibrio desulfuricans [27]. Bio-Pd/Ru was used in this study using D. desulfuricans in comparison to post-fermentation E. coli cells for catalyst fabrication, following the liquid–liquid extraction for ‘one-pot’ 5-HMF separation from the fermentation liquor and its catalytic upgrading.
The potential viability of a biorefinery approach for a dual product stream is limited by hydrogen production. For the E. coli dark fermentation (DF) this is (via an irreversible reaction) maximally 2 mols H2 per mol glucose (see [39,40]), which maintains sufficient partial pressure of H2 to drive a fuel cell directly [5]. Higher H2 yields can be achieved (but via a reversible reaction) using other microorganisms. However, this self-limits the H2-partial pressure (discussed in [41]). As an alternative, a secondary photo-fermentation (PF) of the aqueous DF stream increases the hydrogen yield by several-fold. The PF converts the organic acids (OA) fermentation co-products of E. coli into a second biohydrogen stream [39,40,41]. This method employs the electrodialytic extraction of OAs into a coupled DF/PF ‘biohydrogen from waste’ system, which outperformed other sustainable technologies [39,40], but with no comparable energy balances were made. In this integrated system, the DF/PF would be further value-added by catalytically upgrading in parallel the hydrolysis side stream into a liquid biofuel precursor as we describe in the current study, the overall goal being to estimate the overall energy balance using obtained data from cellulose to underpin a future Life Cycle Analysis (LCA) and economic assessment, which would also factor-in the additional benefits of carbon mitigation and waste minimisation, which, along with circularity, do not feature in standard LCAs. The integration of fermentation (bio-H2) and parallel chemical catalysis (for DMF production) would also realise additional benefits, which include mitigation of waste disposal costs (i.e., used bacteria and the hydrolysis side-stream), while realising resources from waste, working towards process circularity, especially since fermentation waste CO2 has the potential to be recycled into catalysing cellulose hydrolysis upstream (see above). Onward assessments would also consider the value of the ‘saved’ carbon (EU emission trading value of carbon is $50/t June 2021; Source Carbonpricingdashboard.worldbank.org, accessed on 23 January 2022).
The hypothesis driving this study is that a positive energy balance can be achieved by the integration of the component chemical and biotechnologies and the specific objective is to prove or disprove this hypothesis.
The overall goal of this program is energy efficient waste conversion via integrated bio-H2 and 2,5 DMF generation, the aim of this study being to examine a model cellulose HCW hydrolysate suitable for, and utilised into, parallel downstream valorisations: (i) dark and photo-fermentations to produce bio-H2 (after 5-HMF removal) and (ii) catalytic up-conversion of harvested 5-HMF into 2,5 dimethylfuran (DMF). The ‘parasitic’ energy demand is mainly driven by the energy consumed in the HCW hydrolysis and in the hydrogenation reaction of 5-HMF into 2,5 DMF, underpinning the objective—to assess whether a positive energy balance is (or could become) feasible.

2. Results and Discussion

2.1. Hydrolysate Production, Detoxification, and Fermentation into Biohydrogen by E. coli

2.1.1. Cellulose Hydrolysis under N2 and CO2 Using Hot Compressed Water and Preparation of Hydrolysate

The thermal pathways from 220–260 °C (58 to 80 bar) are shown in Figure 1. The heating rate (~7 °C/min) exceeded the minimum 2 °C/min required for maximum sugar yields [13,42].
The conversion of cellulose (Xc, %) into products under N2 and CO2 is shown in Figure 2a,b, respectively, increasing exponentially with respect to temperature.
Substituting CO2 for N2 increased Xc, e.g., at 250 °C the conversion was 59.2 ± 12.8% and 73.3 ± 8.1% under N2 and CO2, respectively (mean ± SEM; 3 experiments). The use of CO2 shifted down the products and TOC profiles by 10 °C and were comparable (Figure 3) with respect to the production of glucose, fructose or 5-HMF, which was approximately equimolar with glucose. The low fructose (Figure 3b) is consistent with it being the precursor of 5-HMF. Some decomposition of products occurred above 260 °C under N2 and 250 °C under CO2 (Figure 3).
During hydrolysis, the pH of the hydrolysates fell from pH 6.3 to between 2.6 and 3.6. The solution chemistry of carbon dioxide in water and formation of carbonic acid (e.g., potentially giving some acid hydrolysis) are poorly understood under HCW reaction conditions; carbonic acid exists only transiently in solution [43]. The hydrolysate pH fell as low as pH 2.6 and would lead to different proportions of fully ionised/fully associated organic species over the conditions studied [44].
The glucose yield under CO2 was, maximally, 0.225 g C/g C initial cellulose, (at 250 °C) while that under N2 was 0.198 gC/gC initial cellulose at 260 °C, i.e., ~20% glucose yield in each case, which is ~half of the yield reported in similar experiments at 200 °C using starch [45], attributable to the recalcitrance of crystalline cellulose (see Introduction). The results shown in Figure 2 suggest that glucose contained in the hydrolysate would be suitable for fermentation provided that significant removal of toxic 5-HMF could be achieved.

2.1.2. Detoxification with AC

Glucose and organic acids were not removed significantly by AC treatment while, in contrast, more than 70% of 5-HMF co-product was removed from 37 mM (250 °C) and 35 mM (260 °C) 5-HMF produced under CO2 and N2, respectively (Figure 4). This is consistent with previous 5-HMF detoxification experiments with starch hydrolysates [45].

2.1.3. Growth of E. coli on Native and AC-Treated Hydrolysates, and Biohydrogen Production

AC-treated hydrolysates supported twice the growth of E. coli in comparison with the control with hydrolysates prepared at 240 °C and 250 °C (Figure 5a).
From the latter, untreated and treated samples gave 74% and 33% inhibition, respectively, indicating residual inhibitory degradation products (c.f. Figure 4). Growth was increased using AC by 1.5-fold using HCW cellulose hydrolysis under CO2 at 230 °C (Figure 5a) but samples made at 240 °C showed no benefit on growth (Figure 4) even though much of the 5-HMF had been removed (Figure 4). Inhibition was attributed to the low pH; undissociated formic acid (pKa of 3.75) is highly toxic as it enters cells, dissociates in the neutral cytoplasm and H+ kills the cells via low internal pH [46]. Against this, the hydrolysate was diluted 10-fold into the growth medium, which was buffered (pH 6.5 [45]). However, the extent of growth is relatively unimportant since bio-H2 is made by resting cells, while biomass overgrowth may be undesirable as this requires disposal or up-valorisation (see later).
No H2 was produced without AC treatment of cellulose hydrolysates made at temperatures above 230 °C and 220 °C with N2 and CO2, respectively (Figure 5b) attributed to inhibitory effects of co products of cellulose hydrolysis. In contrast, AC-treated hydrolysates supported H2 production (following removal of co-product 5-HMF).
Hydrogen yields increased in parallel to the hydrolysis temperature in these bottle tests to, maximally, 10–11 mL H2/g of cellulose from samples prepared at 250 and 260 °C in both cases (attributable to their higher glucose content), but with correspondingly ~31 to 37% lower yields as compared with those previously reported from starch hydrolysate made at 200 °C under CO2 [45]. This H2 yield is important, being key to the overall energy balance for process viability (see later).

2.2. Scope for Sustainable Biohydrogen Process for Bioenergy from Cellulose

2.2.1. Hydrogen Yield from the E. coli ‘Dark’ Fermentation

A simple calculation shows that biohydrogen production by E. coli, as shown here in the bottle tests, is not economically or energetically feasible for use with an example biomass such as Miscanthus: a production of 10–11 mL H2/g of cellulose (above)(i.e., 100–110 kJ H2/kg of cellulose) falls well short of the 184 kJ/kg required for biomass comminution alone, before factoring-in the lower yields obtained from real waste hydrolysate (above and Supplementary Materials Figure S2). The high energy density of liquid H2 (if liquefaction is the chosen path for hydrogen utilisation) comes at the expense of significant compression and cooling costs [47]. Clearly the H2 yield must be increased.
The simple ‘bottle’ fermentation tests in this study, yielding ~11 mL H2 per g of cellulose (Figure 4), used cells fermenting, unmixed, for 24 h. A detailed study reported elsewhere using a well-mixed fermenter [48] showed that, via re-assimilation and re-consumption of part of the acetate produced from glucose, the hydrogen yield was increased by 50% between 24 and 48 h. By factoring this improvement into the current study, a hydrogen yield of ~16 mL per g of cellulose would be anticipated, but the projected energy yield (~160 kJ H2/kg) still falls below the break-even point for the example comminution energy demand (see above).
Fermentative bio-H2 production by E. coli is limited biochemically to 2 mol H2 per mol of glucose (discussed in [41,49]). In addition to H2 and CO2, the fermentation products are ethanol and organic acids (OAs). An analysis of products was made for four strains of E. coli (R.L. Orozco and A.J. Stephen unpublished work) from the optimised 48-h fermentation [48], which takes into account the greater amount of hydrogen and a correspondingly lower acetate content due to its reassimilation and secondary fermentation into additional H2 [48]. The concentration of ethanol product was 0.9 mol/mol sugar (comparable to the production of carbon in the OAs), which is too low for recovery by distillation. However, other studies have shown that a bio-Au catalyst bioresourced from jewellery waste by E. coli oxidises alcohol to the corresponding OA (see [34]) and a bio-Au ‘trap’ would approximately double the acetate content (from ethanol) of the spent fermentation medium for electrodialytic extraction and downstream secondary photo-fermentation (see below). If not removed, ethanol build-up in the dark fermentation inhibits the cells, requiring a ‘fill and draw’ as opposed to a continuous process.
The optimised dark fermentation biohydrogen process [48,49] was evaluated in the current context; note that the H2 production shown in Table 1 is only ~18% of the maximum yield attainable from E. coli fermentation of hexose sugar (2 moles H2/mole sugar), which was attributed to sub-optimal conditions in the small bottle tests. In contrast, ~80% conversion from glucose (a four-fold increase) was reported using a commercial well-mixed fermenter with automated pH regulation and agitation [41,49]. Hence, the working yield adjusted to optimal conditions was taken as 1.6 mol H2/mol glucose for onward calculations.

2.2.2. Downstream Secondary Photo-Fermentation for a Coupled System

The ability of Rhodobacter sphaeroides, a purple-non-sulfur (PNS) bacterium, to generate bio-H2 from the organic acid mix (OA mix) produced by E. coli [49] was evaluated by photo-fermentation, as described in Section 3. The synthetic OA mix concentration was adjusted to 20 mM as follows: acetate 6.7; lactate 3.8; succinate 5.2; butyrate 4.4. The hydrogen produced by this test was 1973 ± 252 mL H2 (mean ± SD; 3 experiments).
A useful evaluation for a biohydrogen process is its hydrogen production potential (HPP) [50] defined as total hydrogen gas (mole) that can be produced per mole of OA. The HPP values are: acetate (2 carbons), 4; lactate (3 carbons), 6; butyrate (4 carbons) 10; succinate (4 carbons) 7 [41]. Hence, by taking the hydrogen yields obtained in this study in the above bottle photo-fermentation tests, the HPP of hydrolysates can be compared and the efficiency of the process calculated (HPP: potential mol hydrogen per mol of OA). This comparison is shown in Table 1 (below):
Table
Table 1. H2 generated in this study compared to HPP of the OA mix. STDEV with n = 3 (number of experiments).
Table 1. H2 generated in this study compared to HPP of the OA mix. STDEV with n = 3 (number of experiments).
Organic Acid (OA)mmol in OA MixH2 Produced by Mix (mol/mol OA Mix)HPP (mol H2/mol OA)HPP by OA Mix (mol)
Acetate6.70.081 ± 0.01140.027
Lactate3.8 60.023
Succinate5.2 70.036
Butyrate4.4 100.044
Total20.10.081 ± 0.011 0.130
From these values, the yield of H2 produced in this study was 4.05 mol H2/mol OA. The conversion efficiency of the photo-fermentation test (EP) is calculated by Equation (1):
EP = Hydrogen produced × PPmix × 100.
From Equation (1), 62.3% of the total theoretical maximum H2 production was achieved (under the conditions of the tests) giving good scope for process improvement through improvements in better controlled operation conditions (e.g., mixing, pH, culture density), light utilisation (which was calculated to be less than 2%: R.L. Orozco, unpublished work) and through better photobioreactor design and light ‘upgrading’. The latter offers significant potential for photosynthetic process improvement. Promising results were achieved in a paradigm algal system by using quantum dots (made from wastes [51]) to up-convert unused wavelengths of sunlight into usable light, doubling the photo-synthetic yield [52]. However, unlike algae, PNS bacteria absorb light maximally in the near infrared (NIR) part of the spectrum. The development of NIR-emitting quantum dots is a relatively new technology and currently uneconomic in this application. The use of low energy light emitting diodes (LEDs) emitting at photo-synthetically active wavelengths is now common in horticulture. As with use of quantum dots, the application of an LED array more than doubled the photo-synthetic productivity using the microalga Chlorella vulgaris (R.L. Orozco, unpublished), but was then limited by the transfer of CO2 to the cells; additional 2% CO2 enhanced the photo-productivity by approximately 7-fold compared to air (R.L.Orozco, unpublished). Since PNS bacteria use fixed carbon (OAs), this limitation will not apply in the coupled bio-hydrogen fermentation. Biogas from R. sphaeroides is virtually all hydrogen under these conditions.
H2 yields from photo-fermentation are significantly higher than from ‘dark’ fermentations. Hence, DF is more important as a supply of organic acids into the photo-fermentation than for its bio-H2 production per se. Previous work (R.L. Orozco and A.J. Stephen, unpublished) showed that the H2 yield in the photo-fermentation was independent of the organic acid proportions in the feed from the mixed-acid dark fermentation. Hence, any source of organic acids could be potentially used from a dual system or, indeed, in a stand-alone photo-fermentation (see later discussion), anticipating similar H2 yields after photo-fermentation as those found in this study. This is important as the organic acid mix from other high producing E. coli strains (e.g., FTD67, FTD89, RL009, MC4100) can also be used as a source of OAs which can be efficiently transferred to the photo-fermentation system via electro-separation [39,41].

2.2.3. Scope for Development of a Dual Fermentation for Up-Conversion of Cellulose into a Positive Energy Balance

The hydrothermal treatment of cellulose (model compound) was used as the hydrolysis method to extract energy products. The actual energy yield from the dual fermentation tests using the extracted glucose can be estimated from their bio-H2 yield. Dark fermentation in the bottle tests used in this study yielded 11 mL H2/g of cellulose from samples prepared between 250 and 260 °C, with a total energy content equivalent to 0.031 kW·h/Kg cellulose. Photo-fermentation yielded 4.05 mol H2/mol OA with a ten-fold higher total energy content equivalent to 0.363 kW·h/Kg cellulose, clearly reinforcing the conclusion that dark fermentation is more important as a supply of organic acids into the photo-fermentation than for its bio-H2 per se. The energy gained from DF alone still falls short of the 183 kJ/kg (0.051 KW·h/Kg) required for comminution of the example Miscanthus biomass (containing lignocellulosic compounds). However, the energy from photo-fermentation significantly surpasses comminution energy requirements. Process product yields and energy balance, taking into account additional energy demands, will be presented later.

2.3. Catalytic Upconversion of Cellulose Hydrolysate Side Stream into Liquid Fuel Precursor

2.3.1. Solvent Extraction of 5-Hydroxymethyl Furfural for One-Pot Catalysis

The liquid/liquid extraction method for 5-HMF extraction developed in previous studies [47,53] reported a MTHF extraction efficiency of 5-HMF from the aqueous phase from 59 to 63% resulting in 5-HMF concentrations in MTHF of c.a. 21 mM. The calculations of energy balance (later) will be based on this result.

2.3.2. Catalytic Up-Conversion of 5-HMF into DMF

In order to justify the use of post-fermentation E. coli to bio-manufacture and support a metallic catalyst, a comparison was made to the paradigm D. desulfuricans catalyst as the bio-Pd/Ru bimetallic. Although the latter organism makes highly effective metallic nanocatalysts (see Introduction), it is uneconomic to grow D. desulfuricans at scale ‘for purpose’, and is also challenging, due to its obligately anaerobic growth format and production of toxic H2S. Bio-nanocatalysts made by the two organisms were characterised and compared by Gomez-Bolivar [54] and the upgrading of 5-HMF by both is compared in Table 2. The nominal catalyst composition was 5 wt% Pd/5 wt% Ru. Whereas all of the Pd(II) was removed onto the cells into Pd(0), the removal of Ru(III) was incomplete, giving actual catalyst compositions as shown in Table 2.
The atomic weights of Pd and Ru are 106 and 101, respectively, hence, the nominal catalyst composition was ~equimolar but the available amount of each metal was not determined; both core-shell and alloyed nanostructures were reported as well as isolated monometallic ‘islands’ [54]. The E. coli bio-Pd/Ru catalyst was 34% and 33% better than the D. desulfuricans counterpart in the case of 5-HMF obtained commercially and that obtained from the hydrolysis of cellulose, respectively (Table 2). In each case, the DMF obtained from cellulose hydrolysate-5-HMF by the bio-catalysts was approximately half of that obtained from pure 5-HMF, which has adverse implications for bioprocess implementation. However, in contrast, the yield using a commercial Ru/C catalyst was 18-fold less in the hydrolysate than with pure 5-HMF (Table 2). This observation contrasts with an earlier study by Hu et al. [55] using a commercial Ru/C catalyst where the use of pure 5-HMF or that made from bioresourced fructose, other sugars and cellulose had little effect on the yield of DMF from the extracted 5-HMF. Understanding the reason for the contrast between this report and the current work was outside the scope of this study, but it should be noted that, while the current work used hot compressed water only, Hu et al. [55] performed cellulose hydrolysis using a sulfonated carbonaceous catalyst in ionic liquid (1-butyl-3-methylimidazolium) chloride at 160 °C (15 min) followed by 5-HMF extraction. This introduces complexity and cost, whereas the present study (which is not yet optimised) pioneers a simple one-pot system for catalytic 5-HMF up-conversion in the cellulose hydrolysate directly following removal and up-conversion of the fermentable fraction.
Hu et al. [55] surveyed various catalysts, concluding that Ru/C outperformed others (Raney-Ni, Pd/C, Pt/C, Rh/C). An early study using Bacillus benzeovorans [38] showed the promise of bimetallic bio-Pd/Ru as compared to single metals. Table 2 shows that in this study the bio-Ru and bio-Pd/Ru behaved similarly against commercial 5-HMF but the bimetallic had better activity by 20% against the 5-HMF from cellulose hydrolysate (a similar improvement using the bimetallic to that observed by Omajali et al. [38]).
Despite the facility of a one-pot conversion from cellulose to DMF, it is clear that both the 5-HMF conversion and yield must be improved to give comparable activity to that of the more complex process of Hu et al. [55] commensurate with a favourable overall energy balance and cost (see later Discussion).

2.3.3. Calculation of Energy Balance Using Obtained Data

The cellulose hydrolysates yielded 1.25 mol glucose/kg cellulose which, after dark fermentation, delivered 0.45 mol H2/Kg cellulose, which is ~8% of the hydrogen yielded by photo-fermentation (5.4 mol H2/Kg cellulose), which comes from the OA mix generated by dark fermentation (1.33 mol OA mix/Kg cellulose). This highlights the importance of dark fermentation as source of OAs for photo-fermentation as a coupled fermentation process. The 5-HMF extracts (at 65% extraction efficiency) provided 5-HMF concentrations of 0.083 mol 5-HMF/Kg cellulose. The hydrolysate-extracted 5-HMF in MTHF (adjusted to the same concentration of 5-HMF in each test), catalytically upgraded to 2,5-DMF (Table 2) using the bio-Pd/Ru E. coli catalyst, yielded 0.159 mol DMF/Kg cellulose (Table 3).
In terms of the energy balance, Table 4 shows the projected energy output from biohydrogen (see above) and the amount of energy from 2,5-DMF obtained per kilo of starting cellulose, which together totals 0.537 kW·h/Kg cellulose. This amount is 15-fold smaller than the total energy input for the heat-consuming reactions (R1 + R2, Table 4), which amount to 8.18 kW·h/Kg cellulose.
In any reaction that demands a significant amount of energy such as heat, an effective heat recovery system is critical to achieve energy efficiency. Assuming heat losses of 15% in a continuous reaction process, the heat supplied to make-up for these losses would need to be 1.23 kW·h/Kg cellulose. The total energy output would have to at least account for these losses to break even. In this case, it falls significantly short, rendering the integrated process energy inefficient as a means to realise energy from cellulose.
The maximum energy (calorific) output of the products on the basis of this study and other obtained data from related work is 0.537 kW·h/Kg cellulose (Table 4). Assuming no boosting via light ‘upgrading’ or advanced photobioreactor design, photofermentation provides a significant contribution, generating ~68% of the energy output. This is important as OAs (the PF feed) can be efficiently extracted from other abundant sources (see later discussion). The energy contribution from DMF is 27% of the energy output but only 5% comes from dark fermentation. Despite a good contribution of DMF to the total energy output, DMF contributes only 4% of the energy needed in the hydrogenation reaction, which is highly energy intensive, requiring 3.61 kW·h/Kg cellulose, i.e., 41% of the total energy input of the reactions, as seen in Table 4. This result has a significant negative impact on the economic viability of the overall process, considering also that the energy that would be needed for electro-separation and for the chemical extraction process were not included in the analysis. Hence, we conclude that, currently, in-process generation of DMF from the cellulose hydrolysate has limited potential without significant optimisation and, hence, the dual focus must now be on improving the catalysis component in tandem with other factors, working towards circularity.

2.4. Key Factors in Waste Up-Valorisation: Towards an Integrated Biorefinery Energy Process

The above discussion indicates that the energy products generated by the thermal-hydrolysis of cellulose (model compound used in this work) are currently incapable to provide, at the least, the energy needed for the overall process. In addition, in the valorisation of real lignocellulosic wastes, cellulose extraction from lignocellulosic biomass requires high ‘parasitic’ energy demand of upstream comminution, which is required to obtain a suitable particle size for cellulose extraction, which can then be efficiently achieved via delignification using environmentally friendly thermal subcritical water-ethanol-CO2 mediated hydrolysis (Pinitial 55 bar; Tfinal 200 °C) [58,59]. Nontrivially, the energy consumption to mill Miscanthus (moisture content of 15%) to <4 mm was 184 kJ/kg of dry matter [60], which can be compared to a hydrogen energy content of only 10 kJ/litre at 1 atm and 15 °C and, hence, 18.4 L of H2 would be ‘consumed’ per kilo of Miscanthus source material. Potentially mitigating against this total ‘parasitic’ energy demand (mechanical + thermal), the method [58] can drive biomass delignification without destroying the cellulose fibres, thereby generating two high value fractions: cellulose and lignin, the latter being an important source of aromatic compounds and chemicals [61]. In the case of cellulose, structural factors such as the reduction in crystallinity and increase in porosity as a result of the thermal treatment, make it potentially more accessible to hydrolysis treatments, facilitating enzymatic-biocatalytic hydrolysis of the weakened cellulose structure, and/or reducing the generation of degradation products and increasing sugar yields from the thermal hydrolysis. These factors, while potentially further enhancing the cellulose extraction, and its hydrolysis process economics, were beyond the focus of this scoping study, which aimed to ‘benchmark’ the potential of a dual-biofuel production from cellulosic feedstock.

3. Materials and Methods

3.1. Chemicals, Microorganisms and Biomass Resources

D(+) glucose was from Fisher Scientific. Activated carbon (AC) was colorsorb 5 steam activated powder from JACOBI (micropore, 0.19 cm3 g−1; mesopore, 0.37 cm3 g−1; macropore. 1.68 cm3 g−1; total surface area: 900 m2 g−1). All other chemicals, cellulose microcrystalline powder, 5-HMF (≥99%) and 2,5-DMF (99%) were from Sigma-Aldrich, as were commercial metal salts (Na2PdCl4 and RuCl3) and 5 wt% Pd/C and 5 wt% Ru/C catalysts.
Desulfovibrio desulfuricans, a Gram-negative comparator to Escherichia coli, was grown as described previously [27]. Escherichia coli HD701, highly efficient in the dark fermentation of HCW hydrolysate, was obtained from laboratory stock. Bacterial cells, grown fermentatively and harvested as described in Orozco et al. 2012 [45], Rhodobacter sphaeroides (for downstream photo-fermentation of the organic acids produced by E. coli) was obtained, grown and harvested as described previously by Redwood et al. 2012 [39]. For adaptation, the same OA mix from the E. coli HD701 mixed acid fermentation of glucose was included in the synthetic OA media for initial growth and later for H2 production tests. Carbon sources for the growth media were supplied to form a 40 mM solution as follows: acetate 13.3 mM; lactate 7.5 mM; succinate 10.4 mM and butyrate 8.8 mM. Cells were washed twice in 100 mL of phosphate buffered saline (PBS: 1.43 g Na2HPO4, 0.2 g KH2PO4, 0.8 g NaCl, 0.2 g KCl /L, pH 7.0) before storing the cell pellet at 4 °C.

3.2. Hot Compressed Water (HCW) Hydrolysis of Cellulose

HCW hydrolysis used a 250 cm3 reactor (Parr series 4570/80 HP/HT; Parr Instruments Company, Moline, IL, USA equipped with a heat/agitation controller (model 4836) and a cooling system (Grant LTD6/20).
Cellulose (5 g) was suspended in de-ionised water (DW) to a final reactant volume of 125 cm3 (40 g/L), leaving a headspace of 120 cm3. The reactor was sealed and purged with N2 or CO2 gases for 3 min with agitation (850 rpm) before pressurising to 30 bar with N2 (or CO2) and heating to set-point temperature (temperature range from 220–260 °C). Reaction conditions were held for 15 min before rapid cooling to 100 °C; the reactor was removed from the heating vessel and quenched in an ice-water bath. The reactor was depressurised and the hydrolysate was separated from the solid residue by vacuum filtration through two layers of filter paper (Fisherbrand QL100). The residue was washed with distilled water, dried at 60 °C and weighed. The hydrolysates were analysed for total organic carbon (TOC) using a TOC analyser (Model TOC 5050A, Shimadzu Co., Tokyo, Japan). Organic acids were analysed by anion HPLC using a Dionex 600-series system as described previously [62]. Sugars, 5-HMF and furfural were analysed by HPLC (Agilent 1100 series) equipped with on-line degasser, quaternary pump, auto-sampler and RI detector (1200 series) as detailed in [45]; gases were not collected or analysed. For the estimation of residual polyglucose, the residue was acid-hydrolysed [53], and the glucose product was analysed by the dinitrosalicylic assay method [56]. Hydrolysates and samples were stored at −20 °C. When, occasionally, precipitate appeared in the samples after defrosting they were filtered to remove the solid residue which was not analysed. Reaction conversions (Xp) were calculated according to: [(Initial sample concentration (ISC)-residue after the reaction)/ISC] × 100; ISC and residues are expressed in g (dry basis)/L. Several preparations at 260 °C were pooled to give samples for extraction and, as appropriate, onward catalytic upgrading.

3.3. Detoxification of Cellulose Hydrolysates Using Activated Carbon, Fermentative Biohydrogen Production by E. coli and Downstream Photo-Fermentation by Rhodobacter spp.

Cooled cellulose hydrolysates were treated with 5% (w/v) activated carbon as described by Orozco et al., 2012 [45]. The pH, not adjusted, was measured before and after hydrolysis. Treated hydrolysates and untreated controls were vacuum-filtered (Fisherbrand QL100 filter paper). Samples were kept at −20 °C until use.
The E. coli strain HD701 was grown, harvested and used for fermentation tests on untreated and treated cellulose hydrolysate as described previously for starch [45]. Routine measurement of H2 concentration (averages of 3 tests each) used a combustible gas meter (Gasurveyor2, GMI) and was estimated as described in Orozco et al., 2012 [40].
For H2 production tests, washed cells of Rhodobacter spp. (from 3.1) were resuspended in 20 mL of PBS to produce a concentrate containing >4 g dry wt./L (calculated from OD660 via a previously determined conversion) and dispensed at a final concentration of 1 g dry wt./L in 5 mL aliquots of mixed organic acid media containing total 20 mM as follows: acetate 6.7; lactate 3.8; succinate 5.2; butyrate 4.4 into 15 mL glass reactors as described by Redwood [41]. Controls used cells in media omitting carbon sources. Reactors were sealed using butyl rubber stoppers with aluminium crimp seals and purged with Ar (15 min) before incubation (30 °C, static, 75 W/m2 tungsten lamp, 72 h). Hydrogen was measured regularly by measuring the gas accumulated in the headspace of the bottles and the headspace pressure was determined as described by Ishaq [63].

3.4. Liquid–Liquid Separation of Soluble Fermentable Fraction from Degradation Products

Further to fermentability ‘scoping’ tests using native and AC-detoxified hydrolysate, a method for liquid–liquid extraction of 5-HMF was developed from the partition coefficients as given by Blumenthal et al. [25]. The mass transfer of 5-HMF from the aqueous to the organic phase was optimal at 60 °C (routinely adopted), 5-HMF added in a concentration range from 1–5 wt% in the aqueous feed, impacted minimally on the partition coefficients (not shown) into organic extraction solvent 2-methyl tetrahydrofuran (MTHF). For extraction, the cellulose hydrolysates were mixed in equal volumetric proportions with 2-MTHF in an Erlenmeyer flask as described by Mikheenko et al. [27]. The organic phase (top) contained 5-HMF and other components (noted where appropriate) and the aqueous phase (bottom) contained fermentable sugars and OAs. Extracts were stored at −20 °C until use, with samples analysed by GC as described in [27]. Solvent extraction efficiency was calculated according to extraction efficiency (%) = moles of 5-HMF in organic layer/moles of 5-HMF in the aqueous layer.

3.5. Metallic Catalysts Made Using D. desulfuricans and Post-Fermentation E. coli

The bimetallic Bio-Pd-Ru catalyst (nominally 5%Pd/5%Ru, or as shown) on dry D. desulfuricans or E. coli was made as described previously [27,54,57]. The 2 mM Pd (II) solution was reduced to Pd(0) on the cells under high purity H2 supplied by BOC UK, (30 min; complete removal (by assay) of residual soluble metal) to give 5 wt% bio-Pd(0). The bio-Pd(0) was washed twice (distilled water, DW), and added as a concentrated suspension into 1 mM Ru (III) solution (final concentration; volume was adjusted to give the required final metal loading on cells) to give a final loading of (nominally) 5 wt% Pd/5 wt% Ru. The actual composition of the catalyst was determined by assay of the residual solution [56,57] following (sequential) metal uptake (Pd then Ru) and was 5 wt% Pd with the Ru content as shown. The bio-Pd/Ru mixture was left to stand then saturated with H2 (as above; 180 rpm, 30 °C; 96 h). The presumptive bio-nanoparticles were washed three times (DW) and once with acetone (9000× g, 15 min, 4 °C), air-dried and ground manually.

3.6. Catalytic Upgrading of 5-HMF to 2,5-DMF

The reactions (catalytic transfer hydrogenations) were carried out in a 100 mL stainless steel Parr reactor series 4590 directly (260 °C) in the MTHF extract using (nominally) 5%Pd/5%Ru on dry E. coli as described previously [27]. The tests used (i) commercially obtained 5-HMF or (ii) 5-HMF obtained from cellulose hydrolysis following extraction into solvent as above. In (i) the reactor was charged with 250 mg of commercial 5-HMF (80 mM) in 25 mL of MTHF; the other tests used appropriate volumes of 5-HMF in MTHF extracted from the cellulose hydrolysates (to the same concentration of 5-HMF).

3.7. Analysis of Residual 5-HMF, 2,5 DMF and Other Products in the Catalytic Conversion Reaction

Samples were analysed using a GC-FID for quantification and a GCMS-QP2010s for compound identification. All GC-FID analysis was performed on a Shimadzu GC2014 equipped with a Shimadzu AOC-20i autosampler as described previously [27,54,57].
5-HMF conversions and 2,5 DMF yields were calculated as follows:
HMF conversion (%) = 1 − (moles of HMF in products/starting moles HMF) × 100;
DMF yield (%) = (moles of DMF in products/starting moles HMF) × 100;
DMF selectivity (%) = [1 − (moles of DMF in products/(starting moles 5-HMF-final moles 5-HMF)] × 100.

4. Conclusions

The hydrothermal liquefaction of cellulose (model compound) yielded 4.5 mol of H2/Kg of cellulose and 0.159 mol of 2,5 DMF/Kg of cellulose; this is equivalent to a potential energy output of 0.537 KW·h/Kg cellulose. The energy input for the hydrolysis and hydrogenation reactions amounts to 8.18 KW·h/Kg cellulose, which exceeds the energy output by 15-fold. The negative energy balance leads us to conclude that the overall process is energetically unfeasible; the energy balance is rarely considered or discussed in the (large) biotechnology literature. This study stresses the importance of energy projections in advance of a programme of research and development.
When using real lignocellulosic wastes, the energy needed for the comminution (0.051 KW·h/Kg material) (see earlier) and extraction of cellulose from the lignin will significantly further worsen the energy balance compared to that obtained with cellulose, thereby reinforcing the conclusion that the process utilised in this work using real lignocellulosic waste as a source of cellulose will be energetically and economically unviable. This overall conclusion would, without significant advances in the catalytic upgrading of the 5-HMF (and other side stream products—see Discussion and Supplementary Materials), significantly impact upon a key future role for biorefinery as a candidate for energy from lignocellulosic wastes, steering the ‘landscape’ in favour of other approaches such as e.g., supercritical water liquefaction or gasification. These do not require an extraction step but are also energy-demanding; here, a similar analysis awaits the outcome of ongoing studies, implementing the recycling of low-grade heat.
Particular attention should be given to the photo-fermentation of organic acids. In this work, OAs were able to generate 4.05 mol of H2/mol of OAmix, thereby contributing to generate c.a. 67% of the total energy output. These results demonstrated that the organic acids (OAs) are a potentially significant source of bio-H2 and energy.
OAs are typically present in municipal wastewater treatment plants (e.g., 300,000 m3/day; OA content 4 g/L) [55]. The realisation of energy production potential using the photo-fermentation of such waste OAs has been shown; 1000 MW p.a was achievable using biogas from ‘traditional’ wastewater treatment, which was doubled by OA photo-fermentation into H2 by Rhodobacter sphaeroides, bypassing a dedicated dark fermentation [49]. In addition to wastewater, the acetogenesis (first) phase of anaerobic digestion (AD) offers scope for electrodialytic intervention with a photo-fermentation loop of extracted OAs to value-add to the final biogas, e.g., via production of ‘hythane’. This mixture of hydrogen and AD-methane is compatible with current gas grids and yields more energy than methane alone [64]. As such, retrofitting a photo-fermentation ‘loop’ into existing waste treatment processes may be a shorter-term interim advance, pending more productive biorefinery development.
With respect to solid waste, such as we consider here, the Cambi process is an example of a successful application of thermal hydrolysis treatment under milder conditions together with a very effective heat recovery system, bringing a positive energy balance. This process hydrolyses and disintegrates sludge from wastewater treatment plants (up to 18% total solids, 160 °C, 6–10 bars, 20–30 min retention time) generating a low viscosity sterilised liquid that is easily digested in an anaerobic digestion reactor (and, hence, would be potentially useful for adding a photo-fermentation loop; see above). Operating conditions were sustained during 15 days of hydraulic retention time under mesophilic conditions (37 °C); pH 7; digestor load at 100 t dry solids/day. As a result, biogas and electricity production increased by 20% at least, with 40% decrease in final sludge volume as compared to conventional digestion. With optimal operation of the thermal hydrolysis plant, the net energy ratio of the additional biogas is approximately 3 times in relation to the parasitic energy needed in a continuous process [65,66].

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/catal12060577/s1, Figure S1: Increase in sugar content of hydrolysate (%) by the application of ohmic heating (A), Hydrogen production potential of the hydrolysate (B), HPLC chromatogram of OH treated fruit wastes show higher content of organic acids (substrates for downstream photobiological hydrogen production) (C). In addition, a new peak (unidentified) is apparent following ohmic treatment of the waste (open arrow) [67]; Figure S2: Yields of products in brewery grain hydrolysates before AC at different reaction temperatures; Figure S3: Yields of products of wheat straw hydrolysate before AC at 220 and 240 °C. References [67,68,69,70,71,72,73,74,75,76,77,78,79,80] are cited in the Supplementary Materials.

Author Contributions

R.L.O. performed the cellulose hydrolysis (with G.A.L.), developed the liquid–liquid extraction of 5-HMF from hydrolysate, and performed E. coli fermentations of the hydrolysate. R.L.O. authored the first draft and carried out the energy balances. He also oversaw the R. sphaeroides biohydrogen tests, carried out with A.J.S., I.P.M. and M.L.M., oversaw the catalysis studies, carried out by J.G.-B., L.E.M., integrated the work and authored the draft to include the catalysis and up-valorisation, with contributions from all authors into the final manuscript, first-authored by J.G.-B. and R.L.O. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by EPSRC (Grants EP/D05768X/1 and EP/E034888/1), BBSRC (Grants BB/C516195/2 and BB/E003788/1) and NERC (Grant NE/L014076/1), the Royal Society (Industrial Fellowship to L.E.M.), Advantage West Midlands (Grant ref. POC46) and the Government of Mexico (studentship no. 203186 to R.L.O.). This work was partially supported by the Spanish Government Sistema Nacional de Garantía Juvenil grant PEJ-2014-P-00391 (Promoción de Empleo Joven e Implantación de la Garantía Juvenil 2014, MINECO) with a scholarship to J.G.-B.

Data Availability Statement

Not applicable.

Acknowledgments

We thank the Combined Workshop, School of Biosciences, University of Birmingham for technical support and C-Tech Innovation Ltd. for collaboration for the work shown in Figure S1. This paper is dedicated to the memory of the late David Penfold whose work started our hydrogen biorefinery journey in 2000.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Tokarska, K.B.; Gillet, N.P.; Weaver, A.J.; Arora, Y.K.; Eby, M. The climate response to five trillion tonnes of carbon. Nat. Clim. Chang. Lett. 2016, 6, 851–855. [Google Scholar] [CrossRef]
  2. Dębowski, M.; Dudek, M.; Zieliński, M.; Nowicka, A.; Kazimierowicz, J. Microalgal hydrogen production in relation to other biomass-based technologies—A Review. Energies 2021, 14, 6025. [Google Scholar] [CrossRef]
  3. Park, J.-H.; Chandrasekhar, K.; Jeon, B.-H.; Jang, M.; Liu, Y.; Kim, S.-H. State-of-the-art technologies for continuous high-rate biohydrogen production. Bioresour. Technol. 2021, 320, 124304. [Google Scholar] [CrossRef] [PubMed]
  4. Anonymous. Hydrogen Generation Market by Generation & Delivery Mode, Technology, Application & Region—Global Forecast to 2021. Available online: http://www.marketsandmarkets.com/Market-Reports/hydrogen-generation-market-494.html (accessed on 21 June 2016).
  5. Macaskie, L.E.; Baxter-Plant, V.S.; Creamer, N.J.; Humphries, A.C.; Penfold, D.W.; Yong, P. Applications of bacterial hydrogenases in waste decontamination, manufacture of novel bionanomaterials in sustainable energy. Biochem. Soc. Trans. 2005, 33, 71–79. [Google Scholar] [CrossRef]
  6. Sasaki, M.; Ohsawa, K. Hydrolysis of lignocellulosic biomass in hot-compressed water with supercritical carbon dioxide. ACS Omega 2021, 6, 14252–14259. [Google Scholar] [CrossRef]
  7. Dogaris, I.; Karapati, S.; Mamma, D.; Kalogeris, E.; Kekos, D. hydrothermal processing and enzymatic hydrolysis of sorghum bagasse for fermentable carbohydrates production. Biores. Technol. 2009, 100, 6543–6549. [Google Scholar] [CrossRef]
  8. Dadwal, A.; Sharma, S.; Satyanarayana, T. Thermostable cellulose saccharifying microbial enzymes: Characteristics, recent advances and biotechnological applications. Int. J. Biol. Macromol. 2021, 188, 226–244. [Google Scholar] [CrossRef]
  9. Horn, S.J.; Vaaje-Kolstad, G.; Westering, B.; Eijsink, V.G.H. Novel enzymes for the degradation of cellulose. Biotechnol. Biofuels 2012, 5, 45–60. [Google Scholar] [CrossRef] [Green Version]
  10. Duff, S.J.B.; Murray, W.D. Bioconversion of forest products industry waste cellulosics to fuel ethanol: A review. Biores. Technol. 1996, 55, 1–33. [Google Scholar] [CrossRef]
  11. Mok, W.S.; Antal, M.J.; Varhegyi, G. Productive and parasitic pathways in dilute acid-catalyzed hydrolysis of cellulose. Ind. Eng. Chem. Res. 2002, 31, 94–100. [Google Scholar] [CrossRef]
  12. Liao, Y.; Beeck, B.; Thielemans, K.; Ennaert, T.; Snelders, J.; Dusselier, M.; Courtin, C.; Sels, B. The role of pretreatment in the catalytic valorization of cellulose. Mol. Catal. 2020, 487, 110883. [Google Scholar] [CrossRef]
  13. Kamio, E.; Takahashi, S.; Noda, H.; Fukuhara, C.; Okamura, T. Effect of heating rate on liquefaction of cellulose by hot compressed water. Chem. Eng. J. 2008, 137, 328–338. [Google Scholar] [CrossRef]
  14. Minowa, T.; Zhen, F.; Ogi, T.; Varhegyi, G. Liquefaction of cellulose in hot compressed water using sodium carbonate: Products distribution at different reaction temperatures. J. Chem. Eng. 1997, 30, 186–190. [Google Scholar] [CrossRef] [Green Version]
  15. Minowa, T.; Zhen, F.; Ogi, T. Cellulose decomposition in hot-compressed water with alkali or nickel catalyst. J. Supercrit. Fluids 1998, 13, 253–259. [Google Scholar] [CrossRef]
  16. Miyazawa, T.; Funazukuri, T. Polysaccharide hydrolysis accelerated by adding carbon dioxide under hydrothermal conditions. Biotechnol. Prog. 2005, 21, 1782–1785. [Google Scholar] [CrossRef]
  17. Irmak, S.; Kurtuluş, M.; Hasanoğlu, A.; Erbatur, O. Gasification efficiencies of cellulose, hemicellulose and lignin fractions of biomass in aqueous media by using Pt on activated carbon catalyst. Biomass Bioenergy 2013, 30, 102–108. [Google Scholar] [CrossRef]
  18. Pipitone, G.; Zoppi, G.; Frattini, A.; Bocchini, S.; Pirone, R.; Bensaid, S. Aqueous phase reforming of sugar-based biorefinery streams: From the simplicity of model compounds to the complexity of real feeds. Catal. Today 2020, 345, 267–279. [Google Scholar] [CrossRef]
  19. Meng, X.; Ragauskas, J. Recent advances in understanding the role of cellulose accessibility in enzymatic hydrolysis of lignocellulosic substrates. Curr. Opin. Biotechnol. 2014, 27, 150–158. [Google Scholar] [CrossRef]
  20. Mills, T.Y.; Sandoval, N.R.; Gill, R.T. Cellulosic hydrolysate toxicity and tolerance mechanisms in Escherichia coli. Biotechnol. Biofuel. 2009, 2, 26. [Google Scholar] [CrossRef] [Green Version]
  21. Ezeji, T.; Qureshi, N.; Blaschek, H.P. Butanol production from agricultural residues: Impact of degradation products on Clostridium beijerinckii growth and butanol fermentation. Biotechnol. Bioeng. 2007, 97, 1460–1469. [Google Scholar] [CrossRef]
  22. Hodge, D.B.; Andersson, C.; Berglund, K.A.; Rova, U. Detoxification requirements for bioconversion of softwood dilute acid hydrolyzates to succinic acid. Enzym. Microb. Technol. 2009, 44, 309–316. [Google Scholar] [CrossRef]
  23. Palmqvist, E.; Hahn-Hägerdal, B. Fermentation of lignocellulosic hydrolysates. I: Inhibition and detoxification. Bioresour. Technol. 2000, 74, 17–24. [Google Scholar] [CrossRef]
  24. Carter, B.; Gilcrease, P.C.; Menkhaus, T.J. Removal and recovery of furfural, 5 hydroxymethyl-furfural, and acetic acid from aqueous solutions using a soluble polyelectrolyte. Biotechnol. Bioeng. 2011, 108, 2046–2052. [Google Scholar] [CrossRef] [PubMed]
  25. Blumenthal, L.C.; Jens, C.M.; Ulbrich, J.; Schwering, F.; Langrehr, T.; Turek, K.; Kunz, U.; Leonhard, K.; Palkovits, R. Systematic identification of solvents optimal for the extraction of 5-hydroxymethylfurfural from aqueous reactive solutions. ACS Sustain. Chem. Eng. 2016, 4, 228–235. [Google Scholar] [CrossRef]
  26. Aycock, D.F. Solvent applications of 2-methyltetrahydrofuran in organometallic and biphasic reactions. Org. Process Res. Dev. 2007, 11, 156–159. [Google Scholar] [CrossRef]
  27. Mikheenko, I.; Gomez-Bolivar, J.; Merroun, M.L.; Macaskie, L.E.; Sharma, S.; Walker, M.; Hand, R.A.; Grail, B.; Johnson, D.B.; Orozco, R.L. Upconversion of cellulosic waste into a potential ‘drop in fuel’ via novel catalyst generated using Desulfovibrio desulfuricans and a consortium of acidophilic sulfidogens. Front. Microbiol. 2019, 10, 970. [Google Scholar] [CrossRef]
  28. Zhorg, S.; Daniel, R.; Xu, H.; Zhang, J.; Turner, D.; Wyszynski, M.L.; Richards, P. Combustion and emission of 2,5-dimethylfuran in a direct-injection spark-ignition engine. Energy Fuels 2010, 24, 2891–2899. [Google Scholar]
  29. Davis, S.E.; Houk, L.R.; Tamargo, E.E.; Datye, A.K.; Davis, R.J. Oxidation of 5-hydroxymethylfurfural over supported Pt, Pd and Au catalysts. Catal. Today 2011, 160, 55–60. [Google Scholar] [CrossRef]
  30. Tong, X.; Ma, Y.; Li, Y. Biomass into chemicals: Conversion of sugars to furan derivatives by catalytic processes. Appl. Catal. A Gen. 2011, 385, 1–13. [Google Scholar] [CrossRef]
  31. Omajali, J.B. Novel Bionanocatalysts for Green Chemistry Applications. Ph.D. Thesis, University of Birmingham, Birmingham, UK, 2015. [Google Scholar]
  32. Zhu, J.; Wood, J.; Deplanche, K.; Mikheenko, I.P.; Macaskie, L.E. Selective hydrogenation using palladium bioinorganic catalyst. Appl. Catal. B Environ. 2016, 199, 108–122. [Google Scholar] [CrossRef] [Green Version]
  33. Orozco, R.L.; Redwood, M.D.; Yong, P.; Caldelari, I.; Sargent, F.; Macaskie, L.E. Towards an integrated system for bio-energy: Hydrogen production by Escherichia coli and use of palladium-coated waste cells for electricity generation in a fuel cell. Biotechnol. Lett. 2010, 32, 1837–1845. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  34. Murray, A.J.; Mikheenko, I.P.; Deplanche, K.; Omajali, J.B.; Gomez-Bolivar, J.; Merroun, M.L.; Macaskie, L.E. Chapter 9: Biorefinjng of metallic wastes into new nanomaterials for green chemistry, environment and energy. In Resource Recovery from Wastes: Towards a Circular Economy; Macaskie, L.E., Sapsford, D.J., Mays, W.M., Eds.; Royal Society of Chemistry: London, UK, 2020; pp. 215–240. ISBN 978-1-78801-381-9. [Google Scholar]
  35. Archer, S.A.; Murray, A.J.; Omajali, J.B.O.; Paterson-Beedle, M.; Sharma, B.K.; Wood, J.; Macaskie, L.E. Chapter 13: Metallic wastes into new processes catalysts: Life cycle and environmental benefits within integrated analyses using selected case histories. In Resource Recovery from Wastes: Towards a Circular Economy; Macaskie, L.E., Sapsford, D.J., Mays, W.M., Eds.; Royal Society of Chemistry: London, UK, 2020; pp. 315–342. ISBN 978-1-78801-381-9. [Google Scholar]
  36. Hallenbeck, P.C. Microbial Technologies in Advanced Biofuels Production; Springer: Berlin/Heidelberg, Germany, 2012; pp. 3–269. [Google Scholar]
  37. Yong, P.; Mikheenko, I.P.; Deplanche, K.; Redwood, M.D.; Macaskie, L.E. Biorefining of precious metals from wastes: An answer to manufacturing of cheap catalysts for fuel cells and power generation via an integrated biorefinery? Biotechnol. Lett. 2010, 35, 1821–1828. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  38. Omajali, J.B.; Gomez-Bolivar, J.; Mikheenko, I.P.; Sharma, S.; Kayode, B.; Al-Duri, B.; Banerjee, D.; Walker, M.; Merroun, M.L.; Macaskie, L.E. Novel catalytically active Pd/Ru bimetallic nanoparticles synthesised by Bacillus benzeovorans. Sci. Rep. 2019, 9, 4175. [Google Scholar] [CrossRef] [PubMed]
  39. Redwood, M.D.; Orozco, R.L.; Majewski, A.J.; Macaskie, L.E. An integrated biohydrogen refinery: Synergy of photofermentation, extractive fermentation and hydrothermal hydrolysis for food waste. Biores. Technol. 2012, 119, 384–392. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  40. Orozco, R.L.; Mikheenko, I.P.; Omajali, J.B.; Macaskie, L.E. Chapter 3: Biohydrogen production from agricultural and food wastes and potential for side stream valorisation from waste hydrolysates. In Resource Recovery from Wastes: Towards a Circular Economy; Macaskie, L.E., Sapsford, D.J., Mays, W.M., Eds.; Royal Society of Chemistry: London, UK, 2020; pp. 57–86. ISBN 978-1-78801-381-9. [Google Scholar]
  41. Redwood, M.D. Bio-Hydrogen Production and Biomass-Supported Palladium Catalyst for Energy Production and Waste Minimization. Ph.D. Thesis, University of Birmingham, Birmingham, UK, 2007. [Google Scholar]
  42. Orozco, R.L.; Redwood, M.D.; Leeke, G.A.; Bahari, A.; Santos, R.C.D.; Macaskie, L.E. Hydrothermal hydrolysis of starch with CO2 and detoxification of the hydrolysates with activated carbon for bio-hydrogen fermentation. Int. J. Hydrog. Energy 2012, 37, 6545–6553. [Google Scholar] [CrossRef] [Green Version]
  43. Redwood, M.D.; Macaskie, L.E. A two-stage, two-organism process for biohydrogen from glucose. Int. J. Hydrog. Energy 2006, 31, 1514–1521. [Google Scholar] [CrossRef]
  44. Saeman, J.F.; Bubl, J.L.; Harris, E.E. Quantitative saccharification of wood and cellulose. Ind. Eng. Chem. Anal. Ed. 2002, 17, 35–37. [Google Scholar] [CrossRef]
  45. Chaplin, M.F. Monosaccharides. In Carbohydrate Analysis: A Practical Approach; Chaplin, M.F., Kennedy, J.F., Eds.; IRL Press; Oxford University Press: Oxford, UK, 1986; p. 3. [Google Scholar]
  46. Ishaq, F. Trace Metal Supplementation in Wastewater Sludge Digesters. Ph.D. Thesis, University of Birmingham, Birmingham, UK, 2012. [Google Scholar]
  47. Gomez-Bolivar, J.; Mikheenko, I.P.; Orozco, R.L.; Sharma, S.; Banerjee, D.; Walker, M.; Hand, R.A.; Merroun, M.L.; Macaskie, L.E. Synthesis of Pd/Ru bimetallic nanoparticles by Escherichia coli and potential as a catalyst for upgrading 5-hydroxymethyl furfural into liquid fuel precursors. Front. Microbiol. 2019, 10, 1276. [Google Scholar] [CrossRef] [Green Version]
  48. Gomez-Bolivar, J. Synthesis of Palladium and Ruthenium Nanoparticles from Metal Solutions Using Bacteria with Application as Nanocatalyst. Ph.D. Thesis, University of Granada, Granada, Spain, 2019. [Google Scholar]
  49. Sakaki, T.; Shibata, M.; Miki, T.; Hirosue, H.; Hayashi, N. Decomposition of cellulose in near-critical water and fermentability of the products. Energy Fuels 1996, 10, 684–688. [Google Scholar] [CrossRef]
  50. Lam, R.K.; England, A.H.; Sheardy, A.T.; Shih, O.; Smith, J.W.; Rizzuto, A.M. The hydration structure of aqueous carbonic acid from X-ray absorption spectroscopy. Chem. Phys. Lett. 2014, 614, 282–286. [Google Scholar] [CrossRef] [Green Version]
  51. Toews, K.L.; Shroll, R.M.; Wai, C.M. pH-Defining equilibrium between water and supercritical CO2. influence on SFE of organics and metal chelates. Anal. Chem. 1995, 67, 4040–4043. [Google Scholar] [CrossRef]
  52. Hallberg, K.; Johnson, D.B. Biodiversity of acidophilic prokaryotes. Adv. Appl. Microbiol. 2001, 49, 37–84. [Google Scholar] [PubMed]
  53. Stolzenburg, K.; Mubbala, R. Integrated Design for Demonstration of Efficient Liquefaction of Hydrogen (IDEALHY) Hydrogen Liquefaction Report; EU Grant No 2781772013; IDEALHY: Maastricht, The Netherlands, 2013. [Google Scholar]
  54. Orozco, R.L. Hydrogen Production from Biomass by Integrating Thermo-Chemical and Biological Processes. Ph.D. Thesis, University of Birmingham, Birmingham, UK, 2011. [Google Scholar]
  55. Stephen, A.J.; Archer, S.A.; Orozco, R.L.; Macaskie, L.E. Advances and bottlenecks in microbial hydro-gen production. Microb. Biotechnol. 2017, 10, 1120–1127. [Google Scholar] [CrossRef] [PubMed]
  56. Eroğlu, E.; Gündüz, U.; Yücel, M.; Türker, L.; Eroğlu, I. Photobiological hydrogen production by using olive mill wastewater as a sole substrate source. Int. J. Hydrog. Energy 2004, 29, 163–171. [Google Scholar] [CrossRef]
  57. Murray, A.J.; Roussel, J.; Rolley, J.; Woodhall, F.; Mikheenko, I.P.; Johnson, D.B.; Gomez-Bolivar, J.; Merroun, M.L.; Macaskie, L.E. Biosynthesis of zinc sulfide quantum dots using waste off-gas from a metal bioremediation process. RSC Adv. 2017, 7, 21484–21491. [Google Scholar] [CrossRef] [Green Version]
  58. Murray, A.J.; Love, J.; Redwood, M.D.; Orozco, R.L.; Tennant, R.K.; Woodhall, F.; Goodridge, A.; Macaskie, L.E. Chapter 9: Enhancement of photosynthetic productivity by quantum dots application. In Nonmagnetic and Magnetic Quantum Dots; Stavrou, V.N., Ed.; InTech Publications: Rijeka, Croatia, 2018; pp. 147–174. [Google Scholar] [CrossRef] [Green Version]
  59. Hu, L.; Tang, X.; Xu, J.; Wu, Z.; Lin, L.; Liu, S. Selective transformation of 5-hydroxymethyl furfural into the liquid fuel 2,5-dimethylfuran over carbon-supported ruthenium catalyst. Ind. Eng. Chem. Res. 2014, 53, 3056–3064. [Google Scholar] [CrossRef]
  60. Roque, R.M.N.; Baig, M.N.; Leeke, G.A.; Bowra, S.S.; Santos, R.C.D. Study on sub-critical water mediated hydrolysis of Miscanthus a lignocellulosic biomass. Resour. Conserv. Recycl. 2012, 59, 43–46. [Google Scholar] [CrossRef]
  61. Pasquini, D.; Pimenta, M.T.B.; Ferreira, L.H.; Curvelo, A.A.S. Extraction of lignin from sugar cane bagasse and Pinus taeda wood chips using ethanol–water mixtures and carbon dioxide at high pressures. J. Supercrit. Fluids 2005, 36, 31–39. [Google Scholar] [CrossRef]
  62. Miao, Z.; Grift, T.E.; Hansen, A.C.; Ting, K.C. Energy requirement for comminution of biomass in relation to article physical properties. Ind. Crops Prod. 2011, 33, 508–513. [Google Scholar] [CrossRef]
  63. Toledano, A.; García, A.; Mondragon, I.; Labidi, J. Lignin separation and fractionation by ultrafiltration. Sep. Purif. Technol. 2010, 71, 38–43. [Google Scholar] [CrossRef]
  64. Mahant, B.; Linga, P.; Kumar, R. Hydrogen economy and role of Hythane as a bridging solution: A perspective review. Energy Fuels 2021, 35, 15424–15454. [Google Scholar] [CrossRef]
  65. Ringoot, D.; Kleiven, H.; Panter, K. Energy efficient thermal hydrolysis with steam explosion. In Proceedings of the 17th European Biosolids and Organic Resources Conference, Leeds, UK, 19–21 November 2012.
  66. Kepp, U.; Machenbach, I.; Weisz, N.; Solheim, O.E. Enhanced stabilisation of sewage sludge through thermal hydrolysis—Three years’ experience with full scale plant. Water Sci. Technol. 2000, 42, 89–96. [Google Scholar] [CrossRef]
  67. Penfold, D.W.; Redwood, M.D.; Yong, P.; Stratton-Campbell, D.; Skibar, W.; Macaskie, L.E. Microbial hydrogen and electricity from wastes. In Proceedings of the 7th Hydrogen-Power and Theoretical Engineering Solutions (HyPoThESIS VII), Merida, Mexico, 2007; ISBN 968-6114-21-1. [Google Scholar]
  68. Sakr, M.; Liu, S. A Comprehensive review on applications of ohmic heating. Renew. Sustain. Energy Rev. 2014, 39, 262–269. [Google Scholar] [CrossRef]
  69. Delbecq, F.; Wang, Y.; Muralidhara, K.E.; Ouardi, G.; Marlair, G.; Lin, C. Hydrolysis of hemicellulose and derivatives—A review of recent advances in the production of furfural. Front. Chem. 2018, 6. [Google Scholar] [CrossRef] [PubMed]
  70. Shen, G.; Andrioletti, B.; Queneau, Y. Furfural and 5-(hydroxymethyl) furfural: Two pivotal intermediates for bio-based chemistry. Curr. Opin. Green Sustain. Chem. 2020, 26, 100384. [Google Scholar] [CrossRef]
  71. Silva, J.F.L.; Mariano, A.P.; Filho, R.M. Economic production of 2-methyl tetrahydrofuran (MTHF) and ethyl levulinate (EL) produced from hemicellulose-derived furfural. Biomass Bioenergy 2018, 119, 492–502. [Google Scholar] [CrossRef]
  72. Murray, A.J.; Zhu, J.; Wood, J.; Macaskie, L.E. Biorefining of platinum group metals from model waste solutions into catalytically active bimetallic nanoparticles. Microb. Biotechnol. 2018, 11, 359–368. [Google Scholar] [CrossRef] [Green Version]
  73. Deplanche, K.; Merroun, M.L.; Casadesus, M.; Tran, D.T.; Mikheenko, I.P.; Bennett, J.A.; Zhu, J.; Jones, I.P.; Attard, G.A.; Selenska-Pobell, S.; et al. Microbial synthesis of core/shell gold/palladium nanoparticles for applications in green chemistry. J. R. Soc. Interface 2012, 9, 1705–1712. [Google Scholar] [CrossRef]
  74. Nishimura, S.; Ikeda, N.; Ebitani, K. Selective hydrogenation of biomass derived 5-hydroxymethyl furfural (HMF) to 2,5-dimethyl furan (DMF) under atmospheric hydrogen pressure over carbon supported PdAu bimetallic catalyst. Catal. Today 2014, 232, 89–98. [Google Scholar] [CrossRef]
  75. Beauregard, D.A.; Yong, P.; Macaskie, L.E.; Johns, M.L. Use of non-invasive magnetic resonance imaging to assess the reduction of Cr(VI) using a biofilm-palladium catalyst. Biotechnol. Bioeng. 2010, 107, 11–20. [Google Scholar] [CrossRef]
  76. Yong, P.; Liu, W.; Zhang, Z.; Beauregard, D.; Johns, M.L.; Macaskie, L.E. One step bioconversion of waste precious metals into Serratia biofilm-immobilised catalyst for Cr(VI) reduction. Biotechnol. Letts. 2015, 37, 2181–2191. [Google Scholar] [CrossRef] [PubMed]
  77. Orsi, E.; Beekwilder, J.; Eggink, G.; Kengen, S.W.M.; Weusthuis, A. The transition of Rhodobacter sphaeroides into a microbial cell factory. Biotechnol. Bioeng. 2021, 118, 521–541. [Google Scholar] [CrossRef] [PubMed]
  78. Deilami, S. Advanced Studies of Catalytic Upgrading of Heavy Oil and Pyrolysis Bio-Oil. Ph.D. Thesis, University of Birmingham, Birmingham, UK, 2019. [Google Scholar]
  79. Lee, S.Y.; Sankaran, R.; Chew, K.W.; Tan, C.H.; Krishnamoorthy, R.; Chu, D.-T.; Show, P.-L. Waste to bioenergy: A review on the recent conversion technologies. BMC Energy 2019, 1, 4–26. [Google Scholar] [CrossRef]
  80. Zhang, Z.-H.; Sun, Z.; Yuan, T.-Q. Recent Advances in the catalytic upgrading of biomass platform chemicals via hydrotalcite-derived metal catalysts. Trans. Tianjin Univ. 2021, 28, 89–111. [Google Scholar] [CrossRef]
Catalysts 12 00577 g001 550
Figure 1. Reaction heating and cooling pathways delivered by the reactor at different temperatures. There was no difference between tests using CO2 and N2. Set temperatures (°C) were: ⬦, 220; □, 230; △, 240; X, 250; O, 260. Heating pathway is based on average values from two experiments, with variation between replicates within ±5%.
Figure 1. Reaction heating and cooling pathways delivered by the reactor at different temperatures. There was no difference between tests using CO2 and N2. Set temperatures (°C) were: ⬦, 220; □, 230; △, 240; X, 250; O, 260. Heating pathway is based on average values from two experiments, with variation between replicates within ±5%.
Catalysts 12 00577 g001
Catalysts 12 00577 g002 550
Figure 2. (a) Cellulose conversion (Xc) to products with reaction temperature. Two independent experiments under N2, shown ((●), (○)), are for a starting cellulose concentration of 40 g.L−1; (▬) represents average Xc under N2; (△) is the control experiment with no heat. (b) Cellulose conversion (Xc) to products with reaction temperature under CO2; (●), (○) denote two independent experiments under CO2; (▬) represents average Xc under CO2. Reaction pressure was from 60 to 90 bars. Cellulose conversion to products is the average of two experiments with variation between replicates within ±5%.
Figure 2. (a) Cellulose conversion (Xc) to products with reaction temperature. Two independent experiments under N2, shown ((●), (○)), are for a starting cellulose concentration of 40 g.L−1; (▬) represents average Xc under N2; (△) is the control experiment with no heat. (b) Cellulose conversion (Xc) to products with reaction temperature under CO2; (●), (○) denote two independent experiments under CO2; (▬) represents average Xc under CO2. Reaction pressure was from 60 to 90 bars. Cellulose conversion to products is the average of two experiments with variation between replicates within ±5%.
Catalysts 12 00577 g002
Catalysts 12 00577 g003 550
Figure 3. Total organic carbon (a) and product yield (b) in hydrolysates before AC under CO2 (filled symbols) and N2 (open symbols). a): (●,○), TOC total (hydrolysate + residue); (▲,△), TOC residue; (■,□), TOC hydrolysate. b): (▲,△), glucose; (■,□), fructose; (◆,◇), 5-HMF in hydrolysates. The TOC total does not include material lost in some samples due to precipitation, char in residue and gas composition, which was estimated to be less than 5%wt. Yield values are average of two experiments with variation between replicates within ±5%.
Figure 3. Total organic carbon (a) and product yield (b) in hydrolysates before AC under CO2 (filled symbols) and N2 (open symbols). a): (●,○), TOC total (hydrolysate + residue); (▲,△), TOC residue; (■,□), TOC hydrolysate. b): (▲,△), glucose; (■,□), fructose; (◆,◇), 5-HMF in hydrolysates. The TOC total does not include material lost in some samples due to precipitation, char in residue and gas composition, which was estimated to be less than 5%wt. Yield values are average of two experiments with variation between replicates within ±5%.
Catalysts 12 00577 g003
Catalysts 12 00577 g004 550
Figure 4. Concentrations (mM) of 5-HMF before (filled symbols) and after (open symbols) AC treatment for experiments prepared under CO2 (▲,△) and N2 (●,○) at different reaction temperatures. Results shown are average values of two experiments with variation within ±5%.
Figure 4. Concentrations (mM) of 5-HMF before (filled symbols) and after (open symbols) AC treatment for experiments prepared under CO2 (▲,△) and N2 (●,○) at different reaction temperatures. Results shown are average values of two experiments with variation within ±5%.
Catalysts 12 00577 g004
Catalysts 12 00577 g005a 550Catalysts 12 00577 g005b 550
Figure 5. (a) Growth of E. coli in hydrolysates prepared under CO2 at different reaction temperatures (●) without and (○) with AC treatment. The dotted line represents the biomass yield from the glucose control (20 mM); (b) Hydrogen yields after fermentation of the hydrolysates produced at various temperatures. Under CO2 no H2 was produced before AC treatment from samples prepared at temperatures higher than 230 °C; under N2 no H2 was produced from samples prepared at above 240 °C before AC treatment. At 220 °C the concentration of 5-HMF in the hydrolysate is too low to cause total inhibition and H2 is produced. As temperature increases, and the concentration of 5-HMF increases, the inhibitory effect is evident.
Figure 5. (a) Growth of E. coli in hydrolysates prepared under CO2 at different reaction temperatures (●) without and (○) with AC treatment. The dotted line represents the biomass yield from the glucose control (20 mM); (b) Hydrogen yields after fermentation of the hydrolysates produced at various temperatures. Under CO2 no H2 was produced before AC treatment from samples prepared at temperatures higher than 230 °C; under N2 no H2 was produced from samples prepared at above 240 °C before AC treatment. At 220 °C the concentration of 5-HMF in the hydrolysate is too low to cause total inhibition and H2 is produced. As temperature increases, and the concentration of 5-HMF increases, the inhibitory effect is evident.
Catalysts 12 00577 g005aCatalysts 12 00577 g005b
Table
Table 2. One-pot catalytic upgrading of 5-HMF to 2,5-DMF in cellulose hydrolysate. Commercial catalyst (comm); Desulfovibrio desulfuricans (Dd) 5% Pd/5% Ru; Escherichia coli (Ec) 5% Pd/5% Ru. ND: not determined. Errors between technical replicates and between experiments were within 10%. Bold: comparison between the two Gram-negative strains. This table contains average values of two experiments. Actual composition of bimetallic Ec was (5%Pd/4.7%Ru), 5%RuEc (2.6%Ru) [47].
Table 2. One-pot catalytic upgrading of 5-HMF to 2,5-DMF in cellulose hydrolysate. Commercial catalyst (comm); Desulfovibrio desulfuricans (Dd) 5% Pd/5% Ru; Escherichia coli (Ec) 5% Pd/5% Ru. ND: not determined. Errors between technical replicates and between experiments were within 10%. Bold: comparison between the two Gram-negative strains. This table contains average values of two experiments. Actual composition of bimetallic Ec was (5%Pd/4.7%Ru), 5%RuEc (2.6%Ru) [47].
Commercial 5-HMF5-HMF from Cellulose Hydrolysate
CatalystConversion (%)Yield (%)Conversion (%)Yield (%)
No catalyst15.533.4NDND
5% Rucomm10052.41003.0
5% Pdcomm1006.4NDND
BimetallicDd10035.973.416.2
BimetallicEc10054.477.624.1
5% RuE.c1001172.619.1
5% PdEc10051.1NDND
Table
Table 3. Cellulose hydrolysis, fermentation and hydrogenation yields of products. Glucose and 5-HMF yields from cellulose via hydrothermal hydrolysis are reported previously in this study. 5-HMF extraction in MTHF, efficiency and yields are from [48,53]. H2-dark fermentation yields are reported in this study and from [48]. H2-photo-fermentation H2 yields are from [48,49,56] and R.L. Orozco (unpublished work). Catalytic hydrogenation of 5-HMF to 2,5 DMF yields are taken as per in [57]. Hydrothermal Hydrolysis (HTH); Extraction Efficiency (E.E.); Dark Fermentation (D.F.); Photo-fermentation (P.F.); Cellulose (C); Commercial (commer.). STDEV with n = 3 (number of experiments). Values without STDEV are average of two experiments.
Table 3. Cellulose hydrolysis, fermentation and hydrogenation yields of products. Glucose and 5-HMF yields from cellulose via hydrothermal hydrolysis are reported previously in this study. 5-HMF extraction in MTHF, efficiency and yields are from [48,53]. H2-dark fermentation yields are reported in this study and from [48]. H2-photo-fermentation H2 yields are from [48,49,56] and R.L. Orozco (unpublished work). Catalytic hydrogenation of 5-HMF to 2,5 DMF yields are taken as per in [57]. Hydrothermal Hydrolysis (HTH); Extraction Efficiency (E.E.); Dark Fermentation (D.F.); Photo-fermentation (P.F.); Cellulose (C); Commercial (commer.). STDEV with n = 3 (number of experiments). Values without STDEV are average of two experiments.
Cellulose Hydrolysis and Fermentation YieldsCatalytic Hydrogenation of 5-HMF to 2,5 DMF Yields
HTH Products5-HMF Extraction in MTHF (1:1)D.F. YieldsP.F. Yieldg DMF/g 5HMFmol DMF/Kg C
(mol glucose/Kg C)mol 5-HMF/Kg CE.E. (%)mol 5-HMF/Kg Cmol H2/Kg Cmol OA mix/Kg Cmol H2/mol OA mixmol H2/Kg C5% Ru Commer.5% Pd/5% Ru E. coli5% Pd/5% Ru E. coli
1.250.127650.0830.45 ± 0.061.33 ± 0.174.05 ± 0.335.40 ± 0.430.0230.1830.159
Table
Table 4. The energy balance of this work consisting of the energy input for the cellulose hydrolysis and hydrogenation reactions compared to the energy output obtained from the bio-H2 yields of dark and photo-fermentation and 2,5 DMF obtained from the hydrogenation reaction. Dark Fermentation (D.F.); Photo-fermentation (P.F.). All calculations in this table were based on average values from experiments as described in this study.
Table 4. The energy balance of this work consisting of the energy input for the cellulose hydrolysis and hydrogenation reactions compared to the energy output obtained from the bio-H2 yields of dark and photo-fermentation and 2,5 DMF obtained from the hydrogenation reaction. Dark Fermentation (D.F.); Photo-fermentation (P.F.). All calculations in this table were based on average values from experiments as described in this study.
Reactor 1 (R1: Hydrolysis of Cellulose to 5-HMF @ 250 °C)Reactor 2 (R2: Hydrogenation of 5-HMF to 2,5 DMF @ 260 °C) Energy Output (kW·h/Kg Cellulose)
Cell. Load (Kg)Water (kg)Energy Input kW·h/Kg Cellulose5-HMF in Reactor (g)MTHF (Kg)Energy Input kW·h/Kg CelluloseTotal Energy Input in R1 and R2H2 from D.F.H2 from P.F.2,5 DMF from R2Total Energy Output
0.0080.1254.560.2500.04273.618.180.0310.3630.1430.537
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Gomez-Bolivar, J.; Orozco, R.L.; Stephen, A.J.; Mikheenko, I.P.; Leeke, G.A.; Merroun, M.L.; Macaskie, L.E. Coupled Biohydrogen Production and Bio-Nanocatalysis for Dual Energy from Cellulose: Towards Cellulosic Waste Up-Conversion into Biofuels. Catalysts 2022, 12, 577. https://doi.org/10.3390/catal12060577

AMA Style

Gomez-Bolivar J, Orozco RL, Stephen AJ, Mikheenko IP, Leeke GA, Merroun ML, Macaskie LE. Coupled Biohydrogen Production and Bio-Nanocatalysis for Dual Energy from Cellulose: Towards Cellulosic Waste Up-Conversion into Biofuels. Catalysts. 2022; 12(6):577. https://doi.org/10.3390/catal12060577

Chicago/Turabian Style

Gomez-Bolivar, Jaime, Rafael L. Orozco, Alan J. Stephen, Iryna P. Mikheenko, Gary A. Leeke, Mohamed L. Merroun, and Lynne E. Macaskie. 2022. "Coupled Biohydrogen Production and Bio-Nanocatalysis for Dual Energy from Cellulose: Towards Cellulosic Waste Up-Conversion into Biofuels" Catalysts 12, no. 6: 577. https://doi.org/10.3390/catal12060577

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop