Elucidating the role of ferrous ion cocatalyst in enhancing dilute acid pretreatment of lignocellulosic biomass
© Wei et al; licensee BioMed Central Ltd. 2011
Received: 5 July 2011
Accepted: 10 November 2011
Published: 10 November 2011
Recently developed iron cocatalyst enhancement of dilute acid pretreatment of biomass is a promising approach for enhancing sugar release from recalcitrant lignocellulosic biomass. However, very little is known about the underlying mechanisms of this enhancement. In the current study, our aim was to identify several essential factors that contribute to ferrous ion-enhanced efficiency during dilute acid pretreatment of biomass and to initiate the investigation of the mechanisms that result in this enhancement.
During dilute acid and ferrous ion cocatalyst pretreatments, we observed concomitant increases in solubilized sugars in the hydrolysate and reducing sugars in the (insoluble) biomass residues. We also observed enhancements in sugar release during subsequent enzymatic saccharification of iron cocatalyst-pretreated biomass. Fourier transform Raman spectroscopy showed that major peaks representing the C-O-C and C-H bonds in cellulose are significantly attenuated by iron cocatalyst pretreatment. Imaging using Prussian blue staining indicated that Fe2+ ions associate with both cellulose/xylan and lignin in untreated as well as dilute acid/Fe2+ ion-pretreated corn stover samples. Analyses by scanning electron microscopy and transmission electron microscopy revealed structural details of biomass after dilute acid/Fe2+ ion pretreatment, in which delamination and fibrillation of the cell wall were observed.
By using this multimodal approach, we have revealed that (1) acid-ferrous ion-assisted pretreatment increases solubilization and enzymatic digestion of both cellulose and xylan to monomers and (2) this pretreatment likely targets multiple chemistries in plant cell wall polymer networks, including those represented by the C-O-C and C-H bonds in cellulose.
Keywordsdilute acid pretreatment iron cocatalyst ferrous ions metal cocatalyst biomass cellulose corn stover cotton linter filter paper Fourier transform Raman spectroscopy
Enzymatic biomass conversion enabled by dilute acid pretreatment processes has been studied for many years but remains one of the key obstacles to the economical production of lignocellulosic biofuels today. Ferrous ion (hereinafter referred as Fe2+) enhancement of dilute acid pretreatment of biomass is a promising technology that enhances the release or conversion of sugars during pretreatment . The economic benefit of adding Fe2+ ions can be realized by reducing the severity of pretreatment conditions (a composite factor based on acid concentration, temperature and time) while retaining comparable conversion to biomass sugars.
Fourier transform infrared spectroscopy (FTIR) was used to study the interaction between first-row transition metal ions (to which Fe belongs) and glucose in the glassy state. The results showed that all bands of sugar skeletal vibration modes and all C-O-H and C-O vibration modes of glucose were shifted in metal ion/D-glucose complexes . In another study, FTIR spectra also showed that the metal ions (including Fe2+ ions) induced changes in the C-C and C-O ring as well as in the skeletal modes of xylose .
Researchers in recent studies have reported that the ferric form of iron (FeCl3) also is efficient in functioning as a catalyst for releasing hemicellulose from corn stover (CS) by hot water pretreatment  and that the enzymatic digestibility and cellulose recovery can be enhanced after hot ethanosolv/FeCl3 pretreatment of barley straw . Most recently, it was reported that Fe2+ and Fe3+ ions are effective in enhancing the Lewis acid cocatalyzed dilute sulfuric acid pretreatment of lignocellulosic biomass .
However, the precise mechanism underpinning the facilitation of biomass deconstruction as catalyzed by iron ions, either in Fe2+ ion form [1, 6] or in ferric ion form (hereinafter referred to as Fe3+) [4–6], remains unknown. Adding to the technical difficulty of studying the role of Fe2+ ions in biomass pretreatment is its transient nature. Fe2+ ions can easily be oxidized to Fe3+ ions by exposure to air. Thus continuous argon or nitrogen gas needs to be used to purge the Fe2+ ion-containing solutions until it is transferred to a sealed container, such as reactors used in pretreatment. Consequently, there is a scarcity of information about which components of biomass, a complex matrix of celluloses, hemicelluloses and lignins, are affected by iron cocatalysts and which types of chemical bonds are actively engaged during catalytic deconstruction.
The aim of our research was to identify the factors that may contribute to metal-enhanced efficiency during dilute acid/Fe2+ ion pretreatment and initiate the exploration of their mechanisms by using model cellulose substrates (filter paper (FP) and cotton linter (CL)) as well as model biomass feedstock (corn stover). To achieve this goal, we employed high-performance liquid chromatography (HPLC), dinitrosalicylic acid (DNS) assay, Fourier transform (FT) Raman spectroscopy, Prussian blue iron staining, laser dissection microscopy, transmission electron microscopy (TEM) and scanning electron microscopy (SEM) to study dilute acid/Fe2+ ion pretreatment and its effect on the digestibility of biomass residues.
Biophysical roles of Fe2+ions in cellulose pretreatment: precipitation patterns of pretreated filter paper
Meanwhile, the FP strips in tube 4 were pretreated with 5 mM Fe2+ in ddH2O, and the formation of red ferric oxide (Fe2O3, that is, rust) observed throughout the FP strips, especially at the edges of the strips, is likely due to the oxidation of Fe2+ to Fe3+ ions (Figure 2, tube 4).
The difference in precipitation volume and appearance of cellulose residues may indicate that FeSO4 physicochemically modifies the remaining FP in a way that leads to changes in the surface charge and/or hydrophobicity of FP during dilute acid/metal pretreatment.
Sugar released during pretreatment of filter paper and cotton linter
We assessed three aspects of FP and CL pretreatment efficiency: (1) how much sugar was released, (2) the chemical and physicochemical changes in the remaining FP and CL and (3) the enzymatic digestibility of the remaining FP and CL. HPLC was used to measure the released sugars in the hydrolysate liquors of pretreated samples.
Quantification of reducing sugar ends in remaining filter paper and cotton linter
The generation of reducing sugar ends from filter paper and cotton linter after dilute acid and/or metal pretreatments
Reducing sugar endsa
Average increase or decrease over control
FP control (0.5% H2SO4)
1.92 ± 0.07
FP (5 mM FeSO4 in ddH2O)
0.88 ± 0.05
FP (0.5% H2SO4 + 0.5 mM FeSO4)
2.47 ± 0.13
FP (0.5% H2SO4 + 5 mM FeSO4)
2.32 ± 0.16
CL control (0.5% H2SO4)
1.68 ± 0.06
CL (5 mM FeSO4 in ddH2O)
1.04 ± 0.05
CL (0.5% H2SO4 + 0.5 mM FeSO4)
2.23 ± 0.13
CL (0.5% H2SO4 + 5 mM FeSO4)
2.17 ± 0.15
Enzymatic digestibility of the remaining filter paper and cotton linter
Effect of Fe2+ions on cellulose degree of polymerization in acid pretreatment
Effect of iron catalyst on the degree of polymerization of Solka-Floc cellulose after various pretreatments
Weighted average degree of polymerization
Untreated Solka-Floc cellulose (control)
Future studies are needed to confirm the DP measurement of the model cellulose materials, FP and CL. These DP data will be helpful for exploring whether there is any correlation between the DP and the measured reducing sugar ends in the pretreated FP and CL residues (as described in earlier section, "Quantification of reducing sugar ends in remaining filter paper and cotton linter".
Fourier transform Raman spectra of pretreated model cellulose substrates
Frequency and characterization of major Fourier transform Raman peaks in spectra of dilute acid-alone pretreated filter paper and cotton linter samples were flattened in spectra of dilute acid/ferrous ion-pretreated samples
Pretreatment regions in four studies (peaks/cm)
Fechner et al.
δ (CH2) scissors
δ (CH2) wagging, δ (C-O-H)
δ (CH2) twisting
γ (C-O-C) glycosidic bond, symmetric stretching
γ (C-O-C) glycosidic bond, symmetric stretching
519, 457, 438, 380, 341
Assignments of major FT Raman spectral peaks that were different between dilute acid alone-pretreated versus dilute acid/Fe2+ ion-pretreated FP are given in Table 3. As illustrated in Figures 5A and 5B and as listed in Table 3 our FT Raman spectroscopy results suggest that the following bond types are likely to be the targets for the ferrous ions during the dilute acid/Fe2+ ion pretreatment of FP and CL. First, the C-O-C glycoside link between the D-glucose units in cellulose covers the region of 1, 120/cm and 1, 097/cm, ascribed to the C-O-C glycosidic link symmetric and asymmetric stretching modes, respectively. These belong to backbone stretching vibrations. Second, the C-H glycoside bond covers the region of 1, 485/cm to 1, 270/cm as well as 2, 895/cm, corresponding to the C-H bending, stretching, twisting or wagging.
These observations are partially consistent with, and extend the knowledge gained from, a previous FTIR study in which bands of sugar skeletal C-O vibration modes of glucose were affected by metal ion/D-glucose complexes . Furthermore, most bands in the spectra of the dilute acid/Fe2+ ion-pretreated FP were broader than the bands in the spectra of dilute acid alone- or ferrous ion alone-pretreated bands, suggesting that the former contain a higher amount of amorphous regions . Another indicator of the change in the cellulose crystalline structure comes from specific bands of 1, 160/cm, 1, 120/cm and 1, 097/cm, which have been used to trace changes in cellulose crystallinity [12, 13]. Our results demonstrate that the bands of 1, 120/cm and 1, 097/cm were attenuated in dilute acid/Fe2+ ion-pretreated samples compared to those from dilute acid alone pretreatment. Usually, the ranges of crystallinity for FP and CL are 40% to 50% and 65% to 80%, respectively . We suggest that monitoring the possible effects of Fe2+ ions on the crystallinity of cellulose during pretreatment should be considered in future studies.
It is also noteworthy that there is an underexplored range of 550/cm to 300/cm in the FT Raman spectra of cellulose samples. We observed a cluster of peaks in the range of 550/cm to 300/cm in dilute acid alone-pretreated FP (Figure 5B). This cluster of peaks, listed in Table 3 (bottom rows: 519/cm, 438/cm, 380/cm and 341/cm), is in agreement with similar previously published observations in microcrystalline and powdered cellulose . Because these peaks were attenuated by dilute acid/Fe2+ ion pretreatment, they are likely linked to the effects of the Fe2+ ion cocatalyst on FP and CL during pretreatment. However, the scarcity of literature about the assignment of these bands to the structure of cellulose necessitates further investigation before well-supported conclusions may be drawn. Parallel studies were also conducted in which FT Raman spectra of CL after various pretreatments were used. The spectra in Figure 5C show that the CL was affected similarly to the FP (Figure 5B) under both control (acid alone) and other pretreatments.
We performed an additional FT Raman experiment on the non-pre-treated FP and CL. Their spectra, shown in Figure 5D, are similar to the dilute acid alone-pretreated FP and CL in Figures 5A through 5C, confirming the unique effect of dilute acid/Fe2+ ion pretreatment in attenuating several major peaks in the FT Raman spectra of FP and CL. Furthermore, the FT Raman spectra of non-pre-treated FP and CL shown in Figure 5D are similar to the Raman spectra of Sigma cellulose powder (Sigma-Aldrich, St Louis, MO, USA) obtained from CL in a recent study , thereby partly validating our spectral data.
Imaging plant cell wall components of native, non-pre-treated corn stover in the presence of Fe2+ions
In addition to investigating the biochemical and spectral effects of Fe2+ ions on biomass conversion, we applied different imaging analysis approaches to investigate the direct interaction between metal ions and their biomass targets. We began these imaging studies by examining the binding and interaction of Fe2+ ions with native, non-pre-treated CS. We hypothesized that Fe2+ ions are likely to be bound with or adsorbed into one or multiple components of plant cell walls. As we carried out the Prussian blue staining assay, we set out to determine whether Fe2+ ions are bound with or adsorbed into cellulose or lignin or both.
In contrast, Figure 6B, i and iv, show the bright-field images obtained with Prussian blue staining of native CS samples after incubation with 0.5 mM FeSO4 solution followed by washing. This may indicate the distribution of Fe2+ ions following impregnation by soaking biomass in iron and/or acid prior to pretreatment. The Prussian blue staining is obvious and intense (compared with the control of no Fe2+ incubation; Figure 6A, i and iv), showing the reaction with Fe2+ ions present on the surface.
Furthermore, analysis of the other panels in Figure 6B reveals the differential deposition of Fe2+ ions on the various components of plant cell walls in CS. First, a parallel examination of the Prussian blue staining in the bright field (Figure 6B, i and iv) and the red autofluorescence of lignin (Figure 6B, ii and v) shows that the dense staining of Prussian blue aligns with the location of lignin in the vascular bundle area, which is indicated by a rectangle in Figure 6B, i, and shown at high resolution in Figure 6B, iv. Second, the merging of the bright-field image and the red autofluorescence of lignin demonstrates that cellulose/xylan (located where there is very weak or no lignin red autofluorescence, as indicated by the white arrow in the right column of Figure 6B, iii and vi) also adsorbs Fe2+ ions. Thus these results indicate that Fe2+ ions associate with both cellulose/xylan and lignin in natural, untreated CS. Note that the above analysis is based on the assumption that where there is blue color but no red color, there is likely to be cellulose/xylan. Figure 6B shows the representative differential binding of Fe2+ with lignin versus cellulose/xylan, and the images presented do not reflect the overall proportionate amount of cellulose/xylan in the CS biomass.
It is noteworthy that Prussian blue staining has been utilized for subcellular localization of bound iron ions in various organisms, including plants [15, 16], but not to date in thermochemically pretreated biomass. Our current research is the first attempt to localize the metal ion in dilute acid/Fe2+ ion-pretreated biomass.
Imaging precipitation of Fe ions on the biomass surface following pretreatment
The pattern of Fe precipitation may provide clues to the role of iron salts in biomass pretreatment. Thus Prussian blue staining was conducted for detection of Fe ion deposition on pretreated CS. Because the pretreated biomass was not washed before being stained, the localization of Fe ions observed by Prussian blue staining is likely to reflect the in situ association of Fe with biomass, especially on the biomass surface.
Transmission electron microscopy of dilute acid/Fe2+ion cocatalyst-pre-treated corn stover
Scanning electron microscopy revealing surface ultrastructure of biomass
Overall, dilute acid-pre-treated and dilute acid/Fe2+ ion-pretreated CS samples followed a similar structural modification in the vascular tissue, as shown in Figures 9A, B, D and 9E). However, dilute acid-pre-treated CS controls (top panels in Figures 9A through 9C) have smooth surfaces. The inside and outside surfaces of the pretreated samples appear smooth and littered with small to large droplets of lignin. In contrast, the surfaces of dilute acid/Fe2+ ion-pretreated CS samples (lower panels in Figures 9D through 9F) appear to have a large percentage of the matrixing material (hemicelluloses) removed, with the cellulose macrofibers intact but floating above the surface in a delicate, lacelike pattern with interspersed lignin droplets (lower panels in Figures 9F and 9G). Lignin droplets are interspersed throughout the lacelike lattice as well as on the surface of the pretreated biomass sample (lower panels in Figures 9F and 9G).
In Figure 9G, a high-magnification image (×100, 000 original magnification) of a section of Figure 9F, the delicate, lacelike pattern of the putative cellulose macrofibers is readily apparent. Small lignin droplets are interspersed throughout the lacelike pattern. This removal of additional hemicellulose from the dilute acid/Fe2+ cocatalyst-pre-treated CS may account for the enhancement in hemicellulose (xylose) yield following dilute acid/Fe2+ cocatalyst pretreatment of CS compared to the control dilute acid alone pretreatment (Chen X, Shekiro J, Kuhn E, Wang W and Tucker MP, unpublished data). In addition, the lacelike fibrillation of the biomass surface following dilute acid/Fe2+ cocatalyst pretreatment of CS may account for the enhancement in enzymatic digestion of the pretreated substrate compared with the digestion of dilute acid alone-pretreated CS controls (Chen X, Shekiro J, Kuhn E, Wang W and Tucker MP, unpublished data).
Fe2+ions and cellulose component of biomass
In the present study, we investigated the effects and the role of Fe2+ ions in the dilute acid pretreatment of FP and CL. As indicated by Raman spectroscopy, the Fe2+ ions during pretreatment affect C-O bonds in polysaccharide backbone as well as the (-CH2OH) side groups of cellulose. These factors contribute to the depolymerization of cellulose and the reduction of its DP, eventually leading to the release of more simple sugars and the generation of more reducing sugar ends as well as making the biomass residues more digestible by enzymes.
Fe2+ions and lignin component of biomass
Biochemical and imaging analyses revealed that, in addition to cellulose, Fe2+ ions may interact with other plant cell wall components. Prussian blue staining was utilized to visualize the presence of and distribution of iron on the surface of the biomass, whereas SEM and TEM were carried out to reveal the structural details of the biomass after pretreatment. The observations made with Prussian blue staining suggest that Fe2+ ions may remain in close proximity to heavily lignified regions of the biomass during pretreatment, and structural analysis by SEM and TEM revealed that these same regions are strongly affected morphologically by the presence of the Fe cocatalyst.
The formation and migration of spherical lignin deposits on the outer surfaces of fibers during pretreatment of CS have been described previously [17, 18]. Our observation of smaller and more dispersed lignin droplets after dilute acid/Fe2+ ion pretreatment suggests a possible role of iron in affecting the pattern of lignin droplet formation and migration during pretreatment. Further quantitative study is needed to confirm whether the production of smaller lignin droplets is another key contributing factor (in addition to the removal of xylan) that leads to better exposure of the structural carbohydrates that remain in the residual solid material after dilute acid/Fe2+ iron pretreatment, facilitating the subsequent enzymatic digestion.
Potential application of Fe2+ions in biomass pretreatment
First, although the results of this study are derived mainly from FP, CL and CS (which is derived from corn and/or maize, a monocot and a member of the grass family), we believe that the results we present are also applicable to woody biomass, such as poplar biomass. In fact, the initial report of Fe2+ ion enhancement of biomass pretreatment was based on a woody plant biomass, the Douglas fir . Second, in addition to the Fe2+ ions, other transitional metal ions, such as Mn2+ and Cu+, may have similar effects by enhancing biomass conversion during dilute acid pretreatment and are worth exploring. Third, in this study, we used gold-coated stainless steel reactors for the pretreatment of model cellulose materials and CS. This may have increased the cost of equipment required for using this technology at an industrial scale. Alternatively, a HASTELLOY corrosion-resistant alloy reactor (Haynes International, Inc, Kokomo IN, USA) may be an attractive option because it is resistant to sulfuric acid. Although a HASTELLOY C corrosion-resistant alloy reactor is 50% more expensive than a stainless steel SS316 reactor vessel , when the cost of the gold coating and, likely, recoating for stainless steel reactors are taken into consideration, a HASTELLOY corrosion-resistant alloy reactor seems more economical than gold-coated stainless steel in the long run.
To prevent the Fe2+ ions from being oxidized to Fe3+ ions prior to the FP, CL and biomass pretreatments, argon gas was used to purge the dilute acid/Fe2+ solution. This argon-purging step may be a hurdle to clear for application at an industrial scale. The use of other forms of metal ions, such as Fe3+, that remain stable with exposure to air (oxygen) have been studied previously and may be an alternative worthy of further exploration.
We employed a range of biochemical, spectroscopic and imaging techniques to investigate the impact of dilute acid/Fe2+ ion pretreatment on the resultant digestibility of model cellulose substrates. By using a Prussian blue staining technique, we successfully observed the pattern of Fe2+ binding with the cell wall components of native, non-pre-treated CS as well as the iron distribution after dilute acid/Fe2+ ion pretreatment. Our imaging and FT Raman data suggest that the presence of ferrous ions during sulfuric acid pretreatment may affect multiple components of the plant cell wall, including the C-O-C and C-H bonds in cellulose. TEM and SEM imaging data suggest that ferrous ions, during pretreatment of CS, may improve xylan removal and lignin relocation. The comprehensive analyses presented in this article expand our understanding of the effects of dilute acid/Fe2+ ion pretreatment; however, further investigation is needed to elucidate the detailed chemical mechanisms of this process, which will enable the optimization of this pretreatment strategy.
An overview of the experimental approach we used, designed to investigate the metal ion-biomass interaction and effects on biomass conversion, is illustrated schematically in Figure 1.
Model cellulose and biomass feedstock materials
Three types of model celluloses were used in current study. First, FP disks (150 mm in diameter, Whatman grade 1; Whatman Inc, Piscataway, NJ, USA) were cut into 1-cm × 6-cm strips and further trimmed to a final weight of 50.0 ± 0.1 mg for each strip. Second, acid-free and 99.5% lignin-free CL sheets (catalog no. 216; Arnold Grummer's Paper Making, Milwaukee, WI, USA) were cut into small disks using a 9-mm cork borer. The thick 2-mm disks were teased apart into thin disks, approximately 0.1 mm thick, to allow rapid and even diffusion of acid and cocatalyst throughout the CL disks. Third, Solka-Floc powdered cellulose 200 FCC was especially suitable for the DP studies because of its uniform, fine powder nature, which permitted reproducibility of DP measurements (described in the section of "Effect of Fe2+ ions on cellulose degree of polymerization in acid pretreatment"). The CS used as our model biomass feedstock (variety 33A14; Pioneer Hi-Bred International, Inc, Johnston, IA, USA) was harvested at the Kramer Farm in Wray, Colorado. The whole CS was tub-ground at the site and further milled at NREL through a Mitts & Merrill rotary knife mill (model 10 × 12; Mitts & Merrill, Harvard, IL, USA) to pass through a ¼-inch screen.
Pretreatments of model cellulose materials using pipe reactors
One-inch Swagelok 316L stainless steel tubing unions with two caps (Swagelok Co, Solon, OH, USA) were utilized as 20-ml pipe reactors. The inside surface of each reactor was plated with a very inert 99.97% gold coating at approximately 1-μm thickness to resist corrosion. It has previously been reported in the literature that chromium, molybdenum and a wide range of metal ions catalyze the destruction of sugars in solution . The inert gold surface was used to prevent destructive catalysis by these metal ions and to ensure that any enhancements observed in pretreatment were the sole result of the cocatalyst incorporation during dilute acid pretreatments. The total volume of each pipe reactor was 20 ml, with minimal headspace remaining when the caps were tightened in place. For dilute acid/Fe2+ ion cocatalyst pretreatment, the 0.5 wt% H2SO4 was prepared beforehand, then the acid solution was sparged with argon prior to adding the FeSO4 cocatalyst. The dilute sulfuric acid/Fe2+ cocatalyst solution was added to the 250 mg of FP, CL or Solka-Floc powdered cellulose. We sparged the plated reactors with argon prior to adding the acid solution and sealing the reactors. The pipe reactors were immersed in a fluidized sand bath at the desired reaction temperature, and the residence time at the desired reaction temperature for the pretreatment was started when a separate control reactor equipped with a thermocouple came within 5°C of the desired reaction temperature.
After the pretreatments, no corrosion of the gold surface was found upon inspection of the reactors. The digestion mixture was poured out into preweighed 50-ml BD Falcon tubes (BD Biosciences, Franklin Lakes, NJ, USA). The tubes were spun in a centrifuge (model T-J-6; Beckman Coulter, Inc, Brea, CA, USA) at 3, 000 rpm for ten minutes. The supernatant was then transferred to a fresh tube, and the volume was measured. The supernatant (hydrolysate liquor) was used in the HPLC analysis to measure the released sugars, and the remaining FP and CL pellets were used in the DNS assay to measure the reducing sugar ends. For saccharification analysis, the remaining FP and CL pellets were washed three times with ddH2O and centrifuged between the washes. To observe the precipitation pattern of the remaining FP, ddH2O was added to the tubes containing the biomass residues. The tubes were then vortexed and left still overnight.
Acid impregnation of corn stover
For control experiments, acid impregnation was carried out using 120 L of warm (40°C to 50°C) 0.5 wt% H2SO4 contained in a 200-L recirculation tank. For dilute acid/Fe2+ ion cocatalyst impregnation, the acid was equilibrated for four hours in the recirculating bath, after which it was sparged with argon for 20 minutes and used to dissolve a measured amount of FeSO4. A HASTELLOY C-276 wire mesh basket (20-mesh screen; (Haynes International, Inc) was loaded with 14.5 kg of ¼-inch milled CS feedstock (approximately 94% solids) and lowered into the bath of warm dilute acid/Fe2+ ion cocatalyst for two hours. A seal around the top of the basket forced acid to recirculate through the bed of CS. Following acid impregnation, the feedstock in the basket was drained of excess acid to approximately 20% solids and loaded into the mold of a hydraulic dewatering press (also sparged with argon), where the acid-impregnated feedstock was pressed to about 45% solids. The dilute acid/Fe2+ ion cocatalyst-impregnated feedstock was prepared the same day as the pretreatment experiments were conducted to decrease the likelihood of oxidation of Fe2+ to Fe3+, and the impregnated feedstock was kept under argon.
Bench-scale pretreatment of corn stover
A 4-L steam explosion reactor made from HASTELLOY corrosion-resistant alloy was used for larger-scale high solids (approximately 45 wt% solids) pretreatment experiments. The reactor was prewarmed to pretreatment temperature and then sparged with argon for one minute using a lance that reached to the bottom of the hot reactor. Following sparging, the steam explosion reactor was loaded with 500.0 g of dilute acid/Fe2+ ion cocatalyst-impregnated and pressed feedstock (about 45% solids) and quickly heated (approximately five to ten seconds) by direct steam injection to the reaction temperature. At the desired reaction time, the pretreated feedstock was rapidly blown into a flash tank and rapidly depressurized to atmospheric pressure.
HPLC analysis for the sugars in hydrolysate liquors of pretreated filter paper and cotton linter
The pretreatment hydrolysate liquors were filtered through a 0.45-μm syringe filter prior to HPLC analysis for soluble carbohydrates. The filtrates were analyzed using a SP0810 column (Showa Denko America, Inc, New York, NY, USA) with a Micro-Guard De-Ashing column (Bio-Rad Laboratories, Hercules, CA, USA) at a flow rate of 0.6 ml/minute and a temperature of 80°C. The standards were used to identify the unknown HPLC peak areas by their retention time and to determine the sugar concentrations.
Dinitrosalicylic acid assay for reducing sugar ends in filter paper and cotton linter residues
The total reducing sugars in pretreated cellulose residues were measured by DNS assay . The assay was conducted on the scale of 0.2 g (dry weight basis) of washed sample of pretreated FP strips and CL disks. The washing step was performed to remove the iron ions or derived iron compounds from the FP and CL residues and to minimize any potential interference with the DNS assay. Because a glucose standard curve was used to calculate the sugar concentrations, the reducing sugars are expressed as "equivalent glucose."
Enzymatic saccharification of the filter paper and cotton linter residues
The FP and CL residues from the pretreatment experiments first went through the extensive washing procedure critical for the determination of fraction insoluble solids . Enzymatic digestions of washed, pretreated residues were then performed in 50 ml of 50 mM citrate buffer, pH 4.8, containing 1% substrate and 20 mg of GC 220 cellulase protein (Genencor, Rochester, NY, USA) per gram of substrate. The enzymatic reaction mixture was put into 125-ml Erlenmeyer shake flasks and incubated at 50°C and 130 rpm according to NREL LAP-009 . No additional β-glucosidase was added. Samples of the slurries were taken at 1, 24, 48, 72, 96 and 168 hours. The glucose released was measured by HPLC.
Determination of degree of polymerization in pretreated Solka-Floc cellulose
The molecular weight distributions of Solka-Floc cellulose samples were determined by SEC using a procedure modified from methods previously reported in the literature [24–26]. To permit dissolution in the SEC eluent tetrahydrofuran, cellulose samples were first carbanilated. Ten milligrams of a vacuum-dried cellulose sample were placed into a 5-ml reaction vial, and 2 ml of dry pyridine and 0.4 ml of phenylisocyanate were added. The reaction vial was kept at 70°C for 24 hours to complete the reaction. Methanol (0.4 ml) was added to the reaction mixture to react with the excess phenylisocyanate at the end of the reaction. The carbanilated cellulose was then precipitated in 26 ml of methanol/water (7:3 vol/vol), and the precipitate was washed twice with methanol/water (26 ml) before being dissolved in tetrahydrofuran (20 ml).
SEC was performed using five columns, 103, 104, 105, 106 and 107 Å (PLgel columns; Agilent Technologies, Santa Clara, CA, USA), to cover the broad range of molecular weights of the cellulose samples. A calibration curve was obtained that allowed the conversion of retention times into estimated molecular weights by running polystyrene standards of known average molecular weight on the SEC system. Consequently, all molecular weights determined in this work are not absolute, but are relative to the calibration curve. The conditions for chromatography were as follows: flow rate 1.0 ml/minute, UV detector wavelength 235 nm (with a bandwidth of 10 nm), injection volume 50 μl and column temperature 25°C. Cellulose DP was calculated by dividing the apparent molecular weights by 519 (molecular weight of a repeating unit of carbanilated cellulose with the degree of substitution of 3.0).
Fourier transform Raman spectroscopy and challenge in analyzing pretreated biomass
Lignin, which constitutes a significant portion of dilute acid-pre-treated CS (typically containing 27% to 31% lignin in dry weight) , poses a challenge while conducting FT Raman analysis on these samples. The lignin-caused dark coloration of pretreated CS may lead to the burning of biomass at high power levels. Alternatively, it may not generate enough signal at safe levels. In addition, our trial experiments in conducting FT Raman analysis of pretreated CS showed that the Raman spectra vary with different thickness of sample (which affects signal magnitude) and also for different areas of sample, which causes peak shifting (data not shown). For these reasons, we decided to use FP and CL as model celluloses instead of using biomass, such as CS, to conduct FT Raman analysis.
FT Raman data were collected using a Nicolet Raman 960 spectrometer (Nicolet Instrument Corp/Thermo Fisher Scientific, Madison, WI, USA) with 1, 064 nm of Nd:YVO4 laser excitation and a Ge detector. Samples were pressed onto glass coverslips and mounted on an upright holder. Measurement parameters included 256 scans at 8 cm-1 resolution with approximately 200-mW excitation power.
Prussian blue staining
The principle of Prussian blue staining is that in the presence of hydrochloric acid, iron ions will form a "Prussian blue" precipitate of ferric ferrocyanide in reaction with potassium ferrocyanide, the component of Prussian blue staining buffer. This blue precipitate will be deposited, and can be observed, in the cell wall sites where the iron ions are located.
We used Prussian blue staining first to detect the presence of Fe2+ ions in native, non-pre-treated CS and second to examine the deposition of Fe2+ ions on the different components of plant cell walls in dilute acid alone- versus dilute acid/Fe2+ ion-pretreated CS. For the first purpose, dry, native, non-pre-treated CS samples were wetted and hand-sectioned, then quickly transferred to either 0.5 wt% H2SO4 (for control) or 0.5 wt% H2SO4 + 1 mM Fe2SO4 and incubated in a shaker at room temperature for two hours before being subjected to the Prussian blue staining procedure.
For the second purpose, the dilute acid alone- versus dilute acid/Fe2+ ion-pretreated CS samples were directly hand-sectioned and then subjected to the following Prussian blue staining procedure. The hand-cut CS materials were washed twice in 200 mM NaCl and suspended in Prussian blue staining buffer (2% potassium ferrocyanide, 2% hydrochloric acid) for 45 minutes in the dark at room temperature, followed by one or two washes with distilled water. The stained materials were mounted on glass slides and imaged using a PALM laser dissection microscope (Carl Zeiss MicroImaging GmbH, Jena, Germany). Adobe Photoshop software (Adobe Systems, Inc, San Jose, CA) was used to overlay the Prussian blue-stained images with the autofluorescence images of lignin.
Laser dissection microscope
The PALM laser dissection inverted microscope was used for bright-field and fluorescence imaging. This system consists of a UV laser and an inverted microscope equipped with fluorescence optics, including filter sets for work with green, red or other fluorescent markers. The original magnification of the images was ×50, ×200 and ×400. The Prussian blue-stained dilute acid (control) and dilute acid/Fe2+ ion-pretreated CS biomass samples were placed in slides on the stage and illuminated using the bright-field and fluorescence optics. The fluorescence images were obtained where the samples were illuminated using the UV laser.
Scanning electron microscopic imaging
SEM imaging was performed using a Quanta™ 400 FEG high-resolution low-vacuum scanning electron microscope (FEI Co, Hillsboro, OR, USA) under low-vacuum (0.40 to 0.65 Torr) operating conditions with a gaseous solid-state detector. For SEM observations, the pretreated CS samples were prepared for imaging by freezing them in liquid nitrogen and then freeze-drying. This freeze-drying step was performed to prevent potential structural distortion that can be caused by the surface tension of liquid water during regular evaporation at atmospheric conditions. The samples were then mounted on aluminum stubs for SEM examination. Imaging was performed at beam-accelerating voltages ranging from 12.5 to 25 keV. Images were obtained at original magnifications ranging from ×500 to ×100, 000.
Transmission electron microscopy imaging
Samples were fixed twice for six minutes each with variable power in 2.5% glutaraldehyde buffered in 0.1 M PBS (Electron Microscopy Sciences, Hatfield, PA, USA) under vacuum pressure. The samples were dehydrated by treating them with increasing concentrations of ethanol for one minute at each dilution (30%, 60%, 90% and 3 × 100% ethanol). After dehydration, the samples were infiltrated using the EMbed 812 Resin for Microscopy kit (catalog no. 14120; Electron Microscopy Sciences) for three minutes under vacuum pressure with one step at room temperature overnight in increasing concentrations of resin (15%, 30%, 60%, 90% and 3 × 100% resin, diluted in ethanol). Infiltrated samples were transferred to flat-bottomed TAAB capsules (TAAB Laboratories Equipment Ltd, Aldermaston, UK) and polymerized in a nitrogen-purged vacuum oven at 70°C for 24 hours.
We sectioned embedded samples to approximately 60 nm by using a diamond knife (DiATOME US, Hatfield, PA, USA) on an Ultramicrotome Leica EM UC7 (Leica Microsystems, Wetzlar, Germany). Sections were collected on 0.5% Formvar-coated copper slot grids (SPI Supplies, West Chester, PA, USA). All grids were poststained for six minutes with 2% aqueous uranyl acetate and for three minutes with 1% KMnO4 for ten minutes to selectively stain for lignins. Images were taken with a 4-megapixel Gatan UltraScan 1000 camera (Gatan, Inc, Pleasanton, CA, USA) mounted on a Tecnai™ G2 20 Twin 200-kV LaB6 Transmission Electron Microscope (FEI Co).
distilled deionized water
dinitrosalicylic acid assay
degree of polymerization
- FT Raman spectroscopy:
Fourier transform Raman spectroscopy
high-performance liquid chromatography
size exclusion chromatography
scanning electron microscopy
transmission electron microscopy.
This research was funded by the US Department of Energy, Energy Frontier Research Centers (EFRC), the Center for Direct Catalytic Conversion of Biomass to Biofuels (C3Bio) under Award #DE-SC0000997. We thank Sara Havig for providing editing assistance.
- Nguyen QA, Tucker MP: Dilute acid/metal salt hydrolysis of lignocellulosics. US Patent 6423145. 2002Google Scholar
- Xi N, Xu ZH, Wu JG, Xu GX: Interaction of metal ions with D-glucose in a glassy state: an FTIR study. Proc SPIE Vol 1145/HC, 7th Intl Conf on Fourier Transform Spectroscopy. Edited by: Cameron DG. 1989, 405-406.View ArticleGoogle Scholar
- Bruni P, Iacussi M, Tosi G: FT-IR investigation on interactions between sugars and metal ions - I. J Mol Struct. 1997, 408-409: 171-176.View ArticleGoogle Scholar
- Liu L, Sun J, Cai C, Wang S, Pei H, Zhang J: Corn stover pretreatment by inorganic salts and its effects on hemicellulose and cellulose degradation. Bioresour Technol. 2009, 100: 5865-5871. 10.1016/j.biortech.2009.06.048.View ArticleGoogle Scholar
- Kim Y, Yu A, Han M, Choi GW, Chung B: Ethanosolv pretreatment of barley straw with iron(III) chloride for enzymatic saccharification. J Chem Technol Biotechnol. 2010, 85: 1494-1498.Google Scholar
- Degenstein JC, Tucker MP, Ji Y: Lewis acid co-catalyzed dilute sulfuric acid pretreatment of lignocellulosic biomass [abstract 2-04]. 33rd Symposium on Biotechnology for Fuels and Chemicals: Seattle, WA, USA, 2 May 2011. 2011, Fairfax, VA Society for Industrial MicrobiologyGoogle Scholar
- Gao Y, Kim Y, Thevuthasan S, Chambers S, Lubitz P: Growth, structure, and magnetic properties of γ-Fe2O3 epitaxial films on MgO. J Appl Phys. 1997, 81: 3253-3256. 10.1063/1.364355.View ArticleGoogle Scholar
- El-Shall MS, Li S, Graiver D, Pernisz U: Synthesis of nanostructured materials using a laser vaporization-condensation technique. Nanotechnology: Molecularly Designed Materials. Edited by: Chow GM, Gonsalves KE. 1996, Washington, DC: American Chemical Society, 622: 79-101. [ACS Symp Ser Am Chem Soc]View ArticleGoogle Scholar
- Su Y, Gathitu BB, Chen WY: Efficient and cost effective reburning using common wastes as fuel and additives. Fuel. 2010, 89: 2569-2582. 10.1016/j.fuel.2009.12.009.View ArticleGoogle Scholar
- Laxman RS, Lachke AH: Bioethanol from lignocellulosic biomass: Part I pretreatment of the substrates. Handbook of Plant-Based Biofuels. Edited by: Pandey A. 2008, Boca Raton, FL: CRC Press, 121-139.Google Scholar
- Fechner P, Wartewig S, Füting M, Heilmann A, Neubert R, Kleinebudde P: Properties of microcrystalline cellulose and powder cellulose after extrusion/spheronization as studied by Fourier transform Raman spectroscopy and environmental scanning electron microscopy. AAPS PharmSci. 2003, 5: E31-View ArticleGoogle Scholar
- Edwards HGM, Farwell DW, Williams AC: FT-Raman spectrum of cotton: a polymeric biomolecular analysis. Spectrochim Acta A Mol Biomol Spectrosc. 1994, 50: 807-811. 10.1016/0584-8539(94)80016-2.View ArticleGoogle Scholar
- Liu Y: Vibrational spectroscopic investigation of Australian cotton cellulose fibres Part 1. A Fourier transform Raman study. Analyst. 1998, 123: 633-636. 10.1039/a707064k.View ArticleGoogle Scholar
- Szymańska-Chargot M, Cybulska J, Zdunek A: Sensing the structural differences in cellulose from apple and bacterial cell wall materials by Raman and FT-IR spectroscopy. Sensors (Basel). 2011, 11: 5543-5560. 10.3390/s110605543.View ArticleGoogle Scholar
- Liu GS, Greenshields DL, Sammynaiken R, Hirji RN, Selvaraj G, Wei YD: Targeted alterations in iron homeostasis underlie plant defense responses. J Cell Sci. 2007, 120: 596-605. 10.1242/jcs.001362.View ArticleGoogle Scholar
- Vasconcelos M, Datta K, Oliva N, Khalekuzzaman M, Torrizo L, Krishnan S, Oliveira M, Goto F, Datta SK: Enhanced iron and zinc accumulation in transgenic rice with the ferritin gene. Plant Sci. 2003, 164: 371-378.View ArticleGoogle Scholar
- Selig M, Viamajala S, Decker S, Tucker M, Himmel M, Vinzant T: Deposition of lignin droplets produced during dilute acid pretreatment of maize stems retards enzymatic hydrolysis of cellulose. Biotechnol Progr. 2007, 23: 1333-1339. 10.1021/bp0702018.View ArticleGoogle Scholar
- Donohoe B, Decker S, Tucker M, Himmel M, Vinzant T: Visualizing lignin coalescence and migration through maize cell walls following thermochemical pretreatment. Biotechnol Bioeng. 2008, 101: 913-925. 10.1002/bit.21959.View ArticleGoogle Scholar
- van Walsum GP, Jayawardhana K, Yourchisin D, McWilliams R, Lam J, Miller C, Finch L, Castleberry V: Final Technical Report on Carbonic Acid Pretreatment of Biomass (Department of Energy Contract DE-FC36-01GO11070, A000). 2003, Waco, TX: Department of Environmental Studies, Baylor University, [http://www.osti.gov/bridge/servlets/purl/882226-qwcYKD/882226.pdf]View ArticleGoogle Scholar
- Corma A, Iborra S, Velty A: Chemical routes for the transformation of biomass into chemicals. Chem Rev. 2007, 107: 2411-2502. 10.1021/cr050989d.View ArticleGoogle Scholar
- Miller G: Use of dinitrosalicylic acid reagent for determination of reducing sugar. Anal Chem. 1959, 31: 426-428. 10.1021/ac60147a030.View ArticleGoogle Scholar
- Sluiter A, Hyman D, Payne C, Wolfe J: Determination of Insoluble Solids in Pretreated Biomass Material: Laboratory Analytical Procedure (LAP) (Technical Report NREL/TP-510-42627). 2008, Golden, CO: National Renewable Energy Laboratory, [http://www.nrel.gov/biomass/pdfs/42627.pdf]Google Scholar
- Brown L, Torget R: Enzymatic Saccharification of Lignocellulosic Biomass (Chemical and Testing Task: Laboratory Analytical Procedure LAP-009). 1996, Golden, CO: National Renewable Energy Laboratory, [http://www.p2pays.org/ref/40/39116.pdf]Google Scholar
- L'Ecuyer N, Robert S, Lapierre L, Methot M, Bouchard J: Determination of degree of polymerization distribution of Canadian Kraft and sulphite pulps: problems and solutions. 59th Appita Annual Conference and Exhibition: Incorporating the 13th ISWFPC (International Symposium on Wood, Fibre and Pulping Chemistry), Auckland, New Zealand, 16 to 19 May 2005: Proceedings. 2005, Carlton, Vic, Australia: Appita Inc, 217-219.Google Scholar
- Evans R, Wearne RH, Wallis AFA: Molecular weight distribution of cellulose as its tricarbanilate by high performance size exclusion chromatography. J Appl Polym Sci. 1989, 37: 3291-3303. 10.1002/app.1989.070371202.View ArticleGoogle Scholar
- Mormann W, Michel U: Improved synthesis of cellulose carbamates without by-products. Carbohydr Polym. 2002, 50: 201-208. 10.1016/S0144-8617(02)00016-4.View ArticleGoogle Scholar
- Schell D, Farmer J, Newman M, McMillan J: Dilute-sulfuric acid pretreatment of corn stover in pilot-scale reactor. Appl Biochem Biotechnol. 2003, 105: 69-85. 10.1385/ABAB:105:1-3:69.View ArticleGoogle Scholar
- Edwards H, Ellis E, Farwell D, Janaway R: Preliminary study of the application of Fourier transform Raman spectroscopy to the analysis of degraded archaeological linen textiles. J Raman Spectrosc. 1996, 27: 663-669. 10.1002/(SICI)1097-4555(199609)27:9<663::AID-JRS11>3.0.CO;2-E.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.