Elsevier

Biomass and Bioenergy

Volume 119, December 2018, Pages 284-292
Biomass and Bioenergy

Research paper
Catalytic production of sugars and lignin from agricultural residues using dilute sulfuric acid in γ-valerolactone

https://doi.org/10.1016/j.biombioe.2018.09.031Get rights and content

Highlights

  • Conversion of agricultural residues to sugars and lignin using sulfuric acid.

  • The physicochemical properties of the biomass affect the sugars yield.

  • Process variables such as heating ramp have a significant effect on sugars yield.

Abstract

The fractionation of sugarcane bagasse, plantain peel and brewer's spent barley to glucose, xylose and lignin was studied using 0.50 and 0.05 wt% H2SO4 as catalyst in a mixture of 80:20 wt% GVL:H2O. We obtained glucose and xylose cumulative yields with up to 90, 94 and 88% for sugarcane bagasse, plantain peel and brewer's spent barley, respectively. We studied the effects of the reaction conditions and physicochemical properties of the agricultural residues on the production of sugars. The production of sugars depends on the physicochemical properties of each residue. The lignin content, biomass crystallinity and ashes composition have the strongest effects on the production of sugars. The lignin can affect the accessibility to hemicellulose and cellulose during the reaction reducing the amount of sugars that can be obtained from the residue. The ashes present in the biomass may have alkaline species that can neutralize the acid catalyst reducing the rate of sugar production for a given acid concentration. The sugar production decreases as the biomass crystallinity increases.

Introduction

Renewable lignocellulosic biomass is a potential source to produce biofuels and platform molecules because it is abundantly generated from agricultural residues and food industry waste. Moreover, there is available technology to produce fungible biofuels that do not require changes to the combustion engines and the transportation infrastructure due to their similarity to conventional fuels. These unique characteristics of biomass-derived fuels offer advantages over other alternatives in the automotive fuel market.

Many important chemicals and platform molecules (i.e., ethanol, polylactic acid, levulinic acid, levoglucosenone, etc.) may be produced from biomass [[1], [2], [3], [4]]. Dumesic and co-workers, demonstrated that the synthesis of different intermediary molecules for the manufacture of biofuels may be produced from different biomass derived feedstocks (i.e., sugars, cellulose, hemicellulose, etc.) [[5], [6], [7], [8]]. Moreover, they demonstrated the production of soluble carbohydrates from the deconstruction of corn stover, hardwood and soft wood using sulfuric acid (H2SO4) in γ-valerolactone (GVL) and water mixtures [9]. Sugars like xylose and glucose are the main products in the deconstruction of these biomass sources [9]. These sugars may be converted to many platform molecules for the production of fuels and specialty chemicals (i.e., 5-hydroxymethylfurfural, furfural, ethanol, lactic acid, etc.) using heterogeneous catalysts [6,7,9]. Dumesic and co-workers used mixtures of GVL and water because they noted that using polar aprotic solvents such as GVL enhances the catalytic activity of the protons of H2SO4, which promotes the hydrolysis of cellulose and hemicellulose in the presence of water [10]. Also, GVL is a renewable green solvent that can be produced from biomass derived products [11]. Motivated by those discoveries, here we will use different local available biomass residues from the agricultural and food industry such as sugarcane bagasse, plantain peel and brewer's spent barley for sugars and lignin production using H2SO4 in a GVL-water mixture.

The worldwide production of sugar cane reaches 1.6 billion tons per year positioning it in the top three harvests in the world [10,11]. The main uses of sugarcane are for sugar and alcohol production. Those operations generate approximately 300 million tons of biomass residues yearly, mainly composed of bagasse and leaves [10,12]. These residues are mainly burned to produce electricity or run different processes in the industry [12]. Due to the abundance and low efficiency methods to treat the residues of sugarcane production, we consider that sugar cane bagasse is an important candidate for the catalytic production of sugars from biomass deconstruction. On the other hand, plantain and banana are also some of the most important harvests in the world. The production of plantain and banana reaches approximately 144 million tons per year, plantain contributing 38 million tons and banana 106 million tons [13]. The peel represents around 35% of the fruit weight for each case and this is translated to 50.4 tons of peel residues generated per year [14,15]. The main uses of plantain and banana peel are as compost and for cattle feed [14,16]. The local availability and global production of plantain and banana peel make them potential biomass sources for sugars production. Additionally, the beer breweries are producing large quantities of biomass residues. These residues are mostly the spent grain that comes from the mash or sugar extraction process for beer production [17]. The main grain used for beer production is barley. The global beer industry produces around 6 million tons of dry spent grain per year that is mainly used as animal feed [17,18]. Currently, the continuous increase in production of the craft beer industry and the limited uses for the breweries spent grain also makes this residue a potential candidate for sugar production from biomass.

In this regard, we studied the catalytic conversion of sugarcane bagasse, plantain peel and brewer's spent barley to sugars and lignin. We report sugars yields up to 90, 94 and 88% for bagasse, plantain peel and brewer's spent barley, respectively. The effect of the reaction conditions and the physicochemical properties of the biomass on the production of sugars was investigated. The physicochemical properties of each biomass source are important variables that define the process conditions for sugar production.

Deionized water (DI), γ-valerolactone (GVL) (ACROS Organics, 98% purity) and denatured ethanol (ACROS Organics, HPLC grade) were used as solvents. Sulfuric acid (H2SO4) (Fisher Chemical, 95–98% purity) was used as solvent and as catalyst. High purity He (Praxair, 99.999% purity) was used for the reactor pressurization. The feedstocks used were sugar cane bagasse (SCB), plantain peel (PP), and spent brewer's barley (BSB). Microcrystalline cellulose with an average particle size of 50 μm (ACROS Organics) was used for comparison. Granular fused silica (Sigma-Aldrich, 4–20 mesh) was mixed with the biomass samples to decrease heat and mass transfer effects and quartz wool (Chemglass, fiber size 8–15 μm) was used to keep the packed bed in place in the reactor. Polyvinylidene fluoride (PVDF) membrane filters (Millipore Durapore GVWP01300) were used for the recovery of lignin. The agricultural residues used were obtained from local sources.

The sugarcane bagasse used in this study came from sugar cane previously pressed in a sugar cane mill. In this step, the sugar cane juice was extracted from the cane. The barley was previously treated with 1 L of distilled water per pound of grain at 340 K for 1 h. The plantain peel was not pretreated. The biomass samples were dried at 318 K for 48 h in a conventional oven. The dried biomass was milled and sieved to obtain particles with a size of approximately 1.42 mm. At this point, the biomass was ready for fractionation.

The biomass characterization was performed according to the procedures published by the National Renewable Energy Laboratory (NREL). Before all procedures, the biomass sample was dried using the convection oven drying method published by the NREL [19]. Then, the biomass was milled and sieved using 20 and 70 mesh sieves. The resulting particles have particles sizes between 0.21 and 0.84 mm. The biomass samples were placed with a maximum depth of 1 cm and weighted. Then the sample was placed in a vacuum oven at 318 K until the weight did not change, this normally happened after 12–48 h. Finally, the sample was weighted.

We used the convection oven method published by the NREL for the determination of total solids in the biomass [20]. About 0.5–2.0 g of biomass previously dried was used for the analysis. This biomass was weighed to the nearest 0.1 mg in an aluminum pan previously weighed. The pan containing the biomass was placed in the vacuum oven at 378 K for 4 h. A small piece of aluminum foil was weighted and used as a cover for the pan with the biomass when it was extracted from the vacuum oven. The pan, biomass and the cover were weighted to the nearest 0.1 mg.

The inorganic composition of the biomass was determined using the procedure established by NREL [21]. About 1.0 and 2.0 g of dried biomass were weighed in an aluminum pan previously weighed. The sample was placed in a furnace at 378 K for 12 min with a heating ramp of 8 min from room temperature. The temperature was raised to 523 K with a heating ramp of 15 min and held for 30 min. Then the temperature was raised to 848 K using a heating ramp of 16 min and held constant for 3 h. The sample was cooled to 378 K and removed for weighting. The sample remaining in the pan is the inorganic content of the biomass.

The extractives composition were determined using the procedures published by NREL [22] to determine the biomass extractives soluble in water and ethanol. The extraction was done using a Soxhlet apparatus. It consists of a condenser tube at the top, a Soxhlet tube in the middle and a boiling flask with the solvent in the bottom. About 2–5 g of biomass were weighted in an extraction thimble. The exact quantity depends on the bulk density of the biomass and the Soxhlet tube used. The height of the biomass in the extraction thimble should not exceed the height of the Soxhlet siphon tube. The extraction thimble with the biomass sample is placed in the Soxhlet tube and 190 mL of deionized water are added to a boiling flask and the Soxhlet apparatus is assembled. The Soxhlet apparatus, specifically the boiling flask, is placed in an oil bath at 423 K for 24 h. When the extraction time is completed, the Soxhlet apparatus is removed from the oil bath. The boiling flask containing deionized water and the extractives is changed with one containing 190 mL of ethanol. The apparatus is placed again in the oil bath but this time at 383 K for 24 h. When the extraction time is completed, the Soxhlet apparatus is removed from the oil bath. The Soxhlet tube is connected to a line to perform vacuum filtration overnight. Once the filtration is finished, the extraction thimble with biomass is removed from the Soxhlet tube and placed in a vacuum oven at 318 K for 24 h. After drying is completed, the thimble with the biomass is weighted. Following the removal of extractives using water and ethanol, the biomass is used for the determination of cellulose, hemicellulose and lignin using the method proposed by the NREL [23]. For this procedure, we used 300 ± 10 mg of the corresponding biomass. The sample is placed in a vial with 3.00 mL of 72% sulfuric acid and stirred for 1 min. The closed vial is placed in a water bath at 303 K for 1 h under stirring. After this process is completed, the sample is transferred to a PTFE autoclave and 84 mL of deionized water are added. This reactor is placed in a conventional oven at 394 K for 1 h. Then, the reactor is removed from the oven and allowed to cool at room temperature. The sample is vacuum filtered, washed with at least 25 mL of distilled water and vacuum filtered for 24 h. After that the sample is removed and placed in a vacuum oven at 318 K for 24 h. The dried sample is weighted to the nearest 0.1 mg and placed in a furnace at 378 K for 12 min with a heating ramp of 8 min. The temperature is raised to 523 K with a heating ramp of 15 min and held constant for 30 min. Then the temperature is raised to 848 K using a heating ramp for 16 min and held constant for 3 h. At this point the organic material in the biomass sample is calcined and the remains correspond to inorganic material present in the biomass, which is not soluble in water and/or ethanol. The soluble inorganic material present in the biomass is subtracted from the extractives. The soluble inorganic material is calculated by subtracting the amount of non-soluble inorganic materials determined after the recovery of the extractives from the total inorganic material determined above. The remains are cooled to 378 K and the sample is removed for weighting. Finally, the absorbance at 240 nm of the liquid remaining is determined using a UV–Visible spectrophotometer (Thermo Scientific) to determine the amount of acid-soluble lignin in the biomass sample. The total amount of lignin is determined by summing the acid-soluble lignin and non-acid-soluble lignin, which is determined by calculating the difference in mass of the sample before and after calcination. The amount of cellulose and hemicellulose is determined by the difference in mass of the dried biomass and the sum of the extractives, lignin and inorganic material.

The biomass and cellulose X-ray diffraction patterns were obtained using a Siemens Diffraktometer D5000 X-ray diffractometer equipped with cross beam optics and a Cu Kα target operating at 40 kV and 44 mA. Standard powder diffraction patterns were gathered for 2θ angles ranging from 15 to 75° at a scanning speed of 2°/min and a step size of 0.1°. The biomass crystallinity index (CrI) was calculated as follows [24]:CrI=I002IAMI002were I002 is the intensity of the peak corresponding to the crystalline portion of biomass that in our case is at a diffraction angle of around 22°. The IAM is the peak intensity for the amorphous portion of biomass, which in our case appears at a diffraction angle of around 17°.

The reaction system consists of a 316 stainless steel (SS) tubular reactor 12.7 mm of external diameter and 254 mm of height, a 150 mm variable area flow meter with a 316 SS valve (Cole-Palmer EW-03269-18), and a backpressure regulator (Tescom 44–2200), that controls the pressure of the system. The reactor was heated using a cylindrical aluminum oven equipped with four heating rods (Omega CSH-10320). The liquid was fed using an HPLC pump (Hurst model A-30-SW). The reactor was packed vertically with 2.5 g of dry biomass, previously mixed with 5 g of silicone dioxide fused particles, between two plugs of quartz wool and silicone dioxide particles (SiO2). The reactor was packed with SiO2 particles to reduce the mass and heat transfer limitations during the reaction. The SiO2 particles were larger than the biomass particles and fixed between two plugs to avoid changes in the reactor volume and residence time. The biomass sample was heated to 430 K with a linear ramp for 20 min in flowing He at atmospheric pressure to remove the remaining moisture. The temperature was equilibrated for 5 min before the reactor was pressurized to 21 bar. Then, a mixture of 80 wt% GVL, 20 wt% H2O and 0.05 or 0.5 wt% H2SO4 was fed to the reactor with a flow of 1 mL/min using an HPLC pump, while the reactor was heated from 430 K to 490 K using a heating rate of 0.5–1.0 K/min. We drained the resulting accumulated liquid from the liquid collector every 10 min. Each sample was filtered using a 0.2 μm PVDF membrane filter to remove insoluble solids. Afterwards, the samples were diluted with 20 times its weight in DI water to precipitate lignin, centrifuged at 10,000 rpm, washed with DI water, filtered using a 0.2 μm PVDF hydrophilic membrane filter to recover the precipitated lignin that was dried at 358 K. Finally, the diluted liquid samples were analyzed using an HPLC and the filtered solids were dried and weighted.

The samples taken during the reaction were analyzed using a high performance liquid chromatograph (Waters Alliance 2690) equipped with a refractive index detector (Waters 2410) used for sugar analysis. The separation of the reaction products was achieved using a Bio-Rad Aminex HPX-87H column. The mobile phase was an aqueous solution of 5 mM H2SO4 with a flow rate of 0.3 mL/min. The column temperature was held at 353 K. The sample injection volume was 10 μL. The sugar yields were determined as the ratio of the accumulated sugar's carbon moles obtained during the reaction and the sum of cellulose and hemicellulose carbon moles present initially in each biomass sample.

Galactose and mannose may be obtained from hemicellulose fractionation [[25], [26], [27], [28], [29], [30], [31]]. The amount of these sugars produced is commonly low compared to xylose [[25], [26], [27], [28], [29], [30], [31]]. The galactose and mannose molar percent obtained from hemicellulose were around 3%, 15% and 11% for SCB, PP and BSB, respectively [[27], [28], [29], [30], [31]]. The HPLC column used for product analysis in the present study is not able to separate xylose, galactose and mannose. However, the amount of galactose and mannose that may be produced during the fractionation of biomass in this study is expected to be low. Hence, we assume that the contribution of galactose and mannose to the peak area corresponding to xylose is low; and, the amount of galactose and mannose that may be produced is lumped with xylose and reported as xylose.

Section snippets

Biomass characterization

Sugar cane bagasse, plantain peel and brewer's spent barley were used to produce sugars and lignin. Each residue has a different composition depending on its species, type and origin. Hence, the residual biomass samples used for this study were previously characterized following the NREL guidelines. Fig. 1 shows the composition results on a mass basis for SCB, PP and BSB. Two different samples of BSB and PP were used. The samples were named as BSB-1 and BSB-2 for brewer's spent barley and PP-1

Conclusions

The fractionation of sugarcane bagasse, plantain peel and brewer's spent barley to glucose, xylose and lignin was studied using H2SO4 as catalyst in a mixture of GVL and water. The reaction conditions such as heating rate and acid catalyst concentration affect the yield and products distribution. The use of high heating rates reduces the time needed to convert the biomass to sugars, but promotes the degradation of sugars during the reaction reducing its yield. The temperature at which the

Acknowledgements

This work was supported through the NSF CREST: Nanotechnology Center for Biomedical, Environment and Sustainability Applications - Phase II [grant number HRD-1345156]. We wish to also acknowledge partial support from the NSF funded RII Track-2 FEC: Center for a Sustainable Water, Energy, and Food Nexus (SusWEF) [grant number OIA-1632824].

References (52)

  • S.H.A. Rahman et al.

    Production of xylose from oil palm empty fruit bunch fiber using sulfuric acid

    Biochem. Eng. J.

    (2006)
  • A. Rodríguez-Chong et al.

    Hydrolysis of sugar cane bagasse using nitric acid: a kinetic assessment

    J. Food Eng.

    (2004)
  • C. Li et al.

    Comparison of dilute acid and ionic liquid pretreatment of switchgrass: biomass recalcitrance, delignification and enzymatic saccharification

    Bioresour. Technol.

    (2010)
  • National Renewable Energy Laboratory, Biomass Research- what Is a Biorefinery?

    (2009)
  • B. Kamm

    Biorefineries – industrial processes and products

    Ullmanns Encycl. Ind. Chem.

    (2012)
  • D.M. Alonso et al.

    Catalytic conversion of biomass to biofuels

    Green Chem.

    (2010)
  • F. Cao et al.

    Dehydration of cellulose to levoglucosenone using polar aprotic solvents

    Energy Environ. Sci.

    (2015)
  • J.N. Chheda et al.

    Production of 5-hydroxymethylfurfural and furfural by dehydration of biomass-derived mono- and poly-saccharides

    Green Chem.

    (2007)
  • Y.J. Pagán-Torres et al.

    Production of 5-hydroxymethylfurfural from glucose using a combination of Lewis and Brønsted acid catalysts in water in a biphasic reactor with an alkylphenol solvent

    ACS Catal.

    (2012)
  • J.S. Luterbacher et al.

    Nonenzymatic sugar production from biomass using biomass-derived γ-valerolactone

    Science

    (2014)
  • A.K. Chandel et al.

    Sugarcane bagasse and leaves: foreseeable biomass of biofuel and bio-products

    J. Chem. Technol. Biotechnol.

    (2012)
  • FAO, Food and agricultural commodities production....
  • J.M. Paturau, Alternative uses of sugar cane and its byproducts in agroindustries, FAO Agric. Consum. Prot....
  • M. Pillay et al.

    Banana Breeding: Progress and Challenges

    (2011)
  • E. Agama-Acevedo et al.

    Potential of plantain peels flour ( Musa paradisiaca L.) as a source of dietary fiber and antioxidant compound

    CyTA - J. Food

    (2016)
  • G. Tchobanoglous et al.

    Integrated Solid Waste Management: Engineering Principles and Management Issues

    (1993)
  • Cited by (9)

    • Potential applications of brewery spent grain: Critical an overview

      2022, Journal of Environmental Chemical Engineering
      Citation Excerpt :

      Studies have also used BSG to supplement culture media to obtain biocides [34,35]. The conversion of BSG into value-added bioproducts, such as sugars, lignin, and levulinic acid, has also been accomplished through acid-catalyzed hydrolysis [160] or assisted by microwaves [161]. Also, the BSG has been used in the delignification of lignocellulosic materials studies [138] and obtention of furfural, widely used in the chemical industry to the synthesis of many compounds such as liquid alkanes and fuel additives [162].

    • Catalytic conversion of cellulose to levoglucosenone using propylsulfonic acid functionalized SBA-15 and H<inf>2</inf>SO<inf>4</inf> in tetrahydrofuran

      2022, Biomass and Bioenergy
      Citation Excerpt :

      Biomass is an abundant renewable carbon source that includes wood, agricultural crops, and agricultural residues [1,2].

    • The cornerstone of realizing lignin value-addition: Exploiting the native structure and properties of lignin by extraction methods

      2020, Chemical Engineering Journal
      Citation Excerpt :

      When the cocatalyst is acid, biomass with high mineral or lignin content will lead to a low sugar conversion rate. Because metal oxides formed by minerals neutralize acid catalysts, and high lignin content hinders the conversion of cellulose and hemicellulose [126]. In addition, some auxiliary methods (microwave, steam explosion, etc.) combined with GVL can also obtain the same results.

    View all citing articles on Scopus
    View full text