An assessment of the potential of invasive weeds as multiple feedstocks for biofuel production

Abstract

The present study assessed the feasibility of five invasive weeds, namely, Arundo donax, Saccharum spontaneum, Mikania mikrantha, Lantana camara and Eichhornia crasspies, as a feedstock for biofuels production. The yield of total fermentable sugars from the pretreatment and enzymatic hydrolysis of these biomasses was assessed. However, the pretreatment and enzymatic hydrolysis were carried out at conditions optimized for the biomass of P. hysterophorus and thus, the conditions of pretreatment/enzymatic hydrolysis were not specifically optimized for any of the invasive weeds. Despite this, it was revealed that the average yield of total fermentable (hexose + pentose) sugars from all the weeds was 43.85 g per 100 g of raw biomass, which corresponds to a theoretical yield of 27.36 g ethanol and 17.96 g butanol. These yields are comparable to bioalcohol yields from the biomass of P. hysterophorus under optimized pretreatment conditions. Characterization of the biomass was carried out using X-ray diffraction, FTIR and SEM micrographs. The high yields of fermentable sugars obtained herein from invasive weeds, even under un-optimized pretreatment conditions, clearly point towards the feasibility of biorefinery using these weeds as multiple feedstocks for the production of alcoholic biofuels.

---

1. Introduction

The quest for alternative and renewable liquid transportation fuel has intensified over the past few years due to the fast and continuing depletion of fossil fuels threatening energy security, and due to increasing concerns about climatic change due to the large emission of greenhouse gases through vehicular exhausts. Waste biomasses in the form of noxious invasive weeds (in addition to conventional lignocellulosic substrates of agro- and forestry residues)^1,2 could form a large potential feedstock source for economic biofuel production.^3,4 Large numbers of such invasive species exist in India. These species infest millions of hectares of arable and non-arable (or infertile) land, leading to enormous monetary loss due to a reduction in crops and forage production. The actual biomass produced by these noxious weeds is in the range of 15–20 tons per hectare. Nonetheless, these biomasses can form a feedstock for liquid biofuels due to their significant content of holocellulose (in addition to lignin), which can be hydrolyzed to produce fermentable monomeric sugars. Hexose and pentose sugars produced from the hydrolysis of cellulose and hemicellulose can be used to produce alcoholic biofuels, such as ethanol and butanol.

The pretreatment of biomass prior to fermentation encompasses the removal of lignin and the acid/enzymatic hydrolysis of hemicellulose and cellulose fractions to pentose and hexose sugars. Removal of the lignin matrix in biomass causes a better exposure of the cellulose and hemicellulose fraction to acid/enzymatic action, which leads to enhancement of the sugar yield. The pretreatment of biomass is also aimed at a reduction in the crystallinity of cellulose and in increasing the biomass porosity and surface area, which all contribute to a faster and higher hydrolysis of the cellulose/hemicellulose, thus maximizing the yield of fermentable sugars.

1.1 Aim and approach

The use of a cheap feedstock and optimization of the cost-intensive pretreatment techniques are crucial aspects of the economic feasibility of a sustainable biofuels process. In the present study, we dealt with the pretreatment of multiple invasive species or weeds (which are essentially waste biomasses), and determined the yield of total (hexose + pentose) reducing (or fermentable) sugars, which is a measure of their potential as a feedstock for biofuels. The invasive weeds considered in this work are: (1) Arundo donax (AD), (2) Saccharum spontaneum (SS), (3) Mikania micrantha (MM), (4) Lantana camara (LC) and (5) Eichhornia crassipes (EC). A number of previous authors have also addressed the matter of the pretreatment of these waste biomasses and a summary of some representative studies in optimization of the pretreatment and fermentation of these invasive weeds is given in Table 1. Most of these studies include the following components: acid hydrolysis, delignification, enzymatic hydrolysis and fermentation of the acid/enzyme hydrolyzates to bioethanol. It could be inferred from Table 1 that the optimum pretreatment conditions differ significantly for each biomass.

Table 1 Summary of literature on pretreatment of invasive weeds and fermentation of the hydrolyzate

Reference	Biomass	Type and details of pretreatment	Major observations and conclusions
Komolwanich et al.^5	Arundo donax	Microwave assisted alkali pretreatment	120 °C/5 min/5% w/v NaOH. Total sugar release: 6.8/100 g biomass
		Two stages pretreatment microwave/dilute NaOH followed by microwave/dil. H2SO4	120 °C/5 min/5% w/v NaOH and 180 °C/30 min/0.5% w/v H2SO4 31.99/100 g biomass
			Glucose is the main monomeric sugars
Scordia et al.^6	Arundo donax	Optimization of enzymatic hydrolysis (EH) and simultaneous saccharification and fermentation (SSF). Optimization of EH and SSF using response surface methodology (RSM) with 2 input parameters: severity factor (SF) and oxalic acid concentration (OA)	Xylan content after dil. oxalic acid pretreatment reduced with increasing SF and OA. Glucan and lignin showed opposite trend with respect to xylan content after dil. OA treatment
		Temp. range: 150–190 °C	Final results: glucan conversion in EH = 95%, and ethanol production = 18 g L^−1 (75% of maximum theoretical yield) for pretreatment conditions of SF = 4.05 and OA = 5% w/w
		Treatment time range: 10–40 min
		Dil. oxalic acid concentration range = (2–8% w/v)
You et al.^7	Arundo donax	Cost effective pretreatment of biomass with protic acid resin Amberlyst 35DRY and ionic liquid 1-butyl-3-methylimidazolium chloride	Reduction in cellulose crystallinity and increased porosity due to extensive swelling of undissolved biomass and partial depolymerization of longer cellulose chains of dissolved biomass by Amberlyst catalyst
		Two stage treatment: (1) 160 °C with ionic liquid for 1.5 h, (2) slurry of ionic liquid and Amberlyst 35DRY resin at 90 °C for 1 h	High glucose yield of 92.8% than 42.8% yield for single ionic liquid treatment for enzyme loading of 20 FPU per g of substrate. Reusability of the ionic liquid solid catalyst system
Anderson et al.^8	Arundo donax	1.75% w/v H2SO4 (8.5 mL), 121 °C h^−1	Xylan associated monosaccharide yield: 201 mg g^−1
		1.2 mL of 10% w/v Ca(OH)2	Ethanol produced from A. donax: 109 mg g^−1
		GC 220 cellulase 5FPU per g biomass
		12U Novozyme 188 cellubiase
Scordia et al.^9	Arundo donax	Dilute acid pretreatment for following conditions:	S. carlsbergensis results: max ethanol concentration of 15.9 g L^−1 after 48 h at pH 5
		Effect of temperature (170 °C–190 °C)	S. stipitis results: ethanol concentration of 15.9 g L^−1 in 96 h. Increasing pH to 6 reduced lag phase and attained 18 g L^−1 after 72 h
		Acid loading (2–10% w/v); reaction time (15–40 min)
		Simultaneous saccharification and fermentation using commercial enzymes and two yeast strains: S. stipitis CBS 6054 and S. carlsbergensis FPL-450
Kuila et al.^10	Lantana camara	Laccase treatment followed by simultaneous saccharification and fermentation (SSF); crude laccase 400 IU per mL, pH 6.5; 10 g dry substrate liq : solid (2 mL g^−1) kept 8 h at 37 °C	Maximum bioethanol production: 5.14% v/v
		SSF using cellulase from T. reesei. Optimization of SSF through CCD based response surface methodology	Optimum substrate concentration: 17% w/v
			Optimum inoculum volume: 9% v/v
			Inoculum age 60 h and 144 h of incubation time
			Enhanced bioethanol concentration of 6.01% v/v using mutant strain of S. cerevisiae
Gupta et al.^11	Lantana camara	Biomass pretreatment with acid, alkali and chlorite to improve enzymatic saccharification of cellulose	Chlorite treatment removes maximum lignin with 90% w/w residual holocellulose
		(1) For 100 g substrate: acid concentration 1–5% w/v, time 15–16 min, temperature 121 °C	Results of enzymatic hydrolysis:
		(2) For 100 g substrate: alkali concentration 1–5% w/v for 2 h. Thermal pretreatment at 121 °C; time 15, 30, 45 and 60 min	(1) Chlorite treatment: 90% w/w initial holocellulose, saccharification of 86–92% fraction
		(3) Chlorite pretreatment: sodium chlorite 1–5% w/v for 121 °C for 15, 30, 45, 60 min followed by washing and drying at 60 °C	(2) Alkali treatment: 66–76% w/w initial holocellulose, saccharification of 55% fraction
			(3) Acid treatment: 39.5–48% w/w initial holocellulose, saccharification of 38–48% fraction
Kuhad et al.^12	Lantana camara	Dilute acid treatment at 3% w/v H2SO4 at 120 °C/45 min followed by delignification using sodium sulphite (5% w/v) and sodium chlorite (3% w/v)	Maximum sugar yield with dil. acid hydrolysis: 187.14 mg g^−1 total sugar with inhibitors such as phenolics, furfurals and hydroxyl methyl furfurals
		Detoxification of acid hydrolyzate using overliming and activated charcoal	87.2% lignin removal with sodium sulphate and sodium chlorite treatment
		Enzymatic hydrolysis of pretreated and delignified biomass; fermentation of acid and enzymatic hydrolyzate with P. stipitis and S. cerevisiae	Fermentation of acid and enzyme hydrolyzate with P. stipitis and S. cerevisiae yielded 5.16 g L^−1 (yield 0.32 g g^−1) and 17.7 g L^−1 (yield 0.48 g g^−1) ethanol in 24 and 16 h, respectively
Kataria et al.^13	Saccharum spontaneum	Biomass loading 5% (w/v); alkaline (NaOH) pretreatment at different concentrations (0.5, 1, 1.5, 2% w/v); period of treatment: 30, 60, 90, 120 min at 121 °C. Enzymatic hydrolysis of biomass with crude mixture of cellulase enzymes (20 units per g biomass); CMCase, FPAse, xylanase activities of 1.41, 1.12 and 6.23 units, respectively	Optimization results: (1) 70.75% lignin removal for 0.5% w/v NaOH treatment at 120 °C with total reducing sugar yield in enzymatic saccharification: 350 mg g^−1
			(2) Holocellulose increase in biomass from 64.7% to 79.61%
			(3) 79.3% lignin removal for 2% w/v NaOH in 90 min treatment. Holocellulose content increases to 76.7% with 70 mg g^−1 reducing sugar yield in enzymatic saccharification
Chaudhury et al.^14	Saccharum spontaneum	Different alkaline pretreatment methods (NaOH, NaOH + 10% urea and aqueous ammonia) were optimized for maximum delignification. Solubilization of solid residue using H2SO4 60% (v/v), 10% biomass loading at 30 °C for 4 h	Maximum delignification with alkaline treatment: (1) 47.8% from 7% NaOH, 48 h, and 10% biomass loading; (2) 51% from NaOH + urea (7% NaOH + 10% urea, 48 h and 10% biomass loading); (3) 48% from 30% ammonia (40 days and 10% biomass loading)
		Real hydrolysis of cellulose and hemicellulose with diluted slurry with acid concentration of 10% at 100 °C for 1 h
			Best result for reducing sugar yield with ammonia treated biomass: 0.58 g reducing sugar per g of initial biomass after acid hydrolysis. This accounts for nearly 85% of the total sugars present in the biomass
Chandel et al.^15	Saccharum spontaneum	Three different pretreatment approaches viz. dil. sulfuric acid (1.5% v/v at160 °C), dil. sodium hydroxide (0.4% w/v or 0.1 N at 120 °C), and aq. ammonia (15%) treatment at 50 °C and 24 h followed by enzymatic hydrolysis (5–35 FPU per g of dry substrate)	A max. sugar yield of 631.5 ± 3.25 mg g^−1 with 89.38% hydrolytic efficiency (HE) after enzymatic hydrolysis of aq. ammonia pretreated biomass
			Fermentation results: yields of 0.36 g g^−1 from acid hydrolyzate, 0.384 g g^−1 from enzymatic hydrolyzate of acid pretreated substrate, 0.391 g g^−1 from enzymatic hydrolyzate of alkali pretreated substrate and 0.4 g g^−1 from enzymatic hydrolyzate of aq. ammonia pretreated substrate
Yan et al.^16	Eichhornia crassipes (water hyacinth)	NaOH/H2O2-pretreated water hyacinth	Reducing sugar yield of 223.53 mg g^−1 dry biomass with reduced cellulose crystallinity
		1.5% v/v H2O2 and 3% (w/v) NaOH at 25 °C
Phothisantikul et al.^17	Eichhornia crassipes (water hyacinth)	Temperature range 160–220 °C, hydrothermal pretreatment using ball-mill reactor followed by enzymatic hydrolysis	Glucose yield at 220 °C in absence of CH3COOH and K2CO3 = 0.267
		Effects of CH3COOH and K2CO3 on the liquid composition were investigated experimentally	Glucose yield at 200 °C with 0.75 wt% CH3COOH and 10% biomass = 0.855
			Glucose yield at 220 °C with 0.5 wt% K2CO3 = 0.195
			Addition of K2CO3 did not suppress hydrolysis in hydrothermal treatment
Satyanaga-lakshmi et al.^18	Eichhornia crassipes (water hyacinth)	Preliminary pretreatment with different acids (HCl/H2SO4, 2% v/v) and organic acids (acetic/formic acid, 30% v/v) and autoclaving (121 °C, 15 lb) for 60 min and 10 g biomass	Most optimum conditions for pretreatment: 4% w/v H2SO4 pretreatment at 10% w/w biomass loading at temperature of 121 °C for 75 min to produce 0.356 g g^−1 reducing sugars
		Further optimization with H2SO4 concentration: 1–7% w/v; biomass loading 5–30% w/w, temperature 80, 100 and 121 °C; incubation time 15–90 min	Optimum condition for hydrolysis of pretreated biomass: 12.5% w/w biomass loading, incubation period 24 h, surfactant concentration 0.1%, commercial cellulase concentration 70 FPU
		Enzymatic saccharification of biomass with commercial cellulase Zytex (30 FPU per g biomass) and surfactant	Final ethanol yield: 0.292% w/v with actual efficiency of 59.3%

---

The approach in the present study is somewhat different from the earlier studies listed in Table 1, which studied the pretreatment of individual biomasses. In a previous paper,² we presented an extensive study on the assessment and optimization of as many as 17 pretreatment techniques (physical/chemical/physico-chemical) for the invasive species of Parthenium hysterophorus for the maximum production of reducible sugars that could be fermented to produce alcoholic fuels. In the present study, we carried out pretreatment of the five invasive weeds mentioned above at optimized conditions determined for Parthenium hysterophorus,² and assessed the yield of reducible sugars. It could be expected that the optimum pretreatment conditions for the five invasive weeds could be different than those for Parthenium hysterophorus. The major contemplation underlying the approach of pretreating the invasive weeds listed above at conditions optimized for Parthenium hysterophorus was to assess the output of a bioprocess with feedstock flexibility. This can be explained in greater details as follows: depending on the availability of biomasses in different parts of the year, the biofuel industry for the large-scale production of alcoholic biofuels may require to change the feedstock or to use a mixed feedstock comprising several biomasses as sufficiently large quantities of a single biomass may not be available throughout the year. In such a situation, it may not be feasible or practical to perform comprehensive optimization of the pretreatment conditions for each biomass used as feedstock. Moreover, the optimum pretreatment conditions (such as acid/alkali concentrations or temperature/pressure of the autoclaving) may show significant variations for different biomasses. Thus, the specifications of processing equipment designed for the pretreatment of one biomass may not be suitable for other biomasses. Obviously, the replacement of process equipment for different biomasses is rather impractical and so under this limitation, it is inevitable to treat different biomasses at conditions optimized for the representative biomass that was considered for the process design. In such a situation, it is necessary to make a preliminary estimate of the alterations in the quality of the hydrolyzates in terms of the concentrations of pentose and hexose sugars with changing feedstock. The present study essentially attempts to paint a picture of such variations by pretreatment of the five selected invasive weeds at conditions optimized for the weed of Parthenium hysterophorus.

2. Materials and methods

2.1 Biomass collection and processing

Biomasses of all five weeds, namely AD, SS, MM, LC and EC, were collected from our institute campus. The substrate for pretreatment was the whole plant body except for the roots. After collection, the biomass was dried in ambient air, followed by chopping it into small pieces of a few mm lengths. The chopped biomass was washed with water and again dried in a hot air oven at 60 °C for 24 h. Prior to pretreatment; the particle size of the dried biomass was further reduced to <1 mm using a domestic mixer grinder. The powdered biomass was then stored in air-tight containers at room temperature for further experiments.

2.2 Proximate and biochemical analysis of biomass

Total solids, volatile matter, the calorific value and the ash content of all five biomasses were determined according to the standard protocols, namely CEN 15104 and ASTM E1755-01. Ultimate analysis of all the biomass samples was performed by using the Euro EA Elemental (C, H, N) analyzer (Eurovector EA 3000). The results of the proximate and ultimate analyses of the invasive weeds are given in Table 2A. Characterization of the chemical composition (i.e. determination of cellulose, holocellulose and lignin content) of raw, pretreated and delignified biomass was performed using standard protocols, namely Updegreff DM, 1969 and standard TAPPI protocols. The results of the determination of the lignocellulosic composition of the invasive weeds are given in Table 2B.

Table 2 Characterization of invasive weeds. (A) Proximate and ultimate analyses of different biomasses. (B) Lignocellulosic composition of different biomasses

(A)
Biomass feedstock	Ultimate analysis^b					Proximate analysis
	C (%)	H (%)	N (%)	S (%)	O (%)	Ash^a (%)	Calorific value^a (MJ kg^−1)
a The values are mean ± SE (n = 3).b An approximate variation of ±5% is expected in these values for various samples drawn from same source.
AD	46.4	5.8	3.0	0.2	44.7	6.5 ± 0.2	18.7 ± 0.4
SS	49.1	6.2	2.7	0.0	41.9	9.0 ± 0.9	15.7 ± 0.3
MM	43.1	5.6	4.6	0.0	46.7	8.5 ± 0.7	19.1 ± 0.7
LC	50.2	6.4	5.1	0.0	38.2	7.2 ± 0.4	18.2 ± 0.3
EC	41.0	5.3	4.2	0.0	49.4	20.6 ± 0.1	9.21 ± 0.3

---

(B)
Sample	Holocellulose^a (wt% raw biomass)	Lignin (wt% raw biomass)	Cellulose content of biomass (wt% of biomass after preceding treatment)
			Raw	Post dil. acid hydrolysis	Post delignification
AD	69.0 ± 2.8	22.5 ± 0.6	55.0 ± 4.2	74.2 ± 1.2	82.3 ± 3.5
SS	61.0 ± 3.5	23.0 ± 3.2	49.3 ± 3.0	76.6 ± 3.2	96.1 ± 3.2
MM	61.5 ± 1.4	19.1 ± 0.7	50.2 ± 2.5	58.2 ± 3.1	93.6 ± 4.3
LC	61.8 ± 1.9	27.2 ± 4.5	38.3 ± 3.7	69.0 ± 3.4	80.7 ± 2.6
EC	82.2 ± 2.1	04.1 ± 0.4	47.3 ± 1.7	72.7 ± 2.1	96.7 ± 3.9

---

2.3 Acid hydrolysis of the biomass

The optimum conditions for the acid hydrolysis of Parthenium hysterophorus were determined by Singh et al.² as follows: 1% (v/v) H₂SO₄ (equivalent to 0.36 N) mixed with 10% w/v biomass, then autoclaved at 121 °C and 15 psi for 30 min, followed by rapid steam release. Dried biomass of all five weed species was pretreated under these conditions. After completion of the pretreatment, the biomass from the reaction mixture was separated by filtration through a double-layered muslin cloth. The residual chemicals left on the biomass surface after acid pretreatment were removed by successive water washes. This procedure was continued until the pH of the wash water became 7 (indicating neutral conditions). This was followed by drying of the biomass residue in a hot air oven for 24 h at 60 °C. The dried biomass containing cellulose and traces of lignin was used for further processing. Acid pretreatment causes hydrolysis of the hemicellulose in biomass, resulting in a release of pentose sugars. The filtrate of the acid pretreatment or the acid hydrolyzate was thus analyzed for the sugar content. The hydrolysate was detoxified to remove the inhibitory compounds. The composition of the hydrolyzate was determined using HPLC. The pentose sugars in the hydrolysate predominantly comprise xylose, but other sugars such as arabinose, mannose and galactose are also present; however, their content is negligible as compared to xylose.

2.4 Detoxification of acid hydrolyzate

Detoxification of the hydrolyzate from the acid hydrolysis was carried out in two steps. Initially, the pH of the hydrolysate was increased to 10 with the addition of Ca(OH)₂, followed by stirring for 30 min. Next, the hydrolyzate was neutralized with the addition of concentrated H₂SO₄, with subsequent centrifugation at 10 000 g for 15 min for the removal of suspended solids. 1.5% w/v activated charcoal was added to the hydrolyzate with continuous stirring for 30 min at room temperature. Inhibitory compounds formed during the acid hydrolysis adsorb on the activated charcoal. The particles of activated charcoal were then removed by vacuum filtration of the hydrolyzate.

2.5 Delignification

The delignification of the biomass obtained after acid hydrolysis was carried out using the procedure outlined by Bharadwaja et al.¹⁹ This procedure makes use of sonication (or ultrasound irradiation) during delignification. The delignification process is significantly intensified by the physical and chemical effects induced by ultrasound and cavitation. The physical effects of ultrasound and cavitation includes the generation of intense micro-mixing through the phenomena of microstreaming, microturbulence and acoustic waves, while the chemical effects of transient cavitation include the generation of highly reactive radicals through dissociation of gas and vapour molecules entrapped in the bubble.^20–23 A probe-type programmable and micro-processor controlled ultrasonic processor (Sonics & Materials, Model VCX 500) with a maximum power of 500 W and frequency of 20 kHz was used for sonication of the reaction mixture. The reaction was carried out in a 100 mL beaker. The total volume of the reaction mixture was 80 mL, with an alkali concentration of 1.5% w/v NaOH and a biomass loading of 2% w/v. The ultrasound probe was set at 30% amplitude, with a theoretical power consumption of 150 W at a duty cycle of 83% (50 s on and 10 s off in 1 min of sonication). The actual power consumption of the ultrasound probe was determined using a calorimetric technique.^22,24 The total sonication time was 10 min, during which the temperature of the reaction medium was maintained at 30 °C. After the completion of sonication, the reaction mixture was filtered through a double-layered muslin cloth for removal of the solid biomass residue. This cellulose-rich biomass residue (after removal of the lignin and hemicellulose) was washed with hot water several times until the pH of the wash water was neutral, which ensured no chemicals were left on the biomass surface. The biomass residue was dried for ∼12 h in a hot air oven at 60 ± 3 °C, and was then used for the enzymatic hydrolysis.

2.6 Biomass saccharification using commercial enzymes

The enzymatic hydrolysis (or saccharification) of delignified biomasses of all five species was carried out using commercial cellulase and cellobiase enzymes (Sigma Aldrich) at the optimum conditions reported by Bharadwaja et al.¹⁹ The hydrolysis was performed in an incubator shaker (Orbitek, Scigenics Biotech) in 50 mM citrate phosphate buffer solution (pH 4.8) at 50 °C and at 150 rpm. The reaction mixture was taken in a 150 mL Erlenmeyer flask with a total reaction volume of 20 mL. The concentration of pretreated biomass in the reaction mixture was 4.2% w/v, with cellulase and cellobiase concentrations of 135 and 75 FPU per g biomass, respectively. The hydrolysis was carried out for 120 h. 0.005% w/v sodium azide solution was added to the mixture to avoid external microbial contamination. 0.1 mL samples of the reaction mixture were withdrawn periodically during hydrolysis and were analyzed to assess the release of sugar.

2.7 Determination of reducing sugar in the hydrolyzate

Both pentose-rich acid hydrolyzate and hexose-rich enzyme hydrolyzate were subjected to centrifugation for 10 min at 10 000 rpm (26 832 g) at 4 °C. The total reducing sugar in the hydrolyzate was estimated using NS (Nelson and Somogyi) method of. The presence of individual sugars in the hydrolyzate was confirmed through HPLC analysis (Perkin Elmer series 200). The HPLC instrument comprised a pump, a refractive index detector, a vacuum degasser and a Hi-plex-H column (Varian, 300 mm × 5 μm × 4.6 mm). Deionized Milli Q water at a flow rate of 0.4 mL min⁻¹ was used as the mobile phase. Prior to injection in to the HPLC column, samples withdrawn from the reaction mixture were diluted and filtered through a 0.2 μm membrane filter to remove any suspended particulate matter.

2.8 Characterization of the biomasses

2.8.1 SEM analysis. The morphologies of the five biomass species at various stages of pretreatment, namely raw biomass, post acid pretreatment and post alkaline delignification, were analyzed with a Scanning Electron Microscope (JEOL, Model: JSM-6360). For SEM analysis, the samples were prepared by drying the biomass at 60 °C for 24 h and then spreading the dried samples onto carbon tape placed over the surface of the SEM stub. The samples were sputtered with 10 nm gold in a sputter-coater. The SEM micrographs of raw, acid-pretreated and delignified biomass were taken at a similar magnification for comparison of the different micrographs to discern the effects of pretreatment on the biomass structure and morphology.

2.8.2 FTIR spectroscopy. The raw, acid-pretreated and delignified biomasses of all five species were characterized for the change in structural composition following pretreatment. An FTIR spectrophotometer (Perkin Elmer, Spectrum Two) was used to characterize the samples. The samples for analysis were prepared by mixing a small quantity of biomass (10 mg) and KBr in a ratio (w/w) of 1 : 100. The mixtures were ground well and the spectra were recorded in the range of 400–4000 cm⁻¹ using 200 mg of biomass + a KBr mixture in the form of pellets.

2.8.3 X-ray diffraction. The effects of acid pretreatment and delignification on the crystallinity of residual cellulose in pretreated biomass were assessed using an X-ray diffractometer (D8 Advance, Bruker, Germany). The diffractometer was operated at 40 kV and 40 mA using Cu-Kα (λ = 1.54 Å) radiation. Samples of the pretreated and delignified biomasses of all five species were scanned in the range of 2θ = 5–35° s⁻¹ with a step size of 0.05° s⁻¹. The crystallinity index (CrI) of the residual cellulose in biomass was determined with the formula of Segal et al.:²⁵
where I_crystalline = intensity of the crystalline peak at 2θ = 22° and I_amorphous = intensity of the amorphous peak at 2θ = 18°.

3. Results and discussion

3.1 Results of the biomass characterization

The characterization of the raw biomass and the biomass after two stages of pretreatment, namely acid pretreatment and alkaline delignification, was carried out using three techniques, namely scanning electron microscopy (SEM), FTIR and X-ray diffraction. The results of these analyses are presented below.

3.1.1 SEM analysis. The surface morphology of the biomass after acid pretreatment and alkaline delignification was assessed using SEM micrographs. The representative micrographs of the raw biomass, the acid-pretreated biomass and the delignified biomass of all five biomass are shown in Fig. 1A.1 to E.3. Some distinct features of the surface morphology that are evident from the SEM micrographs are given below.

Fig. 1 SEM micrographs of five biomass species in native or raw state and after various stages of pretreatment. (1) raw biomass, (2) biomass after acid hydrolysis with autoclaving (1% v/v H₂SO₄, 121 °C, 15 lb pressure), (3) delignified biomass (alkaline treatment with 1.5% w/v NaOH with sonication). (A) Arundo donax; (B) Saccharum spontaneum; (C) Mikania micrantha; (D) Lantana camara; (E) Eichhornia crassipes.

Irrespective of the type of biomass, the observed effects of pretreatment on the fibre structure of all the biomasses are almost similar. The total amount of residue on the sample surface reduces after pretreatment. Dilute acid pretreatment mainly removes the hemicellulosic fraction. Removal of this fraction creates microspores or holes on the surface of the biomass, as evident from the micrographs shown in Fig. 1C.2 and D.2 corresponding to the biomasses of Mikania micrantha and Lantana camara. Another common feature of the micrographs of all the biomasses is the presence of globular structures on the surface of biomass after dilute acid pretreatment. These structures are associated with lignin condensation and agglomeration. As noted by Lima et al.,²⁶ formation of the globular structures is related to the severity of the pretreatment conditions for that particular biomass. During acid pretreatment at elevated pressure and temperature, the lignin molecules become fluid and coalesce, giving rise to the formation of droplets within the cell matrix. Due to hydrostatic pressure within the cell wall layers, some lignin droplets get redeposited on the biomass surface during cooling of the bulk liquid medium.

The SEM micrographs after alkali treatment or delignification reveal the disappearance of the lignin globules from the biomass surface, with a concurrent rise in the surface roughness of the biomass. This result indicates that alkaline treatment causes degradation of the fibrillar structure or tissue of cellulose and lignin. Destruction of the tissues with removal of the lignin helps in gaining a better access of the cellulose to enzyme action, which results in a faster and higher yield of fermentable sugar from the hydrolysis.

3.1.2 FTIR analysis. The changes in composition of the lignocellulosic biomass after different pretreatments can be monitored with FTIR spectroscopy. The effect of dilute acid hydrolysis and alkaline delignification on the composition of the five biomass species was monitored in terms of the relative change in absorbance at specific band positions representing certain biomass components. The vibrational frequencies of different functional groups in the IR spectrum of biomass, along with percentage change in the intensities of these bands after pretreatment, in comparison to raw biomass, are listed in Table 3.^2,12,27,28 The definition of the percentage relative change is given as:

Table 3 Summary of characterization of pre-treated biomass using FTIR spectroscopy (relative changes in intensities for various bands)^a,b

Band position and assignment^c (cm^−1)	Relative change in intensities (%)
	AD AH	AD DLG	SS AH	SS DLG	MM AH	MM DLG	LC AH	LC DLG	EC AH	EC DLG
a Relative change (%) = (intensity of untreated biomass − intensity of pretreated biomass)/intensity of untreated biomass × 100.b Abbreviations: AH-post acid hydrolysis; DLG-post delignification.c Data taken from Kumar et al.,^28 Singh et al.,^2 Kuhad et al.,^12 Sun et al.^27
900 (band of cellulose)	3.6	34.7	−0.4	19.7	−6.8	16.7	0.5	29.9	33.5	76.6
1098 (amorphous to crystalline cellulose ratio)	2.4	27.1	11.5	16.5	−10.8	2.7	−6.9	15.8	32.7	74.5
1059 (C=O stretching due to carbohydrate–lignin linkage)	1.8	26.4	11.9	15.7	−11.7	2.3	−5.3	14.1	33.4	73.7
1238 (hemicellulose–lignin linkage)	1.1	34.8	−4.2	20.3	−5.6	14.9	−4.3	31.3	31.6	76.5
1245 (C=O absorption resulting from acetyl group cleavage)	0.2	35.0	−4.3	21.2	−5.3	16.1	−4.6	32.0	31.3	76.5
1260 (ester absorbance related to removal of uronic acid)	−0.3	36.9	−0.5	26.6	−5.2	18.5	−5.5	33.8	31.2	77.2
1378 (band of hemicellulose)	−2.5	37.7	−7.7	25.4	−8.6	19.2	−13.4	32.0	29.3	77.4
1428 (band of cellulose)	−1.4	40.9	−11.1	28.1	−11.2	21.4	−12.5	38.4	27.6	78.3
1458 (aromatic ring vibration related to lignin removal)	−4.0	19.6	5.4	9.0	−8.0	−1.3	−11.0	13.0	25.7	68.9
1508 (aromatic ring vibration related to lignin removal)	−6.1	15.2	−2.5	1.2	−8.3	−7.8	−14.7	8.1	25.1	65.7
1595 (aromatic ring stretch related to lignin removal)	−4.3	30.3	−3.4	19.8	−11.5	7.6	−20.1	26.2	25.5	75.3
1720 (carboxylic acids/ester groups)	−5.9	27.7	−23.6	11.7	−0.8	14.3	−6.2	30.1	26.7	72.8
1738 (C=O stretching due to carbohydrate linked with lignin)	−9.0	28.6	−31.5	12.5	−4.1	16.1	−9.6	32.1	24.0	72.7
1745 (carbonyl bonds related to lignin side chain removal)	−11.9	29.2	−40.5	11.1	−6.0	15.0	−12.7	33.3	22.9	72.8
2900 (C–H stretching related to rupture of methyl/methylene group of cellulose)	11.5	37.2	−21.5	25.5	−5.2	20.9	−15.5	40.7	18.8	76.5
3348 (O–H stretching related to rupture of cellulose–hydrogen bonds)	14.3	17.6	−14.8	9.6	−11.1	−9.4	−21.7	20.8	18.2	74.9

---

As per the above definition, a positive relative percentage change of intensity at all specific bands indicates a reduction in the particular component assigned to that band. The IR spectra of biomasses at three stages, namely raw biomass, post dilute acid pretreatment and post delignification, are given in the ESI.† For most absorption bands, the values of the relative change in the intensity were revealed to be positive after completion of both pretreatments. The changes indicate the efficient removal of hemicellulose and lignin after the biomass treatments. Positive changes for the band positioned at 1378 cm⁻¹ indicates the removal of hemicellulose during acid hydrolysis. Positive changes for the band positioned at 900 and 1098 cm⁻¹ indicate a reduction in cellulose crystallinity, which assists in enzymatic hydrolysis. The overall conclusion of the FTIR analysis is that the treatment of acid hydrolysis at 1% v/v H₂SO₄, 121 °C and 15 psi pressure, and alkaline delignification at 1.5% w/v NaOH assisted by ultrasound are able to induce desired changes in the biomass structure; although these conditions have not been individually optimized for each of the five biomasses considered in the present work.

3.1.3 XRD analysis. Changes in the biomass composition and structure induced by the two pretreatments, i.e. dilute acid hydrolysis and alkaline delignification, can also be mapped by X-ray diffraction. The crystallinity index (CrI) determined by XRD analysis is an excellent tool to assess the extent of removal of the two components of biomass, i.e. hemicellulose and lignin, which have a relatively more amorphous structure as compared to cellulose. A rise in crystallinity index essentially indicates an increase in the percentage of the cellulose content of biomass after pretreatment. The crystallinity indices of the five biomass species at various stages, namely raw biomass, dilute acid treatment and alkaline delignification, are given in Table 4. The actual X-ray diffractograms of the five biomasses are given in the ESI.† It could be seen from Table 4 that the crystallinity of the biomass increases after treatment with 1% v/v H₂SO₄ at 121 °C and 15 psi pressure, which indicates removal of the hemicellulose. The extent of the rise in crystallinity is, however, different for the five biomass species. The biomasses of Eichhornia crassipes and Mikania micrantha show close to a 100% increase in crystallinity, while the biomass of Saccharum spontaneum shows a very marginal rise. For the biomass of Lantana camara, the crystallinity shows a slight reduction with dilute acid treatment.

Table 4 Characterization of pre-treated biomass by XRD in terms of crystallinity index

Biomass	Crystallinity index, CrI (%)
	Raw	Post dil. acid hydrolysis	Post delignification
AD	37.5	60	64.70
SS	52.00	56.52	62.50
MM	20.00	38.70	48.57
LC	50.00	47.50	57.14
EC	16.67	33.33	42.85

---

The crystallinity index shows a further rise with alkaline delignification. Alkali pretreatment in the presence of ultrasound leads to delignification, with saponification of the intermolecular ester bonds cross-linking the xylan hemicelluloses and lignin, in addition to breakage of the linkages between cellulose and lignin. Singh et al.² observed that the alkaline treatment of lignocellulosic biomass results in a swelling of the biomass, also with an increase in surface area, a decrease in the degree of polymerization, separation of the structural linkages between lignin and carbohydrates and disruption of the lignin structure. All of these effects contribute to a reduction in the amorphous fraction of biomass and exposure of the crystalline cellulose fraction, which is reflected in an increase in crystallinity of all the five biomasses after delignification. These results are in concurrence with the results of Singh et al.,² who observed an increase in the crystallinity of Parthenium hysterophorus after 1% v/v H₂SO₄ treatment at 121 °C and 15 psi pressure and delignification with 1.5% w/v NaOH treatment in the presence of ultrasound. The results of the X-ray diffraction analysis are additional corroborations that, although conditions for the dilute acid hydrolysis and delignification are not optimized for the five invasive weeds used in the present study, the pretreatment of the weed biomasses at these conditions was still able to induce desired changes in their composition.

3.2 Results of dilute acid and enzymatic hydrolysis

The experimental results on the pretreatment and enzymatic hydrolysis of the five biomass species considered in this work are given in Tables 2B and 5A. It should be noted that in Table 2B, the cellulose content of biomass after dilute acid hydrolysis is expressed as a percentage of raw biomass, while that after delignification is expressed as a percentage of biomass after acid pretreatment.

Table 5 Results of TRS release from biomasses during different pretreatments and subsequent enzymatic hydrolysis. (A) TRS yield in acid and enzyme hydrolysis. (B) Kinetics of TRS release during enzyme hydrolysis. (C) Theoretical yield of alcoholic biofuels from fermentable sugars^a

(A)
Biomass	TRS released in AH (g L^−1)	TRS yield in AH (mg g^−1 of raw biomass)	TRS released in EH till 36 h (g L^−1)	TRS yield in EH till 36 h (mg g^−1 of delignified biomass)^b	TFS (mg g^−1 of raw biomass)	TFS yield (g per 100 g of raw biomass)
a TRS – total reducing sugar; TFS – total fermentable sugar, AH – acid hydrolysis; EH – enzymatic hydrolysis.b For greater details on enzymatic hydrolysis of invasive weeds, refer to Borah et al.^29c Results in Table 2B indicate that for biomasses of AD and LC, complete delignification is not achieved at treatment conditions of 1.5% w/v NaOH and sonication.d Rate of TRS release is calculated ignoring the traces of lignin left in the biomass after alkaline treatment with sonication.e Maximum theoretical yield for ethanol from hexose as well as pentose sugars is 0.51 g g^−1 sugar.^2f Maximum theoretical yield for butanol from hexose as well as pentose sugars is 0.41 g g^−1 sugar.^2
AD	31.60	315.9	30.41	724.0	492.9	49.29
SS	39.45	394.5	35.81	851.7	585.9	58.59
MM	27.21	272.1	24.90	592.0	449.7	44.37
LC	11.80	118.0	27.81	662.2	306.7	30.66
EC	23.86	238.6	31.86	758.6	363.6	36.36

---

(B)
Biomass	TRS released in EH^c (mg per g-delignified biomass per h)	TRS released in EH^d (mg per g-cellulose per h)
AD	23.36	28.52
SS	26.24	27.34
MM	17.83	19.05
LC	21.47	26.61
EC	24.75	25.60

---

(C)
Biomass	Theoretical ethanol yield (g per 100 g raw biomass)^e	Theoretical butanol yield (g per 100 g raw biomass)^f
AD	25.14	20.20
SS	29.88	24.02
MM	22.62	18.19
LC	15.63	12.51
EC	18.54	14.90

---

For the biomass of Saccharum spontaneum and Eichhornia crassipes, the percentage cellulose content of biomass after pretreatment is above 95%, indicating almost the complete removal of lignin and hemicellulose. This was also confirmed by the large positive values for the percentage relative changes to various bands in the IR spectrum of Saccharum spontaneum and Eichhornia crassipes (as depicted in Table 3) corresponding to different functional groups in the biomasses. However, for the biomasses of Arundo donax and Lantana camara, the percentage cellulose content after delignification and acid hydrolysis was less then 85%, which indicates the presence of residual lignin in biomass after pretreatment. The results given in Table 2B and demonstrate the potentials of the weed species for biofuel production. The holocellulose content of Eichhornia crassipes is the highest, while Lantana camara has the highest lignin content. The lignin hinders the hydrolysis of hemicellulose during dilute acid pretreatment. The subsequent alkali pretreatment at 1.5% w/v NaOH assisted with sonication has also not been able to remove the lignin completely. As a consequence, the pretreated biomass undergoing enzymatic hydrolysis comprises only 80% cellulose. An interesting result was seen for the biomasses Arundo donax and Saccharum spontaneum, where the initial lignin content of these biomasses is almost similar, yet the biomass of Saccharum spontaneum after pretreatment had a far higher content of cellulose than Arundo donax. An explanation for this discrepancy can be given in terms of the structural composition of lignin in Arundo donax. You et al.⁷ recently published an analysis of the lignin structure from the stems and foliage of Arundo donax, and described how lignin is an amorphous polymer, and is made up of three aromatic alcohols or monolignols, namely p-coumaryl, coniferyl and syringyl. These monolignols form distinct lignin units during the lignification process, known as p-hydroxy phenyl (H), guaiacyl (G) and syringyl (S) units. The analysis of You et al.⁷ clearly indicated that the milled wood lignin (both stem and foliage) of Arundo donax were HGS-type lignin with an S/G ratio in the range of 0.15–0.62, with a strong predominance of G units. The main lignin interunit linkages were β-O-4 alkyl–aryl ethers followed by β–β′, β-1′ and α,β-diaryl ethers together with cinnamyl alcohol and cinnamaldehyde end groups. The foliage lignin with higher condensed G units contained a greater amount of tricin. The linkages between lignin and tricin could be alkaline stable. As a result of its structural characteristics, the lignin in Arundo donax is not removed completely during alkaline pretreatment at 1.5% w/v NaOH.

3.2.1 Enzymatic hydrolysis. The results of enzymatic hydrolysis of the pretreated biomasses of the five invasive weeds are depicted in Fig. 2. The kinetics of sugar release during enzymatic hydrolysis with respect to both pretreated biomass (post acid hydrolysis and delignification) and only cellulose is given in Table 5B. The trends in sugar release in 36 h of treatment for the five biomasses are as follows: Saccharum spontaneum > Eichhornia crassipes > Arundo donax > Lantana camara > Mikania micrantha. Comparing this trend against the cellulose content of the pretreated biomasses, we find an anomaly that the biomasses of Arundo donax and Lantana camara (which have 18% and 20% residual lignin, respectively) have a higher release of sugars than the biomass of Mikania micrantha, which has 96.7% cellulose. A possible explanation for this inconsistency can be given in terms of hydrolysis of some portion of the cellulose (in addition to hemicellulose) during the dilute acid treatment. The reason underlying this could be the lesser crystallinity of the biomass of Mikania micrantha, as indicated by the crystallinity index in Table 4. As a consequence, the absolute quantity of cellulose in the pretreated biomass available for enzymatic hydrolysis is reduced. This is manifested in terms of a lower and slower sugar release during the enzymatic hydrolysis. For greater details on the enzymatic hydrolysis of invasive weeds, we refer readers to our recent paper.³⁰

Fig. 2 Time profile of TRS release during enzymatic hydrolysis of delignified biomasses of different invasive weeds for 36 h of treatment.

Some other observations on the results of the enzymatic hydrolysis and their plausible explanations on the basis of characterization of the biomasses are as follows:

(1) Delignification causes depolymerization of lignin through homolytic cleavages of the phenyl ether β-O-4 and α-O-4 bonds, resulting in exposure of the cellulose moieties in the biomass. The residual biomass after delignification is rich in cellulose and has an increased crystallinity due to removal of the amorphous lignin and hemicellulose. However, we find that the order of crystallinity of delignified biomass (i.e. AD > SS > LC > MM > EC) does not follow the order of the cellulose content of biomass (i.e. EC > SS > MM > AD > LC). This disparity is attributed to two factors: (1) loss of crystallinity during acid hydrolysis due to partial hydrolysis of the cellulose (in addition to hemicellulose), as noted earlier, and (2) the presence of residual lignin in biomass after pretreatment.

(2) The lesser the crystallinity of the cellulose, the faster its reactivity during enzymatic hydrolysis, thus resulting in high hexose sugar yields. However, the order of sugar yields in enzymatic hydrolysis (i.e. SS > EC > AD > LC > MM) does not follow the order of the post-delignification crystallinity of the biomass (i.e. AD > SS > LC > MM > EC). This disaccord is also ascribed to other factors affecting the reaction between the enzymes and the cellulose substrate, such as the presence of residual lignin in the biomass, which may compete with cellulose for adsorption of the enzyme, and hindrance offered by the ash content of the cellulose.

3.2.2 Total fermentable sugar (TFS) yield. Some peculiar features of the total reducible sugar (TRS) release from different biomasses under dilute acid pretreatment and enzymatic hydrolysis are evident from the results presented in Table 5A. The highest sugar release is seen (in terms of concentration in g L⁻¹, and yield in mg g⁻¹ of raw biomass and mg g⁻¹ of delignified biomass) for the biomass of Saccharum spontaneum. Although the biomasses of Saccharum spontaneum, Mikania micrantha and Lantana camara have similar holocellulose contents (as a percentage of raw biomass), the extent of sugar release during their acid pretreatment and enzymatic hydrolysis differs drastically. This difference is attributed to the presence of lignin in the biomass after pretreatment, which hinders the access to hemicellulose and cellulose during acid/enzymatic hydrolysis. Table 5A also lists the total fermentable sugar (TFS) produced by each biomass during pretreatment. A peculiar result is seen for the biomass of Eichhornia crassipes. The biomass has more then 96% cellulose after acid pretreatment, yet the yield of total fermentable sugar is 36.36 g per 100 g of raw biomass. This discrepancy could be a consequence of the high ash content of Eichhornia crassipes, which may hinder both (acid and enzyme) hydrolyses, especially the enzymatic hydrolysis, which has relatively slower kinetics and stronger limitations in the mass transfer. For all the biomasses except Lantana camara, more than 60% of the TFS per unit raw biomass is released during the acid hydrolysis itself. For Lantana camara, about 38% of the TFS per unit raw biomass is released during acid hydrolysis and the rest during the enzymatic hydrolysis. Therefore, the trend in TFS yields per unit raw biomass (i.e. SS > AD > MM > EC > LC) follows the trend in the reducible sugar yield after acid hydrolysis (i.e. SS > AD > MM > EC > LC). The highest TFS yield was obtained for Saccharum spontaneum, while the least yield was obtained for Lantana camara. We assign this result to the high lignin content of LC (∼27%), which may hinder both the acid and enzymatic hydrolysis. The highest TFS yield after acid and enzymatic hydrolysis was observed for Saccharum spontaneum, although the initial holocellulose content was ∼61–64%. This is possibly an outcome of the effective delignification during alkaline pretreatment due to which the resultant biomass comprises almost pure cellulose (∼96%), with a moderate-to-low crystallinity index of 62.5%. These features contribute to the efficacy of the enzymatic hydrolysis, leading to high yields of hexose sugars.

3.3 Assessment of biofuels production potential

The potential of the five invasive weeds as feedstock for biofuels can be assessed on the basis of the maximum theoretical production of alcoholic biofuels, such as ethanol and butanol from fermentation of the hydrolyzates resulting from acid pretreatment and enzymatic hydrolysis. The theoretical yields of ethanol and butanol resulting from the total fermentable sugars (inclusive of both the pentose and hexose hydrolyzates obtained from the acid and enzymatic treatments) from each of the five invasive weeds are listed in Table 5C. It can be seen that the biomass of Saccharum spontaneum had a maximum yield of 29.88 g ethanol and 24.02 g butanol per 100 g of raw biomass. The least yields of 15.63 g ethanol and 12.51 g butanol per 100 g of raw biomass were observed for the biomass of Lantana camara. For the biomass of Mikania micrantha, which is a relatively novel feedstock for bioalcohol production, the theoretical yields were 22.62 g ethanol and 18.19 g butanol per 100 g of raw biomass. The average yield of ethanol from all five invasive weeds was 22.36 g per 100 g of raw biomass, while the average yield of butanol was 17.96 g per 100 g of raw biomass. These yields are at par with the yields reported by Singh et al.³ for Parthenium hysterophorus, i.e. 20 g ethanol per 100 g raw biomass and 16 g butanol per 100 g raw biomass. The similar yields of ethanol and butanol for the five invasive weeds, as compared to Parthenium hysterophorus, are indicative of the promising potential of these weeds for biofuels production – even under un-optimized pretreatment conditions for each of the biomass. This result also highlights that these invasive weeds could be used as multiple feedstocks in a biorefinery, and could be expected to support an overall consistent and sustainable production of alcoholic biofuels. For the efficient production of alcoholic biofuels with a maximum yield per unit raw biomass, it is essential that the hydrolyzates from both the acid and enzymatic treatments can be utilized for fermentation. These hydrolyzates could be fermented separately (as demonstrated by Bharadwaja et al.¹⁹) or simultaneously using mixed cultures. The theoretical bioalcohol yields from the five tested invasive weeds (with a proper utilization of pentose and hexose sugars) are on par with the yields from conventional fermentation substrates, such as corn and molasses, and also second-generation lignocellulosic substrates, such as agro-residues of rice/wheat crops. However, the invasive weeds being waste biomasses (with no other outlet such as cattle feed or domestic fuel) are likely to be available at far cheaper prices than agro-residues. Thus, the processes for biofuels production with invasive weeds as a feedstock are expected to offer significantly lower production costs and contribute to an attractive economy that should interest the stakeholders in the biofuels industry.

4. Conclusion

The present study investigated the bioalcohol production potential of five invasive weeds, namely Arundo donax, Saccharum spontaneum, Mikania mikrantha, Lantana camara and Eichhornia crasspies in terms of the yield of total fermentable sugars from pretreatment and enzymatic hydrolysis. The pretreatment comprised two steps: dilute acid hydrolysis and alkaline delignification with sonication, followed by enzymatic hydrolysis of the pretreated biomass. These pretreatments were carried out at conditions optimized for the waste biomass of Parthenium hysterophorus. Thus, the pretreatment conditions were not optimized specifically for any of the five biomasses considered in this study. Despite this limitation, the average TFS yield from the five invasive weeds was 43.85 g per 100 g of raw biomass, which is on par with the yield from Parthenium hysterophorus for optimized pretreatment conditions. This TFS yield corresponded to a (theoretical) bioethanol yield of 22.36 g and a biobutanol yield of 17.96 g per 100 g of raw biomass. These results vividly demonstrate that a consistent production of alcoholic biofuels could be achieved from the use of invasive weeds as multiple feedstocks, even if the pretreatment conditions are not optimized for each of the weeds. The results of the present study have also demonstrated the feasibility of the large-scale production of alcoholic biofuels employing multiple biomass feedstocks of different invasive weeds.

Acknowledgements

The authors gratefully acknowledge use of SEM facility at Sophisticated Analytical Instruments Facility (SAIF), Northeastern Hill University (NEHU), Shillong. Use of XRD facility (procured through FIST grant No. SR/FST/ETII-028/2010 from Department of Science and Technology, Government of India) and the HPLC and FTIR facility at Department of Chemical Engineering, IIT Guwahati is also acknowledged. One of the authors (AJB) acknowledges help in the experimental work from Mr S. T. P. Bharadwaja. Authors would like to thank the anonymous referees of the manuscript for their meticulous evaluation and constructive criticism.

References

1. A. Ranjan, S. Khanna and V. S. Moholkar, Appl. Energy, 2013, 103, 32–38.
2. S. Singh, S. Khanna, V. S. Moholkar and A. Goyal, Appl. Energy, 2014, 129, 195–206.
3. A. Ekman, O. Wallberg, E. Joelsson and P. Borjesson, Appl. Energy, 2013, 102, 299–308.
4. J. van Eijck, B. Batidzirai and A. Faaij, Appl. Energy, 2014, 135, 115–141.
5. T. Komolwanich, P. Tatijarern, S. Prasertwasu, D. Khumsupan, C. Thanyalak, A. Luengnaruemitchai and S. Wongkasemjit, Cellulose, 2014, 21, 1327–1340.
6. D. Scordia, S. L. Cosentino and T. W. Jeffries, Ind. Crops Prod., 2013, 49, 392–399.
7. T. You, L. Zhang, S. Zhou and F. Xu, Bioresour. Technol., 2014, 167, 574–577.
8. W. F. Anderson, B. S. Dein, S. K. Brandon and J. D. Peterson, Appl. Biochem. Biotechnol., 2008, 145, 13–21.
9. D. Scordia, S. L. Cosentino, J. W. Lee and T. W. Jeffries, Biomass Bioenergy, 2011, 35, 3018–3024.
10. A. Kuila and R. Banerjee, Bioprocess Biosyst. Eng., 2014, 37, 1963–1969.
11. R. Gupta, Y. P. Khasa and R. C. Kuhad, Carbohydr. Polym., 2011, 84, 1103–1109.
12. R. C. Kuhad, R. Gupta, Y. P. Khasa and A. Singh, Bioresour. Technol., 2010, 101, 8348–8354.
13. R. Kataria and S. Ghosh, Energy Sources, Part A, 2014, 36, 1028–1035.
14. G. Chaudhury, L. K. Singh and S. Ghosh, Bioresour. Technol., 2012, 124, 111–118.
15. A. K. Chandel, O. V. Singh, L. V. Rao, G. Chandrasekhar and M. L. Narasu, Bioresour. Technol., 2011, 102, 1709–1714.
16. J. Yan, W. Zhilei, Q. Wang, M. H. He, S. Li and C. Irbis, Bioresour. Technol., 2015, 193, 103–109.
17. P. P. Phothisantikul, R. Tuanpusa, M. Nakashima, T. Charinpanitkul and Y. Matsumura, Ind. Eng. Chem. Res., 2013, 52, 5009–5015.
18. K. Satyanagalakshmi, R. Sindhu, P. Binod, K. U. Janu, R. K. Sukumaran and A. Pandey, J. Sci. Ind. Res. India, 2011, 70, 156–161.
19. D. M. Updegraff, Anal. Biochem., 1969, 32, 420–424.
20. S. T. P. Bharadwaja, S. Singh and V. S. Moholkar, J. Taiwan Inst. Chem. Eng., 2015, 51, 71–78.
21. H. A. Choudhury, R. S. Malani and V. S. Moholkar, Chem. Eng. J., 2013, 231, 262–272.
22. R. Patidar, S. Khanna and V. S. Moholkar, Ultrason. Sonochem., 2012, 19, 104–118.
23. S. Chakma and V. S. Moholkar, AIChE J., 2013, 59, 4303–4313.
24. V. S. Moholkar and M. M. C. G. Warmoeskerken, AIChE J., 2003, 49(11), 2918–2932.
25. T. Sivasankar, A. W. Paunikar and V. S. Moholkar, AIChE J., 2007, 53, 1132–1143.
26. L. Segal, J. J. Creely, A. E. Martin Jr and C. M. Conrad, Text. Res. J., 1962, 29, 786–794.
27. M. A. Lima, G. B. Lavorente, H. K. P. Da Silva, J. Bragatto, C. A. Rezende, O. D. Bernardinelli, E. R. De Azevedo, L. D. Gomez, S. J. McQueen-Mason, C. A. Labate and I. Polikarpov, Biotechnol. Biofuels, 2013, 6, 75.
28. X. F. Sun, F. Xu, R. C. Sun, P. Fowler and M. S. Baird, Carbohydr. Res., 2005, 340, 97–106.
29. R. Kumar, G. Mago, V. Balan and C. E. Wyman, Bioresour. Technol., 2009, 100, 3948–3962.
30. A. J. Borah, M. Agarwal, M. Poudyal, A. Goyal and V. S. Moholkar, Bioresour. Technol., 2016 DOI:10.1016/j.biortech.2016.02.024.

---

Footnote

† Electronic supplementary information (ESI) available: (1) FTIR spectra of native or raw biomass, biomass post dilute acid hydrolysis and biomass post alkaline delignification. (2) X-ray diffractograms of five biomass species (native biomass and the biomass after different pretreatments). See DOI: 10.1039/c5ra27787f

---