Experimental determination of the effects of pretreatment on selected Nigerian lignocellulosic biomass in bioethanol production

In the present study, five lignocellulosic biomass namely, corn cobs (Zea mays), rice husks (Oryza sativa), cassava peels (Manihot esculenta), sugar cane bagasse (Saccharum officinarum), and white yam peels (Dioscorea rotundata) of two mesh sizes of 300 and 425 microns and a combination of some and all of the biomass were pretreated using combined hydrothermal and acid-based, combined hydrothermal and alkali-based and hydrothermal only processes. The raw and pretreated biomass were also characterized by Fourier transform infrared spectroscopy (FT-IR), Brunauer–Emmett–Teller (BET), X-Ray diffraction (XRD), and Scanning electron microscopy (SEM) to determine the effects of the various pretreatments on the biomass being studied. The cellulose values of the raw biomass range from 25.8 wt% for cassava peels biomass to 40.0 wt% for sugar cane bagasse. The values of the cellulose content increased slightly with the pretreatment, ranging from 33.2 to 43.8 wt%. The results of the analysis indicate that the hydrothermal and alkaline-based pretreatment shows more severity on the different biomass being studied as seen from the pore characteristics results of corn cobs + rice husks biomass, which also shows that the combination of feedstocks can effectively improve the properties of the biomass in the bioethanol production process. The FTIR analysis also showed that the crystalline cellulose present in all the biomass was converted to the amorphous form after the pretreatment processes. The pore characteristics for mixed corn cobs and rice husks biomass have the highest specific surface area and pore volume of 1837 m2/g and 0.5570 cc/g respectively.

Pretreatment methods. In this study, three pretreatment methods were adopted to pretreat the biomass for comparison purposes. The pretreatment methods used include combined hydrothermal and acid-based pretreatment, combined hydrothermal and alkaline-based pretreatment, and hydrothermal only pretreatment.
Combined hydrothermal and acid pretreatment. The method used by Utama et al. 22 was adopted in this work with slight modification. H 2 SO 4 (98% analytical grade JHD) with 98.08 g/mol molecular weight was used. 80 ml of the sulphuric acid was measured and transferred into a 2000 ml volumetric flask and distilled water was used to make up the volume to 2000 ml. 0.75 M solution of the sulphuric acid was thus obtained. The prepared biomass (individual and combinations) were then soaked with the prepared acid solution in batches and for each case, about 4 to 6 L of distilled water was used alongside the 2 L of the 0.75 M solution of the H 2 SO 4 based on the absorbing capacity of the biomass. The mixture was then thoroughly mixed in the vessel before being transferred to a pressure pot and allowed to boil for about an hour. It was allowed to cool to ambient temperature before filtering and then kept inside sampling plastics. The filtrate from each batch was also stored separately.
Combined hydrothermal and alkaline pretreatment. 160 g of NaOH pellets were dissolved in a beaker and then moved to a 2000 ml volumetric flask containing 60 ml of H 2 O 2 and then thoroughly mixed in the volumetric flask before making it up to the 2000 ml mark with distilled water. The same procedure as that of acid pretreatment was then followed on all the biomass.
Hydrothermal pretreatment. Hydrothermal only Pretreatment was also done on all the biomass as described in the combined hydrothermal and acid pretreatment as control.
Characterization of the raw and pretreated biomass. Characterization of biomass is essential to establish its capability for bioethanol production 23 . The raw and pretreatment biomass samples were then subjected to characterization to establish the impacts of the different pretreatment methods on the biomass meant for fermentation for bioethanol production. The following physicochemical analysis was carried out on the raw and pretreated biomass: proximate analysis, ultimate analysis, FT-IR, XRD, BET, and SEM. The characterization was done at the multiuser laboratory of Ahmadu Bello University, Zaria Nigeria, and the Chemical Engineering Department laboratory of Federal University of Technology Minna, Nigeria.
Proximate analysis. The gross composition of the biomass pre and post pretreatment was determined using the proximate analysis to ascertain their moisture content, ash content, lignin content, cellulose content, and hemicellulose content.
Determination of lignin content. The acid detergent fiber (ADF) residue earlier obtained was immersed in chill sulphuric acid. The mixture was blended to a smooth paste to breakdown all the arms. The residue in the crucible was dehydrated for 24 h at 100 °C and then allowed to cool to around ambient temperature. It was then weighed and labeled (W1). The crucible plus oven-dried residue was moved to a muffle furnace fixed at 550 °C to ash for 3 h till a white greyish residue was obtained, cooled in a desiccator, and then weighed and labeled (W2). The lignin content was then calculated using the equation; Determination of holocellulose. A solution of 80 ml acetic acid and 1 g of sodium chloride were put into 2.5 g of extractive free sample (after extracting the sample) in a water bath hourly for six (6) h in a process known as chlorinating. Subsequently, after six (6) h of chlorinating, the samples were then allowed to stay for a while in a water bath to lower the temperature, and then the holocellulose was filtered using a Buchner funnel. The initial and final weight of the holocellulose were taken and the holocellulose content was calculated using the following formula: www.nature.com/scientificreports/ where: W1-the weight of the sample before the process, W2-the weight of the sample after the process.
Determination of hemicellulose. 2 g of holocelloluse which had earlier been dried in the oven was transferred into a 25 ml glass beaker, afterwards, 10 ml of 17.5% sodium hydroxide (NaOH) solution was mixed with the holocelloluse, it was then made to stay in a water bath, a flat end glass rod was employed in stirring it for it to be soaked with the NaOH, after adding of the initial portion of 17.5% (NaOH) solution every five (5) min, additional 5 ml of NaOH solution was then introduced and thoroughly agitated using a glass rod. Sodium hydroxide was continuously added until all the NaOH solution was used up. Thereafter, the mixture was made to stay in a water bath for thirty (30) min.
Determination of cellulose. The cellulose content of the biomass was determined using the formula: Determination of moisture content. The moisture content of the biomass was gotten by the method of oven drying. This was done at a temperature of 103 ± 2 °C following ASTM D1037 (1991). The moisture content was thereafter determined by using the equation: where Wi-the initial mass of the sample, Wf-the final mass of the sample.
Determination of the ash matter. The Ash matter in the biomass samples was determined by the method described in ASTM D2017 (1998). 1 g each of the samples was put in a pre-weighed crucible and was then burnt in a muffle furnace at 760 °C until ashing was completed after which the container was then moved into a desiccator to lower the temperature. Three replicates were made. The samples were then weighed after cooling. The ash matter was determined by using the following equation: where: W0 = Weight of the container, W1 = weight of the container + biomass sample before burning, W2 = weight of the container + biomass sample after burning.
FT-IR, SEM, BET and XRD characterization of the raw and pretreated samples. The virgin and pretreated biomass were also characterized by Fourier transform infrared spectroscopy (FT-IR), Brunauer-Emmett-Teller (BET), X-Ray diffraction (XRD), and Scanning electron microscopy (SEM).
Fourier transform infrared (FT-IR) spectroscopy. The main function of FT-IR spectroscopy is for the detection of the different functional groups present in the virgin and pretreated biomass 24 . The main result of FTIR assays mainly deals with the lignin content of the biomass 25 . For this work, Agilent Cary 630 FTIR Spectrometer was used to characterize the virgin and the pretreated biomass. The FTIR was analyzed using Agilent MicroLab PC software equipped with the equipment.
Scanning electron microscopy (SEM). This is an analytical procedure that scans a sample with an electron beam to produce a magnified image for assessment. SEM analysis makes the samples structures to be assessed and their elemental make-up determined 26 . The Phenom ProX desktop SEM with a magnification range of 20-100,000× and element detection range of C-Am and acceleration voltage of 10 kV was used in this study.
Brunauer-Emmett-Teller (BET) analysis. The specific surface area of a biomass sample and the pore size distribution can be measured with this analysis, which can be used to forecast the dissolution rate as this rate is proportional to the specific surface area and the surface area, in turn, can be used to forecast bioavailability. The Nova 4200e BET analyzer was used in this study. The BET analyzer used nitrogen as an analysis gas. The outgas time was 3 h at a temperature of 250 °C. The pressure tolerance for the analysis was 0.100/0.100 (ads/des). The equilibrium time was 60/60 s and the equilibrium time out was 240/240 s. The analysis time was 111.8 min.
X-ray diffraction (XRD) analysis. This is a swift analytical procedure used to detect phases of lignocellulosic biomass and can also provide information on the cell composition of the biomass. The lignocellulosic biomass meant for analysis should be in fine particulate form and thoroughly mixed before the mean bulk components are determined. This test is centered on the applied intrusion of monochromatic X-rays and a lignocellulosic biomass sample. The X-ray is generated by the Cathode ray tube, which has a monochromatic radiation produc- www.nature.com/scientificreports/ tion. The X-rays are also collimated to meet and directed at the lignocellulosic biomass sample. The incident rays interact with the lignocellulosic biomass sample to enable the creation of positive interference and a diffracted ray after conditions meet Bragg's law (nλ = 2 d sin θ). The diffracted X-rays are then detected, sorted out and calculations carried out. The lignocellulosic biomass samples, with a range of 2θ angles for all likely diffraction paths of the lattice to be achieved, as the grounded lignocellulosic biomass is unsystematically orientated. As a result of every mineral having a set of exclusive d-spacings, diffraction peaks were converted to d-spacing to spot the minerals existing in the biomass. A comparison of the d-spacings with a standard reference pattern was then carried out.
Life cycle assessment (LCA). The LCA is equipped with the data collection section, which is meant for the identification and accounting of all the input and output of a process. In building and analyzing LCA models, thorough, holistic, and generally admitted stock data is needed for the materials and processes used. SimaPro has been a leading LCA software package that is used to extract considerable amounts of related data from diverse production handling reports, which includes chemical and food production facilities and brings in the data from a varied collection of accessible databanks. Ecoinvent v3 LCI Databank was also employed in this study. This is the most frequently consulted and referenced 27 .
Economic consideration of the bioethanol production process. Bioethanol production process economic assessment usually includes the estimation of yields, financial considerations, and costs related to the investment and operating costs. To determine the efficiency of the pre-treatment process, the bioethanol yield from the hybrid biomass (cassava peels and yam peels biomass) for the acid-based and alkali-based pretreatments were evaluated using the model used by Solarte-Toro et al. 28 . The assessment also included obtaining the utility costs, which comprise the cooling water, steam, fuel, and electricity required for pre-treatment, fermentation, and distillation of the fermented hydrolysates, cost of raw materials and transportation, labour costs, operating charges, Fermentation and distillation equipment Costs, administrative costs, cost of enzymes and depreciation. The data were evaluated using engineering economics and descriptive statistics. The bioethanol cost was then compared with the current market price of bioethanol as obtainable in Nigeria.

Results and discussion
Effects of different pretreatment rigorousness on the physicochemical properties of biomass. The results of the proximate analysis on the raw and pretreated biomass samples of different particle sizes are as shown in Tables 1, 2 29 . The pretreated biomass shows a significant difference from the raw biomass with values ranging from 33.2 to 43.8 wt%. The values of the cellulose content increased slightly with the pretreatment and that of the lignin content decreasing significantly with pretreatment. High cellulose and low lignin contents is a very desirable quality in lignocellulosic biomass for bioethanol production 30 . The hydrothermal and alkaline- This also conforms to the findings of Sabiha-Hanim et al. 31 and Chang et al. 13 that both found out from their works that alkaline pretreatment is more efficient in lignin removal and considerably increases the degradability of cellulose even if only some part of the lignin is removed. The ash content of the pretreated biomass except for rice husks biomass which remained the same were all reduced. It is worthy of note that elevated ash content is a problem as the ash particles in the biomass take up more steam, H 2 O, dilute acid solution, or diluent than the relatively larger lignocellulosic fibers 32 . The consequence of the ash content of the lignocellulosic biomass could also be seen on the moisture content results for the pretreated biomass. Since the ash particles in the biomass absorb the steam, water, dilute acid, and other solvents during pretreatment, the moisture content of the biomass increased slightly after pretreatment with the alkali pretreated biomass having the highest moisture content, again confirming the superior severity of the alkali pretreatment over the other pretreatment methods adopted in this study. The hemicellulose content also shows the same pattern by increasing slightly after the different pretreatments, with the values of the pretreated biomass ranging from 33.5 to 43.5 wt%. For the mixed biomass (cassava peels plus yam peels, corn cobs plus rice husks, and all five biomass combined), there is not much difference in the outcome of the different pretreatment methods. The values obtained for their cellulose and hemicellulose content for the three pretreatments falls within a very narrow range. Total delignification of the biomass is not easy due to the position of lignin in the macromolecular structure 33 .
The proximate analysis parameters for the three pretreatment methods adopted in this study were analyzed statistically using analysis of variance (ANOVA) of Microsoft Excel and the results are as shown in Table 5a-d. The p-values in the ANOVA results for the ash content, lignin content, hemicellulose content, and cellulose content were all greater than 0.05, showing that there is no significant difference in the effects of the different pretreatments on the biomass in this study. The statistical significance of the data obtained for the proximate analysis parameters was tested by F-test and shows that the effects of the three pretreatment methods was highly significant as suggested by the model F values on the tables.
Effect of pretreatment on the composition of biomass. The major purpose of pretreatment is the breaking down of the lignin make-up and the disruption of the crystalline make-up of cellulose for improving enzyme ease of access to the cellulose in the hydrolysis step 34 . Tables 6,7,8,9,10,11,12,13 show the pore characteristics of the different biomass being analyzed in this study. The BET adsorption isotherms are also shown in Figures S1 to 8. The results show that generally for all the biomass under consideration, there is a significant reduction in specific surface area and pore volume with the different pretreatment methods carried out on them, with hydrothermal only pretreatment showing the least values. The results also show that cassava peels biomass of 300 microns particle size has the highest specific surface area and pore volume of 819.6 m 2 /g and 0.2031 cc/g  www.nature.com/scientificreports/ respectively for individual pretreated biomass, thereby upholding the assertion by many researchers that cassava peel biomass is a very good feedstock for bioethanol production 35,36 . For the combined biomass, the combined corn cobs and rice husks biomass with specific surface area and pore volume of 1837 m 2 /g and 0.5570 cc/g respectively show promising potential for improved yield of bioethanol after the fermentation process.
FTIR spectroscopy of raw and pretreated biomass. FTIR is mainly used to characterize the biomass based on the organic groups present 37 . The focal point of FTIR analysis is the lignin content, which is an aromatic biopolymer composed chiefly of phenylpropane substituted components attached to form a giant molecule of Table 7. Pore characteristics of corn cobs biomass (raw and pretreated).

Sample
Specific surface area (m 2 /g) Pore volume (cc/g)   www.nature.com/scientificreports/ non-consistent crystallinity and optical activity 38 . The FTIR spectra of the untreated (raw) as well as the treated biomass are shown in Figs. 1 and S9 to S13. Table 14 displays the functional groups and vibration modes of the raw and pretreated biomass at standard temperature. The absorption band of the raw and pretreated biomass between 3200 and 3600 cm −1 is usually attributed to the O-H stretching vibrations of alcohols, carboxylic acids, and hydroperoxides 39,40 . The FTIR spectrum shows a fingerprint region of 1420-670 for the source identification of the biomass. The results show that the O-H stretching of the hydroxy group of alcohol falls between 3693 and 3008 cm −1 for acid pretreated cassava peels and acid pretreated rice husks respectively. The methyl group of alkanes has a band of 2926 to 2855 cm −1 for water pretreated corn cobs and alkali pretreated corn cobs respectively. C=O stretching vibration can be attributed to ketones with a wavenumber of between 1636 and 1606 cm −1 . The band 1457 cm −1 shows the C-H bending or scissoring of alkanes found in acid pretreated corn cobs, cassava peels, and a combination of yam and cassava peels. Other organic compounds detected include ether, ether, and Β-glucosidic bonds (864 cm −1 ) between sugars 41 . For all the biomass, a very sharp peak of between 1006 to 1028 cm −1 of C-O-C is noted. The strong and sharp peak is attributed to the ether group. The hemicellulose content of the biomass can be ascribed to the wavebands at 1710 and 1028 cm −1 . There are also very sharp peaks noticed on the pretreated combined corn cobs and rice husks biomass, this is an indication of the fortification of the properties of the biomass by their combination.

Pore diameter (nm) V micro /V Total (%) S BET S micro S Ext V Total V micro
Inhibitor analysis of the different pretreatment process. The results of the proximate analysis as shown in Table 2 indicate that the combined hydrothermal and acid pretreated biomass has a very high recovery of hemicellulosic sugars with increased enzymatic convertibility. However, the process brings about the formation of inhibitory derivatives such as aliphatic carboxylic acids, furans, among others. The combined hydrothermal and alkali pretreatment was able to remove lignin and a small fraction of hemicelluloses, however with the formation of some side products such as acetic acid, hydroxy acids among others. Table 15 shows a summary of the inhibitory side products formed during the different pretreatment methods in this study. In the work by Olsson et al. 42 , to reduce the effects of the inhibitors on the hydrolyzates for fermentation, two methods were proposed namely; detoxification and adaptation of the microorganism to the lignocellulosic hydrolysate. The latter is more cost-saving than the former. Other methods include overliming 43 , charcoal adsorption, and ion exchange 44 . Structural changes in the raw and pretreated biomass. The scanning electron micrograph (SEM) was employed to compare, study and analyze the untreated samples as well as morphological changes that had   www.nature.com/scientificreports/ occurred on the pretreated samples as a result of the different pretreatments carried out on them. Figures 2 and S14 to S20 show the images of the morphological analysis of the raw and pretreated biomass by SEM. The images show that all the raw biomass had a smooth intact structure with a rigid and fibrillary morphology, which had not been damaged by the crushing and grinding of the biomass. The images also revealed interesting transformations after the pretreatment processes were carried out on them. The morphologies of the pretreated biomass show that they were broken down and fragmented with the hitherto compacted and finely divided surfaces unsettled. The SEM images of the pretreated biomass also showed a cluster of globe-like micro grains deposited over copious particles, this is an indication that cellulose and hemicellulose were disintegrated to a reasonable extent during the pretreatment processes 45 . However, it can be observed that the images of the alkali pretreated biomass show more of the globe like clusters, indicating the severity of alkali pretreatment over the other pretreatment processes carried out in this study.  46 in their work on lignocellulosic biomass. Lignin, which is a non-saccharide fraction of biomass is more chemically rigid than the saccharide fraction, composed of cellulose and hemicellulose. The lignin was moderately disintegrated, and the initial properties of the grains were conserved. The images also show that pore volume with the surface area was considerably enlarged after the pretreatment process with the alkali pretreatment showing the highest severity. This observation was also supported by the results obtained from the BET analysis. X-ray diffraction (XRD). The X-ray diffraction patterns of the raw and pretreated biomass are illustrated in Fig. 3 and Figures S21 to S26. As can be seen, all the diffractograms showed the typical XRD peaks of cellulose. For the raw biomass, a sharp peak at 2θ values of between 18 and 22° was observed owing to the presence of crystalline cellulose in the biomass samples. This conforms to the results obtained from previous research 47 . The pretreated biomass samples showed diffractograms with peaks of somewhat reduced intensities, a demonstration of incomplete degradation of the cellulose with pretreatment 48 . The pretreated biomass samples had a broader peak at 2θ values of between 22 and 24°, which is an indication of atomic order in them. The non-existence of sharp peaks proves the amorphous texture of the pretreated biomass samples 49 . As opposed to amorphous cellulose, crystalline cellulose is more recalcitrant to enzymatic and microbial reactions 50,51 . Environmental considerations. Many factors influence the determination of the life cycle effects from bioethanol production from different biomass; this includes cultivation of the feedstock, production of the enzymes and chemical, pre-treatment, fermentation, delivery, and utilization of the bioethanol. Emissions to air, soil, and water during the pre-treatment process were assessed using SimaPro. The environmental impact categories include global warming potential (GWP)/climate change, eutrophication (EP), acidification (AP), photochemical oxidation demand (POD), and marine and human ecotoxicity. The emissions from the pre-treatment processes can be classified in terms of corresponding quantities using the CML (Centre of Environmental Science at Leiden University) method 52 . Potential for global warming is measured in kg of CO 2 (eq), eutrophication is measured in kg of PO 4 -3 (eq), acidification is measured in kg of SO 2 (eq), photochemical oxidation demand is measured in kg of C 2 H 4− (eq), while marine and human ecotoxicity are measured in kg of dichlorobenzene (1,4 C 6 H 4 Cl 2 ) (eq). The GWP and the AP were considered on the mixed biomass of cassava peels and yam peels in this work.
Global warming potential (GWP). The global warming potential (GWP) of a process or product and its consequent effect on the climate, have been the current subject of debate and regulation in the environmental performance assessment. GWP is an enumerated amount of the global mean comparative radiative driving effects  With an average value of 15.82 kg CO 2 (eq), the alkaline pre-treatment using sodium hydroxide shows the highest discharge of GHG emissions, while acid pretreatment using dilute sulphuric acid generated an average 8.68 kg CO 2 . Climate change is a very important consideration in lignocellulosic bioethanol production as one of the reasons for utilizing bioethanol for energy purposes is the reduction in GHG emissions.
Acidification potential (AP). Acidification stems from the anthropogenic emissions involving sulfur (iv) oxide (SO 2 ), nitrogen oxides (NO x ), and ammonia (NH 3 ). The assessment of the acidifying potential of the sulphuric acid and sodium hydroxide used for pre-treatment of one of the biomass in this study (cassava peels biomass), show that sulphuric acid emitted 0.035 kg SO 2 (eq), a value lower than that of sodium hydroxide emission, which  www.nature.com/scientificreports/ was 0.087 kg SO 2 (eq). This result is an indication that the acid pre-treatment is more preferable in terms of environmental degradation owing to the effect of the biomass pre-treatment.

Economic consideration results and analysis.
To be viable and economically acceptable, the expenditure for the processing of lignocellulosic biomass to fuel must not be up to the current gasoline cost. This is achievable owing to the efforts of researchers at improving the effectiveness of the biomass processing technologies. The cost of feedstock, feedstock pre-treatment, and enzymes are important considerations for low-cost ethanol production. The use of hybrid (mixed) feedstocks and large-scale processing facilities coupled with low-cost feedstock as well as effective cellulases helps in making the process cost-effective. In converting biomass to bioethanol, some inputs result in environmental costs, the output from the process such as electricity generations and sales of the produced bioethanol would result in the recouping of the expenses of the production process. Table 16 shows the results of the techno-economic assessment of bioethanol produced from acid-based pre-treated cassava and yam peels biomass mixture. The bioethanol price of 0.41 USD/l is a good deal as it compares favorably well with the 0.45 USD/l price of ethanol in the Nigerian market. The produced bioethanol could also augment gasoline from crude oil, this would also reduce the drastic effects of the combustion of gasoline on the environment in terms of emissions and costs. The major advantage derivable from the use of biomass in bioethanol production is the limiting of greenhouse gases' environmental pollution 53 . The lignocellulosic bioethanol production, therefore, comes with more benefits in the long run in terms of economic and environmental considerations.

Conclusion
Bioethanol production from lignocellulosic biomass such as corn cobs, rice husks, cassava peels, yam peels, sugar cane bagasse among others, is an emerging technology in the renewable energy field. However, these biomass must be pretreated before the fermentation process. The main goal aimed to be achieved by carrying out pretreatment is the breaking down of the lignin structure and the interruption of the crystalline make-up of cellulose for enhancing enzyme accessibility to the cellulose during the hydrolysis stage. Selected biomass was sourced locally, dried, and sieved into two mesh sizes and then pretreated using hydrothermal and acid-based based, hydrothermal, and alkali-based and hydrothermal only processes. The raw as well as the pretreated biomass samples were then characterized by proximate analysis, SEM, FTIR, XRD, and BET. The cellulose values of the raw biomass range from 25.8 wt% for cassava peels biomass to 40.0 wt% for sugar cane bagasse. The pretreated biomass shows a significant difference from the raw biomass with values ranging from 33.2 to 43.8 wt%. The cassava peels biomass of 300 microns particle size has the highest specific surface area and pore volume of 819.6 m 2 /g and 0.2031 cc/g respectively for individual pretreated biomass, while for the combined biomass, the combined corn cobs and rice husks biomass with specific surface area and pore volume of 1837 m 2 /g and 0.5570 cc/g www.nature.com/scientificreports/ respectively. The combined corn cobs and rice husks biomass show a promising potential for improved bioethanol yield after fermentation. The organic compounds present in the biomass include hydroxyl, alkanes, methoxy, carboxylic acids, ketones among others. The use of a combined biomass feedstock would be preferred as this will give greater flexibility in the bioethanol production from lignocellulosic biomass, and as was shown from this work, the combination of different biomass materials can result in favorable properties of the combined biomass feedstock after pretreatment.