Physical–chemical–morphological characterization of the whole sugarcane lignocellulosic biomass used for 2G ethanol production by spectroscopy and microscopy techniques

Renewable Energy 87 (2016) 607e617

Contents lists available at ScienceDirect

Renewable Energy journal homepage: www.elsevier.com/locate/renene

Physicalechemicalemorphological characterization of the whole sugarcane lignocellulosic biomass used for 2G ethanol production by spectroscopy and microscopy techniques Sandra C. Pereira a, Larissa Maehara a, b, Cristina M.M. Machado c, Cristiane S. Farinas a, b, * ~o Carlos, SP, Brazil Embrapa Instrumentation, Rua XV de Novembro 1452, 13560-970, Sa ~o Carlos, Rodovia Washington Luiz, Km 235, 13565-905, Sa ~o Carlos, SP, Brazil Graduate Program of Chemical Engineering, Federal University of Sa c ~o Biolo gica s/no, 70770-901, Brasília, DF, Brazil Embrapa Agroenergy, Parque Estaça a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 May 2015 Received in revised form 15 September 2015 Accepted 27 October 2015 Available online xxx

The natural recalcitrance of sugarcane lignocellulosic biomass remains a challenge for second generation (2G) ethanol production. Here, the physicalechemicalemorphological characteristics of the whole sugarcane lignocellulosic biomass (including bagasse, straw, and tops) from commercial sugarcane varieties were evaluated before and after dilute sulfuric acid pretreatment, in order to help predict the behaviors of these materials during the 2G ethanol process. Analyses using NMR, FTIR, XRD, and SEM showed that the properties of the sugarcane varieties evaluated here were very similar. The crystallinity index values calculated from the XRD results were also similar for the different residue fractions, and were higher after pretreatment due to the removal of hemicellulose. The lignin and crystalline cellulose FTIR absorption bands were most intense for bagasse, followed by straw and tops. NMR analysis identified the presence of skeletal aromatic and methoxyl groups, attributed to the lignin structure, with the intensity of the signals following the order: bagasse > straw > tops. SEM images showed that structural disruption followed the order: tops > straw > bagasse. The spectral and morphological differences helped to elucidate the characteristics that made the bagasse fraction of the sugarcane residue less susceptible to enzymatic saccharification. Differences between the spectra for straw and tops indicated that the straw was less easily digested by enzymatic action, as also indicated by the morphological analysis. The results demonstrate that the combined use of spectroscopy and microscopy techniques can contribute to understanding the behavior of different biomasses intended to be used for 2G ethanol production. © 2015 Elsevier Ltd. All rights reserved.

Keywords: Sugarcane lignocellulosic biomass Cellulosic ethanol Compositional analysis Sugarcane straw Sugarcane bagasse Sugarcane tops

1. Introduction Second generation (2G) ethanol production using lignocellulosic biomass offers an attractive means of improving productivity in the biofuels sector. In this process, the enzymatic hydrolysis is still considered one of the most challenging steps, and its efficiency is closely related to the performance achieved in the biomass pretreatment step. Understanding of the physicalechemicalemorphological features of the lignocellulosic biomasses and elucidation of the components and mechanisms influencing their recalcitrance are needed for the design of suitable processes. However, traditional methods of compositional analysis

* Corresponding author. Embrapa Instrumentation, Rua XV de Novembro 1452, 13560-970, S~ ao Carlos, SP, Brazil. E-mail address: [email protected] (C.S. Farinas). http://dx.doi.org/10.1016/j.renene.2015.10.054 0960-1481/© 2015 Elsevier Ltd. All rights reserved.

in terms of the contents of cellulose, hemicellulose, and lignin are not sufficient to understand the complex multi-scale structures of the lignocellulosic materials [1e3]. Moreover, most conventional methods used to determine the composition of lignocellulosic materials are lengthy and laborious, requiring harsh reagents and generating large amounts of waste [4]. For these reasons, there is increasing use of faster techniques that enable more comprehensive and accurate characterization, together with the development of effective screening tools that can be applied to different types of lignocellulosic biomass. Instrumental methods that have been used for this purpose include nuclear magnetic resonance spectroscopy (NMR), Fourier transform infrared spectroscopy (FTIR), X-ray diffraction spectrometry (XRD), and scanning electron microscopy (SEM), which are capable of generating comprehensive qualitative and quantitative data [5e10]. Spectroscopy and microscopy techniques have been used to compare different lignocellulosic materials in terms of their

608

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

potential for bioconversion into ethanol. For instance, Li et al. [11] explored the susceptibility to enzymatic hydrolysis of celluloses from five different biomasses, using XRD to determine the crystallinity index (CrI, %). Duan et al. [12] evaluated the effect of composition and structure of different poplar lines (transgenic and wild type) on the production of bioethanol, using FTIR to determine the lateral order index and total crystallinity index, and SEM for analysis of morphological features. This enabled better understanding of the results of enzymatic hydrolysis and distinction among the raw materials employed. Similarly, Lima et al. [13] were able to demonstrate the potential of novel Brazilian sources of biomass for bioethanol production. Structural differences and the contents of cellulose, hemicellulose, and lignin were evaluated using SEM and solid-state 13C NMR. Techniques including NMR, FTIR, XRD, and SEM have been widely employed to evaluate the structural changes caused by different pretreatments of sugarcane lignocellulosic biomass. The combined use of NMR and FTIR can provide detailed structural elucidation of the effects of pretreatment on this material, by observing the changes in the analytical signals associated with the main components (amorphous and crystalline cellulose, hemicellulose, and lignin). XRD can be used to estimate the crystallinity index, and SEM provides information on the morphology of the biomass. These techniques have been employed to investigate the effects on sugarcane bagasse structure of different pretreatments (sequential acid-base, oxalic acid fiber expansion, steam processing in the presence of CO2 and SO2, and delignification) [14e17]. Elsewhere, FTIR, XRD, and SEM analyses have been used to determine the structural changes in sugarcane residues caused by hydrothermal, dilute acid, and sequential hydrothermal-delignification pretreatments [18e20]. Sindhu et al. [2,21] performed a physicalechemical characterization of sugarcane tops pretreated with dilute acid and alkali in order to optimize enzymatic saccharification for bioethanol production. In a previous study, we undertook a systematic comparison of the use of the whole residual lignocellulosic biomass (including bagasse, straw, and tops) from four commercial sugarcane varieties (SP79-1011, RB867515, SP81-3250, and RB92579) for the production of 2G ethanol [22]. The parameters assessed were chemical composition, susceptibility to saccharification, and fermentability, considering the responses of each sugarcane residue in the different steps of the process (dilute acid pretreatment, enzymatic hydrolysis, and alcoholic fermentation). No significant differences among the varieties of sugarcane were observed, but the residue types showed significantly different responses to the process conditions [22]. Therefore, in order to understand and elucidate the different responses of the sugarcane residue fractions (bagasse, straw, and tops) in the 2G ethanol production process, the aim of the present study was to carry out a systematic physicalechemicalemorphological characterization of the whole sugarcane lignocellulosic biomass, before and after the dilute sulfuric acid pretreatment step. The techniques employed included microscopy (SEM) and spectroscopy (NMR, FTIR, and XRD). It is worth noting that, to the best of our knowledge, the present work is the first to perform such a systematic study of the whole sugarcane lignocellulosic biomass (including bagasse, straw, and tops) from different varieties of sugarcane.

raw materials were obtained from the processing of four commercial varieties of sugarcane (SP79-1011, RB867515, SP81-3250, and RB92579, symbolically represented in this work by K, M, Q, and X, respectively). Prior to use, the materials were dried at 45
C to a moisture content of less than 10%. The samples were then milled and sieved to granulometry smaller than 2 mm prior to storage at room temperature until used in the experiments. Details concerning the agronomic characteristics of these sugarcane varieties and their compositional analysis (in terms of the contents of cellulose, hemicellulose, lignin, and ash) are provided in our earlier work [22].

2. Materials and methods

Changes in the functional groups due to the dilute sulfuric acid pretreatment were evaluated by FTIR analysis. The spectra of the whole sugarcane lignocellulosic biomass (untreated and pretreated) were obtained with a Bruker Vertex 70 FTIR spectrometer equipped with an attenuated total reflectance (ATR) accessory. During the analysis, the samples were pressed against the diamond crystal of the ATR device. The scanning range was from 4000 to

2.1. Sugarcane lignocellulosic biomass The three residue fractions that composed the whole sugarcane lignocellulosic biomass (bagasse, straw, and tops) were kindly provided by the Sumaúma mill (Marechal Deodoro, Brazil). These

2.2. Dilute sulfuric acid pretreatment The lignocellulosic sugarcane residues were submitted to a pretreatment in sulfuric acid solution (1.5%, w/w), at a solid to liquid ratio of 1:10, in an autoclave at 121
C for 30 min. After returning to ambient temperature, the mixtures were filtered under vacuum to recover the solid fractions, with suitable disposal of the liquid fractions. The solids were washed with distilled water to remove the soluble components adhered to the surfaces. At the end of the procedure, the pretreated materials were dried in an oven at 45
C until reaching moisture content below 10%, and then stored at room temperature for later use. 2.3. Physicalechemicalemorphological analysis The whole sugarcane lignocellulosic biomass (bagasse, straw, and tops) obtained from the processing of the four commercial varieties (K, M, Q, and X), before and after dilute acid pretreatment (H2SO4, 1.5% w/w, 1:10 solid:liquid ratio, 121
C, 30 min), was examined by microscopy and spectroscopy techniques, as described below. No additional sample preparation was performed, except for coating of the specimens with a gold layer prior to the SEM analyses. 2.4. X-ray diffraction spectroscopy (XRD) The inherent crystalline nature of the whole sugarcane lignocellulosic biomass (untreated and dilute acid pretreated) was studied using a Shimadzu LAB-X XRD-6000 X-ray diffractometer, with Cu-Ka radiation (l ¼ 1.54 Å) generated at a voltage of 30 kV and a current of 30 mA. The 2q scan range was from 3
to 60
, with a scanning rate of 2
/min and a step size of 0.02
. For the purposes of comparison, the X-ray diffractograms were normalized by the intensity of the diffraction peak at 2q ¼ 22
. The crystallinity index was determined by means of the empirical method described by Segal et al. [23], according to:

CrI ¼

ðI002  Iam Þ *100; I002

(1)

where CrI represents the relative degree of crystallinity (%), I002 is the peak intensity of the 002 crystal plane at 2q ¼ 22
, and Iam is the peak intensity of the amorphous phase at 2q ¼ 18
. 2.5. Fourier transform infrared spectroscopy (FTIR)

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

600 cm1, with a resolution of 4 cm1 and 32 scans per sample. Background scanning and correction were carried out before the acquisition of each spectrum. For comparative purposes, the FTIR spectra were normalized using the intensity of the band at 2852 cm1. 2.6. Solid-state (NMR)

13

C nuclear magnetic resonance spectroscopy

The effects of the dilute sulfuric acid pretreatment on the whole sugarcane lignocellulosic biomass were also evaluated using solidstate 13C NMR with cross-polarization and magic angle spinning (13C CP MAS ssNMR). The spectra of untreated and pretreated samples were recorded with a Bruker Avance III HD 400 MHz spectrometer equipped with a solid-state MAS probe, with two of the channels configured for 1H and 13C. The operational conditions were: 2048 scans, spinning rate of 10 kHz, acquisition time of 40 m s, and contact time of 4.5 m s. The samples were prepared by packing into 4 mm cylindrical zirconia MAS rotors. The chemical shifts were calibrated using the methyl group resonance (at 17.3 ppm) of an external hexamethylbenzene standard. To facilitate comparison, the NMR spectra were normalized using the intensity of the signal at 105 ppm. 2.7. Scanning electron microscopy (SEM) The morphology of the whole sugarcane lignocellulosic biomass was analyzed by scanning electron microscopy before and after the dilute sulfuric acid pretreatment. Prior to analysis, the samples were fixed onto aluminum stubs using carbon tape and then coated with a gold layer using a Leica EM SCD050 sputter coater system. All the prepared samples were kept in a desiccator until the time of analysis. Sample imaging employed a JEOL JSM-6510 scanning electron microscope operated with an acceleration voltage of 10 kV and a working distance of 10 mm. A careful examination was performed of the entire specimen, with a reasonable number of images recorded for different areas, at various magnifications and with at least 10 images per sample, in order to give confidence in the observations. 3. Results and discussion 3.1. Physicalechemicalemorphological analysis There is a need for an effective screening of new feedstocks used for cellulosic ethanol production, together with improved understanding of the physicalechemicalemorphological characteristics of these materials, before and after the pretreatment step. This requires the use of advanced analytical techniques to obtain the qualitative and quantitative data necessary to ensure the efficient bioconversion that is crucial to the development of the biofuels sector. The implementation of spectroscopy and microscopy methods to investigate the raw materials used for the production of biofuels has increased in recent years. Some of these techniques are simpler and faster than conventional methods. Others are time consuming and expensive. An important point is that the methods employed in this study (XRD, FTIR, NMR, and SEM) are relatively simple to perform and require little or no sample preparation. With the exception of NMR, all the techniques are fast, and the sample is not lost (with the exception of SEM). The information obtained from different techniques, as used here, is important for technological advancement because it can facilitate the establishment of methodologies for routine analysis of different lignocellulosic biomasses that could potentially be employed as raw materials in bioprocesses.

609

The contents of cellulose, hemicellulose, lignin, and ash (% w/w, dry weight basis) in the whole sugarcane lignocellulosic biomasses used in this study, before and after the dilute acid pretreatment step, were determined using a standard compositional analysis method and have been reported previously [22]. In general, the compositional analysis showed that there were no substantial differences among the sugarcane varieties, for both untreated and pretreated biomass. On the other hand, there were appreciable differences among the compositions of the sugarcane lignocellulosic residue fractions (bagasse, straw, and tops). Nevertheless, this difference was less marked for the pretreated sugarcane biomass than for the untreated biomass. Hence, the dilute acid pretreatment was able to increase the similarity of the sugarcane residues in terms of their chemical composition, regardless of the variety [22]. Furthermore, the conditions employed in the pretreatment step were effective for solubilizing the hemicellulose fraction, removing on average (among all the varieties) approximately 82% for straw, 80% for tops, and 74% for bagasse. The dilute acid pretreatment also led to an enrichment of the materials in terms of cellulose and lignin, as well as a substantial removal of ash (approximately 57% for straw, 50% for tops, and 37% for bagasse). After the pretreatment, the tops presented the highest cellulose and lowest lignin contents [22]. Overall, the compositional analysis showed that for both untreated and pretreated sugarcane biomass, the contents of cellulose, hemicellulose, lignin, and ash were in agreement with the ranges previously described in the literature [19,21,24]. In a recent study, Rocha et al. [25] conducted a chemical and elemental characterization of 60 different bagasse samples and also concluded that the differences among the sugarcane varieties were not significant. Nonetheless, since lignocellulose is a highly complex raw material that is innately recalcitrant, it is evident that detailed investigations should be carried out using other analytical methods. In the following sections, the use of spectroscopy and microscopy techniques will be presented. 3.2. X-ray diffraction (XRD) Fig. 1 presents the X-ray diffraction patterns for the whole sugarcane lignocellulosic biomasses (bagasse, straw, and tops) from the four commercial varieties, before and after the dilute acid pretreatment step. As can be seen, all the samples (untreated and pretreated) showed the typical XRD peaks of cellulose. There were no pronounced differences among the XRD profiles of the residue fractions (bagasse, straw, and tops) or the sugarcane varieties (K, M, Q, and X) (Fig. 1), indicating that the dilute acid pretreatment did not cause any drastic changes in the crystalline nature of the sugarcane lignocellulosic biomass. In other words, the XRD patterns suggested that the ordered structure of the crystalline region in the remaining cellulose was not disrupted by the action of the dilute acid. Similar XRD profiles for untreated and dilute acid pretreated sugarcane biomass were reported by Chandel et al. [15], with the difference that the present study assessed the whole sugarcane lignocellulosic biomass from different varieties, while the earlier study investigated the molecular changes brought about by the acid-base pretreatment of bagasse from a single sugarcane variety. According to Zhao et al. [26], crystalline cellulose is more recalcitrant to microbial and enzymatic attack, compared to amorphous cellulose. The crystallinity index (CrI, %) [27] is often used as a measure of the relative amount of crystalline material present in a lignocellulosic sample, and can be determined using various techniques including XRD, FTIR, and NMR. For this reason, there is considerable variability in the values found in the literature, depending on the methodology used. In the present study, the method based on XRD was chosen, due to the large body of data available for this technique. The CrI values were calculated

610

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

Fig. 1. X-ray diffraction analyses of the whole sugarcane lignocellulosic biomasses from four different sugarcane varieties: untreated (A, C, E, and G) and dilute acid pretreated (B, D, F, and H).

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

according to the methodology described by Segal et al. [23], which is a fast and simple procedure for estimation of relative crystallinity [16]. Table 1 presents the CrI values for the untreated and dilute acid pretreated biomasses (bagasse, straw, and tops). In terms of relative crystallinity, the effect of dilute acid pretreatment on the sugarcane biomass was the same for all the materials, with increases in the CrI values after pretreatment. However, it was not possible to observe any trends in the magnitude of this increase, or any dependency on the residue type or sugarcane variety (Table 1). An effect of dilute acid pretreatment on relative crystallinity was expected, because the earlier compositional analyses [22] showed that there was substantial removal of hemicellulose, which is an amorphous component of the lignocellulosic biomass, with a resulting enrichment in terms of cellulose. Similarly, Chandel et al. [15] reported an increase in the CrI of sugarcane bagasse after dilute acid pretreatment, and the same has been found for sugarcane straw [19,21]. It is worth mentioning that no relationship was detected between the CrI values (Table 1) and the cellulose contents [22], in contrast to the findings of Rezende et al. [16] for a sugarcane bagasse submitted to delignification pretreatment. The effect of crystallinity (in terms of CrI values) on the enzymatic saccharification of lignocellulosic biomass has been investigated for decades [28e30], although there is no consensus about the effect of crystallinity during the enzymatic hydrolysis step, with other factors being equally or more important. For example, Moutta et al. [19] studied the enzymatic hydrolysis of sugarcane bagasse and straw mixtures pretreated with dilute acid, and reported that the CrI values did not show any direct relationship with the enzymatic digestibility of the materials. On the other hand, Li et al. [11] described a significant negative correlation between enzymatic performance and CrI values, at high enzyme loadings. The same authors highlighted that the influence of CrI on the enzymatic hydrolysis should be evaluated without the availability of enzymes being a limiting factor, and that this condition had been overlooked in earlier studies. The CrI values (Table 1) and the XRD patterns (Fig. 1) indicated that there were no marked differences in the crystallinity of the residue fractions (bagasse, straw, and tops). Furthermore, there were no noticeable differences among the sugarcane varieties. These observations help in understanding the results obtained previously concerning the susceptibility of different commercial varieties of sugarcane to enzymatic saccharification, where the variety of sugarcane was not a significant factor in the 2G ethanol production process [22]. Examination of the data reported by Pereira et al. [22] for the enzymatic conversion of cellulose (ECC, %) and the present XRD results suggested that the relative crystallinity (CrI

611

%) was not a determining factor in the effectiveness of enzymatic saccharification (under the conditions employed in these studies). Other factors must therefore have been involved, such as the accessible surface area of the cellulose, chemical composition in terms of lignin and hemicellulose contents, and structural factors (including particle size and the porosity of the material). Li et al. [11] found that at low enzyme loads (28 FPU/g cellulose), there was no correlation between CrI and ECC. In our previous study [22], an enzyme loading of 30 FPU/g cellulose was used, which could be considered limiting for investigation of the effect of crystallinity. 3.3. Fourier transform infrared spectroscopy (FTIR) FTIR was used as a tool to investigate differences among the types of sugarcane residues and among the varieties of sugarcane, as well as changes induced by the dilute acid pretreatment, by analyzing the absorption bands assigned to the main components of sugarcane lignocellulosic biomass (cellulose, hemicellulose, and lignin). Fig. 2 shows the FTIR spectra of the untreated and pretreated whole sugarcane lignocellulosic biomass (bagasse, straw, and tops) from the four varieties. It is noteworthy that the FTIR spectra of the pretreated samples were similar to those of the untreated samples, suggesting that the pretreatment did not cause any severe changes in the biomass. However, there were some differences in the spectral bands of the three residue fractions, as can be seen in Fig. 2. The FTIR spectra showed features attributed to the polymers (cellulose, hemicellulose, and lignin) that are the major components of lignocellulosic material. There were numerous bands, so the present discussion only considers those that could assist in structural elucidation and understanding of the susceptibility to enzymatic degradation, based on previously reported findings [7,8,14,15,17,31e33]. The assignments of the main bands are highlighted with their respective wavenumbers (cm1) in Fig. 2 (A and B). Since the FTIR spectral profiles were very similar for all the sugarcane varieties, only variety K is considered here. A band at 3386 cm1 was characteristic of OH groups, while a band at 2852 cm1 was attributed to OCH3 groups; both are usually present in lignin. A band at 2920 cm1 could be attributed to CH2 and CH3 groups from cellulose, lignin, and hemicellulose, without providing specific identification of any of these components. A band at 1735 cm1 corresponded to acetyl groups present in hemicellulose. Bands at 1633 and 1604 cm1 were attributed to skeletal aromatics generally found in the lignin structure (C-Ph and C]C, respectively). Bands at 1425, 1375, and 1325 cm1 were assigned to crystalline cellulose, while a band at 898 cm1 was due to amorphous cellulose. A band at 1247 cm1 was due to the stretching of CeO, characteristic of

Table 1 Crystallinity index values (CrI, %) for the whole sugarcane lignocellulosic biomasses (bagasse, straw, and tops) from four varieties of sugarcane (K, M, Q, and X), before and after dilute sulfuric acid pretreatment, determined according to the method of [23]. Variety

K

M

Q

X

Residue

Straw Tops Bagasse Straw Tops Bagasse Straw Tops Bagasse Straw Tops Bagasse

Untreated sugarcane biomass

Pretreated sugarcane biomass

CrI (%)

CrI (%)

54.9 50.8 52.7 59.2 56.2 55.8 57.4 50.2 52.6 63.4 56.7 60.4

61.4 65.9 67.3 68.1 64.6 66.3 64.8 65.5 68.1 68.2 66.7 64.6

612

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

Intensity (u.a.) 3386

TOPS

STRAW

1735

3386

2920 2852 1633 1604

Variety K

BAGASSE

(B)

Intensity (u.a.)

BAGASSE

(A)

1510

1375

4000 3600 3200 2800 2400 2000 1600 1200

1735 2852 1633 1604

Variety K

1247

800

400

-1

BAGASSE

1510

1375 1247

800

400

-1

Wavenumber (cm ) STRAW

BAGASSE

TOPS

STRAW

Intensity (u.a.)

(D)

Intensity (u.a.)

TOPS

898

1425 1325

4000 3600 3200 2800 2400 2000 1600 1200

Wavenumber (cm ) (C)

STRAW

2920

898

1425 1325

TOPS

Variety M

Variety M

4000 3600 3200 2800 2400 2000 1600 1200

800

400

4000 3600 3200 2800 2400 2000 1600 1200

-1

BAGASSE

STRAW

BAGASSE

(F)

TOPS

STRAW

Intensity (u.a.)

TOPS

400

Wavenumber (cm )

Intensity (u.a.)

(E)

800

-1

Wavenumber (cm )

Variety Q

Variety Q

4000 3600 3200 2800 2400 2000 1600 1200

800

400

4000 3600 3200 2800 2400 2000 1600 1200

-1

BAGASSE

STRAW

BAGASSE

(H)

TOPS

STRAW

Intensity (u.a.)

TOPS

400

Wavenumber (cm )

Intensity (u.a.)

(G)

800

-1

Wavenumber (cm )

Variety X

Variety X

4000 3600 3200 2800 2400 2000 1600 1200 -1

Wavenumber (cm )

800

400

4000 3600 3200 2800 2400 2000 1600 1200

800

400

-1

Wavenumber (cm )

Fig. 2. FTIR spectra of the whole sugarcane lignocellulosic biomasses from four different sugarcane varieties: untreated (A, C, E, and G) and dilute acid pretreated (B, D, F, and H).

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

hemicellulose and lignin, while a band at 1510 cm1 was due to C] C stretching of the aromatic ring in lignin. Although not highlighted in Fig. 2, intense absorption bands between 1200 and 1000 cm1 could be assigned to hemicellulose and cellulose, with maximum absorption at 1035 cm1, characteristic of CeO stretching, while a band at 1164 cm1 was characteristic of CeOeC asymmetric stretching. One of the most marked differences between the untreated and pretreated sugarcane biomasses concerned the absorption band at 1735 cm1, which decreased considerably in intensity after the pretreatment, for all residue types and regardless of the variety of sugarcane, demonstrating the effectiveness of dilute acid pretreatment in almost completely solubilizing the hemicellulosic fraction. This was further supported by the reduction of the band at 1247 cm1. A substantial difference among the types of residue, regardless of sugarcane variety, was related to the lignin band at 1604 cm1. While pretreated straw and pretreated tops showed similar low intensities, pretreated bagasse showed a marked increase in intensity. This was corroborated by the more pronounced intensification of another lignin band at 1510 cm1 for pretreated bagasse, compared to pretreated straw and pretreated tops. The bands in the range 1425e1325 cm1 (assigned to crystalline cellulose) showed distinct increases, especially in the case of the band at 1325 cm1 for pretreated bagasse. Finally, the band at 898 cm1, attributed to amorphous cellulose, displayed an appreciable intensification, as expected due to the enrichment in cellulose. These observations help in understanding the results obtained previously using the whole sugarcane lignocellulosic biomass for cellulosic ethanol production, after dilute acid pretreatment [22]. As mentioned above, the order of susceptibility to enzymatic saccharification was tops > straw > bagasse, irrespective of sugarcane variety. The FTIR spectra (Fig. 2) demonstrated that the variety of sugarcane was not a determining factor in the process, because the spectral profiles were closely similar. It should be emphasized that the set of varieties used in this study possess similar characteristics, so this finding cannot be extended to other varieties and crops. The same conclusions concerning the influence of sugarcane variety can be readily obtained from the XRD analysis (Fig. 1). Additionally, the observed FTIR bands (Fig. 2) provide an explanation for the low enzymatic susceptibility of bagasse, because the lignin bands at 1604 and 1510 cm1 were more pronounced in the spectra of pretreated bagasse than in the spectra of pretreated straw or tops. Furthermore, the crystalline cellulose band at 1325 cm1 was also more pronounced for the pretreated bagasse. It remains to be discovered why the tops were significantly more susceptible to saccharification, compared to the straw (after the dilute acid pretreatment). It is possible that there could have been interference of the band located at 2852 cm1 (attributed to lignin methoxyl groups), which was enhanced for all residue types after dilute acid pretreatment. However, the relative intensity of this band was higher for straw, regardless of the variety of sugarcane. According to Pu et al. [34], lignin is an irregular polyphenolic polymer composed of phenylpropanoid monomers with varying degrees of methoxylation. Therefore, it can be suggested that the presence of more methoxyl groups leads to greater branching in the lignin structure, consequently hindering diffusion and the productive binding of cellulolytic enzymes to cellulose. It is worth noting that the FTIR profiles obtained here for sugarcane bagasse and straw were very similar to those reported previously by Moutta et al. [18], despite the use of hydrothermal pretreatment instead of dilute acid pretreatment. Dilute acid and hydrothermal pretreatments are recognized to cause similar significant removal of hemicellulose, with increased plant cell wall pore volume, but act differently on the lignin structure, depending on the conditions used in each methodology [34,35]. The FTIR analyses showed less

613

obvious differences between the two pretreatment techniques (dilute acid and hydrothermal), due to the similar effects on the absorption bands. On the other hand, large differences among the absorption bands might be observed using other types of pretreatment (alkali, for example), especially with respect to the lignin structure. 3.4. Solid-state

13

C nuclear magnetic resonance (NMR)

Nuclear magnetic resonance is a well-established analytical tool that has been used previously to study lignocellulosic raw materials [36]. The NMR technique was employed here to investigate the chemical composition of the biomasses (untreated and dilute acid pretreated) in terms of changes in the 13C chemical shifts of cellulose (crystalline and amorphous), hemicellulose, and lignin, in order to support the findings of the XRD and FTIR analyses. Fig. 3 shows the 13C CP MAS ssNMR spectra obtained for the different biomasses (bagasse, straw, and tops from the four sugarcane varieties), before and after dilute acid pretreatment. The spectra were typical of lignocellulosic structures, showing specific signals associated with cellulose, hemicellulose, and lignin, as described in detail below. The 13C CP MAS ssNMR spectral profiles were generally similar for the untreated and pretreated biomasses, indicating that the dilute acid pretreatment used here did not dramatically alter the structures of the materials, in agreement with the results of the XRD and FTIR analyses. However, there were differences in the spectra that could provide useful information. There were no notable differences among the 13C CP MAS ssNMR spectra for the different varieties of sugarcane, so subsequent analyses only employed sugarcane variety K. The spectral signals were interpreted based on the findings of a broad set of studies concerning the solid-state NMR technique [5,9,10,15,16,36e38]. Several main signals could be extracted from the 13C CP MAS ssNMR spectra shown in Fig. 3. Signals at 62.5 and 83.5 ppm were attributed to carbons from amorphous cellulose, while signals at 64.8 and 87.9 ppm were typical of carbons from crystalline cellulose. Signals at 72.5, 74.4, and 105 ppm were also due to cellulose carbons. Signals at 21.5 and 173.6 ppm were attributed to acetyl group carbons of hemicellulose, while a signal at 56.2 ppm was characteristic of methoxyl groups (OCH3) in the lignin structure. Signals associated with lignin and hemicellulose carboxyl groups were present between 163.0 and 180.0 ppm. Signals between 110.0 and 155.0 ppm were due to aromatic carbons present exclusively in lignin. According to Rezende et al. [16], lignin carbon signals are broader due to the complex and disordered structure of this biopolymer. The spectra for the pretreated samples showed improved resolution between 50 and 100 ppm, compared to the untreated samples (Fig. 3), due to the removal of hemicellulose and consequent enrichment of the material in terms of cellulose. There were no major changes in the signals related to lignin after the pretreatment, although the signals were more intense due to removal of the hemicellulosic fraction. The most notable changes after the pretreatment were found for the signals attributed to the hemicellulose carbons, at 173.6 and 21.5 ppm, as can be clearly seen in Fig. 3. The 13C CP MAS ssNMR spectral data confirmed the effectiveness of dilute sulfuric acid pretreatment for the removal of hemicellulose, and were in agreement with previous reports [13,15,16]. The spectra obtained for the different sugarcane varieties were very similar, which explains the earlier finding that the variety of sugarcane was not a determining factor in the cellulosic ethanol production process, based on statistical analysis of enzymatic conversion and fermentation efficiency data [22].

614

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

BAGASSE Variety K

STRAW 72.5

64.8

Variety K

Intensity (u.a.)

87.9

21.5 56.2

Aromatic, lignin

160

140 13

120

80

60

40

20

0

200

TOPS

STRAW

Aromatic, lignin

180

160

140

120

100

80

60

40

20

0

C Chemical Shift (ppm)

BAGASSE

TOPS

STRAW

Intensity (u.a.) 180

160

140

120

100

80

60

40

20

0

200

180

160

TOPS

140 13

C Chemical Shift (ppm)

BAGASSE

(E)

STRAW

120

100

80

60

40

20

0

C Chemical Shift (ppm)

BAGASSE

(F)

TOPS

STRAW

Intensity (u.a.)

Variety Q

Intensity (u.a.)

Variety Q

180

160

140 13

120

100

80

60

40

20

0

200

180

160

TOPS

140 13

C Chemical Shift (ppm)

BAGASSE

(G)

STRAW

120

100

80

60

40

20

0

C Chemical Shift (ppm)

BAGASSE

(H)

TOPS

STRAW

Intensity (u.a.)

Variety X

Intensity (u.a.)

Variety X

200

56.2

Variety M

13

200

62.5

21.5

105.0

(D)

Intensity (u.a.) 200

64.8

87.9

13

Variety M

STRAW 72.5

83.5

C Chemical Shift (ppm)

BAGASSE

(C)

100

TOPS 74.4

173.6

105.0

180

BAGASSE

(B) 62.5

83.5

173.6

200

TOPS 74.4

Intensity (u.a.)

(A)

180

160

140 13

120

100

80

60

40

C Chemical Shift (ppm)

20

0

200

180

160

140 13

120

100

80

60

40

20

0

C Chemical Shift (ppm)

Fig. 3. 13C CP MAS ssNMR spectra of the whole sugarcane lignocellulosic biomasses from four different sugarcane varieties: untreated (A, C, E, and G) and dilute acid pretreated (B, D, F, and H).

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

615

Fig. 4. SEM images of the whole sugarcane lignocellulosic biomasses from four different sugarcane varieties: (A) untreated bagasse (top) and pretreated bagasse (bottom); (B) untreated tops (top) and pretreated tops (bottom); (C) untreated straw (top) and pretreated straw (bottom).

616

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617

Small differences were observed among the spectra for the bagasse, straw, and tops, especially for the signals attributed to lignin (13C chemical shifts at 110.0e155.0 and 56.2 ppm), which in all cases showed intensities that decreased in the order: bagasse > straw > tops. These findings, supported by the results of the other analyses performed in the present study, provided a possible explanation for the qualitative order of susceptibility to enzymatic saccharification (tops > straw > bagasse) that was reported previously [22]. The FTIR and NMR spectroscopic analyses indicated that there was almost complete removal of the hemicellulosic fraction, with enrichment of the pretreated material in terms of cellulose and lignin, and that there was close similarity among the varieties of sugarcane (K, M, Q, and X), as can be clearly seen in Figs. 2 and 3. The pretreated materials showed a greater presence of lignin methoxyl groups, which could directly influence the access of cellulases to the substrate in the enzymatic hydrolysis step, as discussed previously. Nevertheless, the observation of different degrees of enrichment of crystalline cellulose among the residue fractions (bagasse, straw, and tops) was only possible using FTIR spectroscopy, due to the greater sensitivity of this technique. The combination of NMR (Fig. 3) and FTIR (Fig. 2) spectroscopy can therefore be an effective way of obtaining insights into the characteristics of lignocellulosic materials. 3.5. Scanning electron microscopy (SEM) The final stage of characterization of the biomasses employed scanning electron microscopy (SEM) to evaluate the effect of dilute acid pretreatment on the morphology of the different residue fractions (bagasse, straw, and tops). Fig. 4 shows SEM images of untreated and pretreated biomass from the four sugarcane varieties. The images of the untreated biomass revealed highly compressed and rough structures, irrespective of the type of lignocellulosic material, while morphological analysis of the pretreated materials showed partial deconstruction of the natural rigid structure of the plant cell walls, with greater disruption and exposure of the fibers. Furthermore, although the structures of the untreated sugarcane biomasses (bagasse, straw, and tops) showed some similar arrangements, the pretreated materials showed no consistent structural patterns, in agreement with the heterogeneity that is characteristic of the feedstocks (Fig. 4). It could be concluded from the SEM micrographs that the tops showed the greatest structural disruption, followed by the straw and bagasse. According to Chandel et al. [15], dilute acid pretreatment acts on the cell walls of plants, causing the separation of fibers from the pith and loosening the fibrous network. Removal of the hemicellulosic fraction therefore enhances the digestibility of the lignocellulosic material. The pretreated materials displayed a less cohesive and more fractured structure after dilute sulfuric acid pretreatment, with an increase in the cellulose surface area and consequently greater access of cellulases to the substrate. Since one of the most important factors for efficient conversion of lignocellulosic biomass into fermentable sugars is the accessibility of the cellulose [39], effective pretreatment is essential for the economic feasibility of bioethanol production. Despite the difficulty in comparing the SEM results for the different samples, the use of SEM in this study provided additional visual information about the structural modification caused by the dilute acid pretreatment (Fig. 4). In addition, the SEM analyses were able to show the main differences among the sugarcane residue fractions (bagasse, straw, and tops) and the similarities among the four commercial varieties (SP79-1011, RB867515, SP81-3250, and RB92579), based on two-dimensional topographical images. It is worth noting that a careful examination was performed of the

entire set of samples, with a reasonable number of images recorded for different areas, at various magnifications and with at least 10 images per sample. Furthermore, the information obtained here using SEM micrographs to evaluate the structural characteristics of the lignocellulosic biomass was in accordance with previous reports [2,3,15]. 4. Conclusions A systematic characterization was performed of the complete residual biomass (bagasse, straw, and tops, before and after dilute sulfuric acid pretreatment) from four commercial sugarcane varieties. A combination of spectroscopy and microscopy techniques was used in order to obtain a better understanding of the physicalechemicalemorphological features of the materials. The results of the spectroscopic analyses showed that dilute sulfuric acid pretreatment did not cause any drastic changes in the structure of the sugarcane biomass and that the properties of the sugarcane varieties evaluated were closely similar. The crystallinity index (CrI) values determined from the XRD analyses revealed that the relative crystallinity of the biomass increased after dilute acid pretreatment, mainly due to removal of the amorphous hemicellulosic fraction. NMR and FTIR analyses provided valuable information concerning the biomass structure and the effects of dilute acid pretreatment, contributing to a better understanding of the susceptibility of each type of residue to enzymatic degradation. The FTIR results confirmed removal of the hemicellulosic fraction and enrichment in terms of lignin and cellulose. The greatest increases in intensity of the absorption bands corresponding to lignin and crystalline cellulose were observed for the bagasse, followed by the straw and tops. NMR analysis confirmed the solubilization of hemicellulose and the presence of typical signals attributed to the lignin structure, which were most intense for bagasse, followed by straw and tops. Observation of the sample morphology using SEM enabled insights into the way that the removal of hemicellulose influenced the enzymatic digestibility of the lignocellulosic biomass, with greater structural disruption observed for the tops, followed by the straw and bagasse. Overall, the differences observed among the sugarcane residue fractions contributed to elucidation of the characteristics that resulted in the bagasse being less easily digested by enzymatic action and that made the tops more susceptible to saccharification, compared to sugarcane straw. The combination of techniques employed here, allied with compositional analysis, offers a valuable technological tool that could assist in the implementation of standard routine tests for the characterization of different biomasses, leading to greater efficiency in bioethanol production. Acknowledgments The authors are grateful for the support provided by the Brazilian research funding agencies CAPES, CNPq, and FAPESP, and the Brazilian Agricultural Research Corporation (EMBRAPA). References [1] S. Chundawat, G. Beckham, M. Himmel, B. Dale, J. Prausnitz, Deconstruction of lignocellulosic biomass to fuels and chemicals, Annu. Rev. Chem. Biomol. Eng. 2 (2011) 121e145. [2] R. Sindhu, M. Kuttiraja, P. Binod, R.K. Sukumaran, A. Pandey, Physicochemical characterization of alkali pretreated sugarcane tops and optimization of enzymatic saccharification using response surface methodology, Renew. Energy 62 (2014) 362e368. [3] U.F. Rodriguez-Zuniga, V.B. Neto, S. Couri, S. Crestana, C.S. Farinas, Use of spectroscopic and imaging techniques to evaluate pretreated sugarcane bagasse as a substrate for cellulase production under solid-state fermentation, Appl. Biochem. Biotechnol. 172 (2014) 2348e2362.

S.C. Pereira et al. / Renewable Energy 87 (2016) 607e617 [4] J.S. Lupoi, S. Singh, B.A. Simmons, R.J. Henry, Assessment of lignocellulosic biomass using analytical spectroscopy: an evolution to high-throughput techniques, Bioenergy Res. 7 (2014) 1e23. [5] L. Fu, S.A. McCallum, J. Miao, C. Hart, G.J. Tudryn, F. Zhang, R.J. Linhardt, Rapid and accurate determination of the lignin content of lignocellulosic biomass by solid-state NMR, Fuel 141 (2015) 39e45. [6] G. Zhou, G. Taylor, A. Polle, FTIR-ATR-based prediction and modelling of lignin and energy contents reveals independent intra-specific variation of these traits in bioenergy poplars, Plant Methods 7 (2011) 9. [7] C.G. Mothe, I.C. Miranda, Characterization of sugarcane and coconut fibers by thermal analysis and FTIR, J. Therm. Anal. Calorim. 97 (2009) 661e665. [8] M. Irfan, Q. Syed, S. Abbas, M.G. Sher, S. Baig, M. Nadeem, FTIR and SEM analysis of thermo-chemical fractionated sugarcane bagasse, Turk J. Biochem. 36 (2011) 322e328. [9] H. Zhao, J.H. Kwak, Z.C. Zhang, H.M. Brown, B.W. Arey, J.E. Holladay, Studying cellulose fiber structure by SEM, XRD, NMR and acid hydrolysis, Carbohydr. Polym. 68 (2007) 235e241. [10] S. Park, D.K. Johnson, C.I. Ishizawa, P.A. Parilla, M.F. Davis, Measuring the crystallinity index of cellulose by solid state C-13 nuclear magnetic resonance, Cellulose 16 (2009) 641e647. [11] L. Li, W. Zhou, H. Wu, Y. Yu, F. Liu, D. Zhu, Relationship between crystallinity index and enzymatic hydrolysis performance of celluloses separated from aquatic and terrestrial plant materials, Bioresources 9 (2014) 3993e4005. [12] X. Duan, C. Zhang, X. Ju, Q. Li, S. Chen, J. Wang, Z. Liu, Effect of lignocellulosic composition and structure on the bioethanol production from different poplar lines, Bioresour. Technol. 140 (2013) 363e367. [13] M.A. Lima, L.D. Gomez, C.G. Steele-King, R. Simister, O.D. Bernardinelli, M.A. Carvalho, C.A. Rezende, C.A. Labate, E.R. deAzevedo, S.J. McQueen-Mason, I. Polikarpov, Evaluating the composition and processing potential of novel sources of Brazilian biomass for sustainable biorenewables production, Biotechnol. Biofuels 7 (2014) 10. [14] A.K. Chandel, F.F.A. Antunes, V. Anjos, M.J.V. Bell, L.N. Rodrigues, O.V. Singh, C.A. Rosa, F.C. Pagnocca, S.S. Silva, Ultra-structural mapping of sugarcane bagasse after oxalic acid fiber expansion (OAFEX) and ethanol production by Candida shehatae and Saccharomyces cerevisiae, Biotechnol. Biofuels 6 (2013) 4. [15] A.K. Chandel, F.A.F. Antunes, V. Anjos, M.J.V. Bell, L.N. Rodrigues, I. Polikarpov, E.R. de Azevedo, O.D. Bernardinelli, C.A. Rosa, F.C. Pagnocca, S.S. Silva, Multiscale structural and chemical analysis of sugarcane bagasse in the process of sequential acid-base pretreatment and ethanol production by Scheffersomyces shehatae and Saccharomyces cerevisiae, Biotechnol. Biofuels 7 (2014) 63. [16] C.A. Rezende, M.A. de Lima, P. Maziero, E.R. de Azevedo, W. Garcia, I. Polikarpov, Chemical and morphological characterization of sugarcane bagasse submitted to a delignification process for enhanced enzymatic digestibility, Biotechnol. Biofuels 4 (2011) 54. [17] R.C.N.R. Corrales, F.M. Teixeira Mendes, C.C. Perrone, C. Sant'Anna, W. de Souza, Y. Abud, E.P.S. da Bon, V. Ferreira-Leitao, Structural evaluation of sugar cane bagasse steam pretreated in the presence of CO2 and SO2, Biotechnol. Biofuels 5 (2012) 36. [18] R. Moutta, M. Silva, R. Corrales, M. Cerulho, V. Ferreira-Leitao, E.P.S. Bon, Comparative response and structural characterization of sugarcane bagasse, straw and bagasse-straw 1:1 mixtures subjected to hydrothermal pretreatment and enzymatic conversion, J. Microb. Biochem. Tech. (2013) 1e8. [19] R. Moutta, V. Ferreira-Leitao, E.P.S. Bon, Enzymatic hydrolysis of sugarcane bagasse and straw mixtures pretreated with diluted acid, Biocatal. Biotransform 32 (2014) 93e100.

617

[20] L. Oliveira, V. Nascimento, A. Goncalves, G. Rocha, Combined process system for the production of bioethanol from sugarcane straw, Ind. Crop Prod. 58 (2014) 1e7. [21] R. Sindhu, M. Kuttiraja, P. Binod, K.U. Janu, R.K. Sukumaran, A. Pandey, Dilute acid pretreatment and enzymatic saccharification of sugarcane tops for bioethanol production, Bioresour. Technol. 102 (2011) 10915e10921. [22] S.C. Pereira, L. Maehara, C.M.M. Machado, C.S. Farinas, 2G ethanol from the whole sugarcane lignocellulosic biomass, Biotechnol. Biofuels 8 (2015) 44. [23] L. Segal, J.J. Creely, A.E. Martin, C.M. Conrad, An empirical method for estimating the degree of crystallinity of native cellulose using the X-ray diffractometer, Tex Res. J. 29 (1959) 786e794. [24] L. Canilha, V.T.O. Santos, G.J.M. Rocha, J. Silva, M. Giulietti, S.S. Silva, M.G.A. Felipe, A. Ferraz, A.M.F. Milagres, W. Carvalho, A study on the pretreatment of a sugarcane bagasse sample with dilute sulfuric acid, J. Ind. Microb. Biotechnol. 38 (2011) 1467e1475. [25] G.J.M. Rocha, V.M. Nascimento, A.R. Goncalves, V.F. Nunes Silva, C. Martin, Influence of mixed sugarcane bagasse samples evaluated by elemental and physical-chemical composition, Ind. Crop Prod. 64 (2015) 52e58. [26] X. Zhao, L. Zhang, D. Liu, Biomass recalcitrance. Part I: the chemical compositions and physical structures affecting the enzymatic hydrolysis of lignocellulose, Biofuels Bioprod. Biorefin 6 (2012) 465e482. [27] S. Park, J.O. Baker, M.E. Himmel, P.A. Parilla, D.K. Johnson, Cellulose crystallinity index: measurement techniques and their impact on interpreting cellulase performance, Biotechnol. Biofuels 3 (2010) 10. [28] V.S. Chang, M.T. Holtzapple, Fundamental factors affecting biomass enzymatic reactivity, Appl. Biochem. Biotechnol. 84 (2000) 5e37. [29] V.P. Puri, Effect of crystallinity and degree of polymerization of cellulose on enzymatic saccharification, Biotechnol. Bioeng. 26 (1984) 1219e1222. [30] M. Hall, P. Bansal, J.H. Lee, M.J. Realff, A.S. Bommarius, Cellulose crystallinity e a key predictor of the enzymatic hydrolysis rate, FEBS J. 277 (2010) 1571e1582. [31] S.Y. Oh, D.I. Yoo, Y. Shin, G. Seo, FTIR analysis of cellulose treated with sodium hydroxide and carbon dioxide, Carbohydr. Res. 340 (2005) 417e428. [32] S. Cao, G.M. Aita, Enzymatic hydrolysis and ethanol yields of combined surfactant and dilute ammonia treated sugarcane bagasse, Bioresour. Technol. 131 (2013) 357e364. [33] R.M. Silverstein, F.X. Webster, D. Kiemle, D.L. Bryce, Spectrometric Identification of Organic Compounds, eighth ed., 2015. [34] Y. Pu, F. Hu, F. Huang, B.H. Davison, A.J. Ragauskas, Assessing the molecular structure basis for biomass recalcitrance during dilute acid and hydrothermal pretreatments, Biotechnol. Biofuels 6 (2013) 15. [35] X. Zhao, L. Zhang, D. Liu, Biomass recalcitrance. Part II: fundamentals of different pre-treatments to increase the enzymatic digestibility of lignocellulose, Biofuels Bioprod. Biorefin. 6 (2012) 561e579. [36] M. Foston, Advances in solid-state NMR of cellulose, Curr. Opin. Biotechnol. 27 (2014) 176e184. [37] S.P.S. Chundawat, B.S. Donohoe, LdC. Sousa, T. Elder, U.P. Agarwal, F. Lu, J. Ralph, M.E. Himmel, V. Balan, B.E. Dale, Multi-scale visualization and characterization of lignocellulosic plant cell wall deconstruction during thermochemical pretreatment, Energy Environ. Sci. 4 (2011) 973e984. [38] K. Wickholm, P.T. Larsson, T. Iversen, Assignment of non-crystalline forms in cellulose I by CP/MAS C-13 NMR spectroscopy, Carbohydr. Res. 312 (1998) 123e129. [39] X. Meng, A.J. Ragauskas, Recent advances in understanding the role of cellulose accessibility in enzymatic hydrolysis of lignocellulosic substrates, Curr. Opin. Biotechnol. 27 (2014) 150e158.