Thermodynamic analysis,synthesis,characterization,and evaluation of 1-ethyl-3-methylimidazolium chloride: Study of its effect on pretreated rice husk

Eileen Katherine Coronado-Aldana,Cindy Lizeth Ferreira-Salazar,Nubia Yineth Pi?eros-Castro,Rubén Vázquez-Medina,Felipe A.Perdomo

1 Universidad de Bogotá Jorge Tadeo Lozano,Facultad de Ciencias Naturales e Ingeniería,Cra.4 22-61,Cundinamarca,Bogotá,Colombia

2 Instituto Politecnico Nacional,CICATA Queretaro,Cerro Blanco 141,Colinas del Cimatario,76090 Queretaro,Mexico

3 School of Engineering,University of Edinburgh,Edinburgh EH9 3FB,United Kingdom

Keywords:Ionic liquids Fluid-phase equilibria Thermodynamics Waste treatment UNIFAC approach

ABSTRACT This work is focused on the determination of the optimal reaction conditions to synthesize the ionic liquid 1-ethyl-3-methylimidazolium chloride([EMIM][Cl])and assess its suitability for the pretreatment of rice husk.The modified UNIFAC (UNIversal quasi-chemical Functional-group Activity Coefficients)approach for ionic liquids is used to develop a thermodynamic model that describes the reactive system methylimidazole (MIM),chloroethane (C2H5Cl) and [EMIM][Cl].The model allows to study the phase equilibria coexistence (vapor-liquid equilibria and solid-liquid equilibria) and yields the theoretically optimal conditions to synthesize the ionic liquid.The model predictions are validated with the available experimental and reported data.By implementing the developed model,a simple way to synthesize ionic liquid [EMIM][Cl] was found allowing to study its influence on the structure and morphology of pretreated rice husk.The lignocellulosic materials involved in this study are characterized by their composition,enzymatic digestibility,scanning electron microscopy,and crystallinity.Compared to untreated material,[EMIM][Cl]-pretreated rice husk produces cellulose that can be efficiently enzymatic hydrolyzed with high sugar yields.This work offers a suitable methodology to include the synthesis and thermodynamics of the solvent media within the design of low-cost ionic liquids for lignocellulosic biomass pretreatment.

1.Introduction

Biomass available from agricultural residues of corn,wheat,sorghum,rice,and other crops can be used for chemical,food,or industrial applications.In order to use this biomass as a reliable energy source,it is mandatory to develop biotechnological or chemical methods that decomposes cellulose chains into easily hydrolyzable material[1].Chemical methods,particularly the ionic liquids-based methods,have been widely studied [2,3].Ionic liquids can be defined as molten salts comprising one organic cation associated with one inorganic or organic anion.Commonly,the molecular structure of ionic liquids is featured by a long chain anion and a cation with low molecular symmetry conferring them unique thermodynamic properties.Their main feature is the wide range of possibilities to combine anion-cation pair as they can be designed to suit any specific task for a specified process [4].

Ionic liquids are distinguished from traditional solvents by their very low vapor pressure,which avoids atmospheric emissions of evaporated solvent.They also have relatively high thermal stability,low flammability,and recyclability,which contributes to waste reduction.Depending on the desired application,they can be designed by appropriately selecting the cation-anion pair.The chemical structure of cations and anions confers ionic liquids distinctive physical-chemical properties and even brief changes in the molecular framework undergo considerable differences [3,5].For instance,ionic liquids composed of 1,3-dialkyl imidazolium cations are featured by low melting point,low vapor pressures at ambient condition,and a good capability to dissolve a wide range of inorganic and organic materials.Ionic liquids of this type are efficient for dissolution of lignocellulosic materials improving the enzymatic hydrolysis [6-10].

Ionic liquids-based imidazole can be used to disrupt the tight structure of the lignocellulosic biomass.However,they exhibit high viscosity,high hygroscopicity,and high cost of laboratorysynthesis limiting its scaling-up on industrial processes [11].The physicochemical properties of an ionic liquid are influenced by the anion type it contains.For instance,1-butyl-3-methylimidazolium chloride ([BMIM][Cl]) and 1-ethyl-3-methylimidazolium chlo ride ([EMIM][Cl]) are solids at room temperature,while 1-butyl-3-methylimidazolium Acetate([BMIM][OAc])and 1-ethyl-3-methylimidazolium acetate([EMIM][OAc]) are liquids.In terms of viscosity,ionic liquidsbased imidazole such as 1-butyl-3-methylimidazolium hydrogen sulfate ([BMIM][HSO4]) and 1-ethyl-3-methylimidazolium hydrogen sulfate ([EMIM] [HSO4]) exhibit values of 3100 cP and 1500 cP,at 293.15 and 303.15 K,respectively [11].The difference in these values has been attributed to the Van der Waals forces between the alkyl substituents in the imidazole ring,which are stronger for longer chains.A good guess about the hygroscopicity in ionic liquids can be obtained from their water content.Several studies have found that imidazole based in ionic liquids exhibit lower water content and reduce the tendency to absorb moisture when are combined with [Cl-] anion [12].

Moreover,dialkyl-imidazolium salts can be synthesized from commercially available materials and on that basis can be promising alternatives in the search of affordable ionic liquids for biomass dissolution.In such wise,rice husk’s cellulose is one of the main agricultural wastes generated as a by-product during the rice milling process [13].Despite of slight differences reported in the composition profile[14],rice husk is generally composed of cellulose (35%),hemicellulose (25%),lignin (20%),crude protein (3%),and ash (17%) [15,16],making this a valuable residual for extraction of biomaterials.

Thus,in the current work,we want to assess the suitability of 1-ethyl-3-methylimidazolium chloride,[EMIM][Cl],for the dissolution of rice husk’s cellulose.To the best of our knowledge[EMIM][Cl] has not yet been implemented in rice husk pretreatment;as can be seen from Table 1,only [EMIM][OAc],1-ethyl-3-methylimidazolium diethyl phosphate ([EMIM][DEP]),and[BMIM][Cl] have been reported so far [17,18].Although some of the mentioned ionic liquids are easier to handle than [EMIM][Cl]as they exhibit lower melting points;they have the disadvantage to be more hygroscopic easily increasing their moisture content and therefore,undermining the performance to dissolve cellulose[12].Ultimately,the main goal in the implementation of ionic liquids as solvents of lignocellulosic biomass is driven by the search of an affordable and low-cost pretreatment.Hence,unlike others imidazole based ionic liquids with more complex anions (i.e.,[OAc] and [DEP]) the synthesis of [EMIM][Cl] is simple,cheaper and can be performed from commercially available materials(e.g.,chloroethane).

Table 1 Capability of ionic liquids to dissolve rice husk cellulose

Nevertheless,there is still a gap in the synthesis of [EMIM][Cl],as there is a lack of consensus about the optimal conditions to carry out the reaction and,moreover,the limited literature available is inconclusive.Because of the consistency of their molecular structure,imidazolium-based ionic liquids have been widely represented implementing theoretical tools and thermo-dynamic models [19-22].Particularly,the molecular structure of [EMIM][Cl] comprises a relatively simple and asymmetric cation,and a monatomic anion,with abundant information in the open literature of physical constants and parameters allowing the description of its physicochemical properties.It means that is possible to develop a thermodynamic analysis that drives the study of the synthesis of [EMIM][Cl] and helps to better understand the phase coexistence of the reactive system.

Consequently,the current work is devoted to developing a thermodynamic study of the reactive system in order to determine the optimal set of conditions at which [EMIM][Cl] can be synthesized.For the corresponding synthesis,we design an experimental setup to test the ionic liquid stability and control the operational conditions in the reactor.Then,a physical-chemical characterization of the products is carried out to determine the purity and reaction yield.Finally,we assess the performance of the synthesized[EMIM][Cl] in the pretreatment of rice husk in terms of lignincarbohydrates linkages disruption,changes in the biomass morphology and structure,and the effect on enzymatic hydrolysis.The study is followed by a physicochemical characterization of the recovered products as well as a comparison between the pretreated rice husk using [EMIM][Cl] and the untreated samples.The analysis is completed by measuring the crystallinity of the materials implementing X-ray diffraction (XRD) and morphology using scanning electron microscopy (SEM).

As we mentioned above,to the best of our knowledge this is the first time that [EMIM][Cl] is implemented in the pretreatment of rice husk and the results obtained in our current work will be used to broad the spectrum of ionic liquids in this regard presented in Table 1.In this work,[EMIM][Cl]was used due to the consideration of reducing sugar yields shown in Table 1,which will be contrasted with results obtained at Section 4.2.

This paper is organized as follows.In Section 2,we describe the thermodynamic model used to calculate the phase behavior of the mixtures comprising the reactive system.In this description,we consider three types of phase equilibria: vapor-liquid equilibria(VLE) MIM+Chloroethane,chemical reaction equilibria (Menshutkin’s reaction equilibria) and MIM+[EMIM][Cl] solid-liquid equilibria (SLE).In Section 3,we describe the experimental setup to carry out the synthesis.Also,we describe the techniques used to characterize the ionic liquid and the procedure to carry out the hydrolysis of the rice husk samples treated with the synthesized [EMIM][Cl].In Section 4,we present our findings and make a comparison between untreated and treated rice husk.Concluding remarks are given in Section 5.

2.Model for Thermodynamic Analysis of the Reactive System

Reactants (methylimidazole and chloroethane) and products([EMIM][Cl]) are featured by a complex thermodynamic behavior in a wide range of pressure and temperature.The classic synthesis of [EMIM][Cl] is led by the Menshutkin’s reaction,in which the quaternization of dialkyl-imidazole by chloroethane takes place followed by the ionic liquid formation.The Menshutkin’s reaction must be carried out in liquid phase,but the compounds involved in the synthesis of [EMIM][Cl] exhibit a remarkable difference of volatility.According to Leeetal.[6],it can be inferred that this reaction must be carried out at high pressures,but there is not a clear consensus about the optimal conditions to archive the best performance and yield.This information is very relevant in order to evaluate the possibility of using this type of ionic liquid at large scale and exploring potential industrial applications.Probably,one of the major barriers to develop a thermodynamic analysis in this type of systems is the lack of reliable experimental data.In order to overcome this disadvantage,we propose a consistent methodology to predict not only the influence of the variables over the reaction performance but also the fluid-phase behavior of the system.

2.1.MIM+Chloroethane fluid-phase equilibria

Our study starts with the description of the vapor-liquid equilibria (VLE) for the mixture 1-methylimidazole ([MIM])+chloroe thane(C2H5Cl).The analysis of VLE slices allows to identify the best conditions in the reactor to guarantee that the reactive mixture remains below the bubble point at any molar ratio of the reactants.According to the classic thermodynamic of mixtures,the compositions of liquid and vapor coexisting in equilibrium are determined by applying the general criteria as follows [23]:

where γiand ?iare the activity and fugacity coefficients,respectively,for theithcompound in the mixture,is the pure compound saturation pressure;andxiandyiare the composition of the liquid and vapor phase,respectively.

Equation (1) comes from the chemical potential equality criteria,but here it is expressed in terms of the fugacity for each one of the involved species.In addition,the mass balance must be satisfied for each phase coexisting in equilibria and consequently,for the VLE case this constrain can be expressed byxi=1 andyi=1.

The activity coefficients are determined using the modified UNIFAC(Do)group-contribution(GC)approach[24],taking the values for the main groups and subgroups between cations and anions following the procedure presented in [24-26] to describe the fluidphase behavior of systems containing ionic liquids.The calculation of the fugacity coefficients is made using the classical Peng-Robinson cubic equation of state [27],where the acentric factors for MIM it is estimated using the Ambrose-Walton corresponding states method[28],for chloroethane it is taken from National Institute of Standards and Technology (NIST) standard reference database [29],and for [EMIM][Cl] it is estimated implementing the group-contribution(GC)methodology developed in[30].

A stability analysis is also implemented by applying the condition given in Eq.(2)to study the miscibility of the mixture at given conditions of pressure,P,temperature,T,and compositionx.

whereGErepresents the excess free energy of the system,and it is estimated implementing the universal quasi-chemical functionalgroup activity coefficient (UNIFAC) model.

2.2.Menshutkin’s reaction equilibria

When the stoichiometry of a chemical reaction is known a priori the chemical reaction equilibria can be calculated by using equilibrium constants as follows:

whereaidenotes the activity of eachispecie in the reaction and it can be expressed as the relation between the fugacity,^fiof theithspecie in the mixture evaluated at given temperatureT,pressureP,and compositionxdivided by the fugacity of the same component,at the selected standard state as is described in Eq.(4)selecting as state the fugacity of the compoundias

Once the criteria given in Eq.(3)is satisfied,the composition of all the species in equilibria is obtained.This information allows to determine the suitable ranges in pressure and temperature to carry out the chemical reaction.It means that once the equilibrium composition,xe,is estimated,the free energy can be evaluated and therefore the optimal conditions for the reaction can be predicted theoretically by finding the minimum for the free energy,ΔG,as follows:

where μiand virepresent the chemical potential and the stoichiometric coefficient of each specieipresent in the reaction.

The chemical potential is expressed as a function ofT,Pandxeas is given in Eq.(6).

where μ0iis the reference chemical potential ofievaluated at same conditions ofPandT.

The analysis can be made by substituting Eq.(6)into Eq.(5).The procedure also requires the knowledge of the standard formation properties for each species that take place in the reaction.In the case of[EMIM][Cl],the formation free energy was estimated using the Joback’s group contribution methodology [28] and the formation enthalpy was taken from [31],while for the methylimidazole and chloroethane the same properties were taken from [29,32].

2.3.MIM+[EMIM][Cl] solid-liquid equilibria

Among the phase changes observed during the synthesis of[EMIM][Cl],the one regards to liquid-solid deserves to be highlighted (see Fig.4).Once the reaction was completed,the mixture leaving the reactor was cooled to environment conditions.It undergoes a solidification process that suggests the presence of a eutectic point between the ionic liquid and the amount MIM remaining in the system.By calculating the intrinsic solubility of MIM in [EMIM][Cl] is possible to have an educated guess about the remaining amount of non-converted reactants (i.e.MIM).

The description of solid-liquid equilibria is typically undertaken by means of a standard thermodynamic cycle,which leads to the following compact working expression(see Eq.(7)),where the temperature dependence of the heat capacity has been neglected[33].

We use Eq.(7) to predict the solid-liquid coexistence of MIM+[EMIM][Cl] mixture,where χjrepresents the maximum amount of solutejthat can be dissolved in the liquid phase,ΔHfis the enthalpy of melting ofjat its melting temperatureTf,j.In a similar fashion,the activity coefficient of the solute γjT,P,χjat the specifiedT,Pand χjwas determined by means the modified UNIFAC (Do) group contribution (GC) method [25].

3.Methods and Experiments

3.1.Ionic liquid synthesis

We purchased the reagents chloroethane (99%) from Aldrich Chemical Co.(China)and[MIM](99%)from MERCK(China),which were used without pretreatments.Having determined the optimal conditions of pressure,temperature,and reactants initial ratio,the synthesis of [EMIM][Cl] was carried out by means of an alkylation with chloroethane.In order to guarantee liquid phase for both reactants we used a high-pressure reactor.The operation of all the streams coming into the reactor was performed in batch mode.The reactor is isolated by a hermetic jacket suitable for working at high pressure and temperature.The device is also equipped with a magnetic stirrer,pressure gauge,a thermo-couple,and a temperature transducer.

3.2.Ionic liquid characterization

3.2.1.Fouriertransforminfrared(FTIR)

We analyzed all the samples at room temperature using an Agilent Cary 630 FTIR Spectrometer obtaining,with a very good resolution,the absorbance spectrum in (500,4000) cm-1for ionic liquids.FTIR spectra of products ([EMIM][Cl]) and reactants allowed easily infer whether the chemical reaction reached the expected yield.

Although Nuclear Magnetic Resonance(NMR)would be a more adequate technique for the characterization of the synthesized IL,unfortunately we were not able to access this technique.However,our analysis of FTIR matches the results presented by Suarezetal.[34],where both FTIR and NMR were implemented for the characterization of 1-n-butyl-3-methylimidazolium.We found a very good agreement between our results and their reported FTIR spectra.

3.2.2.Conductivityanddensity

We measured the conductivity of [EMIM][Cl] using a Mettler Toledo Seven2Go S7 and its density using a Mettler Toledo pycnometer Density2Go.The cell constant was calibrated by using an aqueous solution of KCl (0.01 mol.L-1at 298.15 K).Since[EMIM][Cl]is solid at room temperature,we performed a common measurement practice based on the chemical reaction of [EMIM][Cl] with AlCl3.Therefore,we prepared solutions of ionic liquid([EMIM][Cl]) considering 0.34 g of AlCl3in 1 ml of ionic liquid to obtain reliable density and conductivity measurements at room conditions.

3.3.Rice husk pretreatment

We blended the rice husk samples with a particle size of 250 μm in an Erlenmeyer flask with 10 g of [EMIM][Cl] per 1 g of rice husk.The ionic liquid was previously heated above the melting point.The mixture was heated to 373.5 K and stirred during the following 12 h.Then,the solution was cooled until crystallization followed by the respective regeneration process.In this way,200 ml of aqueous solution of acetone (0.1 ml of acetone in 10 ml of water) were added at constant stirring during 30 min to remove the ionic liquid from biomass.Then,the mixture was centrifuged (6000 r.min-1) during 30 min and the precipitated was washed in order to pull any trace of ionic liquid out of the formed cake and drying to remove the water-acetone.The regenerated fraction was obtained by means of a soft extraction with NaOH.For a detailed description of the procedure,we direct the reader to Ref.[35].Finally,during the biomass fractionation,the ionic liquid was extracted from the solution of acetone by adding 150 ml acidic water (pH 2.0) at constant stirring during 15 min at 500 r.min-1and the mixture was centrifuged at 6000 r.min-1during 15 min and the resulting supernatant was composed of [EMIM][Cl] and acidified water [36].

On the other hand,we have carried out batch enzymatic hydrolysis of pretreated and untreated rice husk samples at 323.15 K and 150 r.min-1in an orbital shaker during 43 h,where 1.25 g of original samples were suspended in 25 ml of sodium citrate buffer 0.1 mol.L-1with pH of 4.8,followed by the addition of 0.3 μl of cellulases(celluclast 1.5 L-15 FPU per 1 g substrate)and 0.1 μl of beta glucosidases (NS50010-12.6 UI per 1 g substrate) of Novozymes.We calculated reducing sugars following the dinitrosalicylic acid method as reported in Ref.[37].

3.4.Structure and morphology measurements

The study of the morphological and structural changes before and after the pretreatment of biomass in ionic liquids is a suitable way to evaluate their impact and role on the pretreatment process[6,7].In order to assess the effect of the [EMIM][Cl] on the molecular structure of the rice husk,we have carried out X-ray diffraction patterns and scanning electron microscope images in untreated and treated materials.

3.4.1.X-raydiffractioncharacterization

The X-ray diffraction patterns for each sample of pretreated and untreated rice husk were acquired with a Rigaku MiniFlex diffractometer operating at 35 kV and 15 mA at room temperature.The samples were scanned from 10°-80° (2θ) with a step size of 0.05°.The apparent crystallinity was calculated with the relation defined by Segal’s method [9,38]:

whereITdenotes maximum intensity of crystalline portion in rice husk samples (2θ=22°-26°),andIAis the intensity due to the amorphous portion at the minimum intensity (2θ=12°-18°).

3.4.2.Scanningelectronmicroscopy(SEM)

In order to determine the morphological change,the variation in the fibrillar structure and the degree of disruption in the rice husk samples after the pretreatment in [EMIM][Cl],we used a JSM-6490 Series Scanning Electron Microscope to obtain the images of treated and untreated samples at different magnifications: ×300,×500,×1000,and ×2500.We acquired the representative images with a 10 kV accelerating voltage.

4.Results and Discussion

We started our investigation focusing on the fluid-phase equilibria of the binary mixture chloroethane+[MIM].As we mentioned in Section 2,to successfully obtain [EMIM][Cl] by following the Menshutkin’s mechanism is mandatory to guarantee liquid phase for both reactants in the reaction vessel.At environment temperature and pressure [MIM] is liquid and chloroethane is gas.Therefore,the conditions of temperature and pressure in the reactor (before starting the reaction) must remain below the bubble point at the initial compositions of the reactants.Be-cause of the lack of experimental data for this system,we developed a thermodynamic model (see Eq.(1)) allowing us to understand the fluid-phase behavior of this mixture and set the suitable conditions to carry out the reaction.

The prediction of the isobaric vapor-liquid equilibria for the mixture chloroethane+[MIM]at several different pressures is presented in Fig.1.The blue continuous line describes the system atP=0.5 MPa,the dot-dashed line atP=0.6 MPa,the red continuous line atP=0.7 MPa,and the dotted line atP=0.8 MPa.The only reliable experimental data available correspond only to vaporpressure data for the pure compounds [39],and as it is evident from Fig.1,the model is able to match these points.

Fig.1.Isobaric vapor-liquid equilibrium of binary mixture (C2H5Cl)+[MIM].Triangles and squares represent experimental data [39] and the curves are phase equilibria calculations represented in a x-T diagram: (a) P=0.6 MPa (dot-dashed line,red squares) and P=0.5 MPa (blue continuous line,red triangles) and (b) P=0.7 MPa (red continuous line,blue triangles) and P=0.8 MPa (dotted line,blue squares).

The main advantage of the proposed model is that it allows one to explore different conditions of the fluid-phase equilibria.In this way,we could identify a suitable region(P,T,andx)to set the right initial conditions prior to start the corresponding chemical reaction.This feasible region is also depicted in Fig.1 as the area enclosed by dotted lines representing the limit temperature at constant pressure and fixed composition,where the system coexists at vapor-liquid equilibria.In this analysis we have fixed the composition limit tox=0.6 liquid molar fraction of chloroethane as it was the composition in excess reported in Ref.[40].We select as optimal pressureP=0.7 MPa and a suitable interval of temperature varying from 348.15 K to 351.15 K to perform the synthesis of[EMIM][Cl].

It is also important to evaluate the miscibility of the reactants,this is because of the fact that only a single phase must exist along the reaction at the given conditions ofPandT.The presence of another liquid phase gives rise to the formation of a liquid-liquid interface affecting the performance of the reaction as the interface minimizes the influence of the kinetics on the reaction.By calculating the change of the free energy of mixing,we can infer the effect of temperature and composition over the miscibility of the mixture.Along these lines,the prediction of the free energy change of mixing for a selected temperature range is presented in Fig.2(a).The curves were obtained by solving Eq.(2),where the activity of the species was calculated using the UNIFAC(Do)approach and the respective parameter values are taken from [24,26,41].It is worth noting that atT=349.15 K the binary systems exhibited total miscibility in the whole range of composition;it means the mixture is homogeneous at the selected working conditions.Here,we also study the chemical reaction equilibria by solving Eqs.(3)to(5) for the selected interval of temperature,where the ternary reactive system (C2H5Cl+[MIM]+[EMIM][Cl]) is supposed to be homogeneous and stable.This analysis allowed not only to evaluate the reaction favorability,but also to identify the optimal conditions to reach the equilibria.From Fig.2(b),it can be noticed that the free energy change in the reaction ΔGr×nis negative,which means that the reactants have more free energy than the products,and the reaction is spontaneous.

Fig.2.Change of reaction free energy.(a)Binary mixture[MIM].Calculation with the UNIFAC(Do)group-contribution approach[24,26,41]at T=333.15 K(black dot-dashed line),T=349.15 K (blue continuous line) and T=363.15 K (red dashed line),(b) reactive system C2H5Cl+MIM+[EMIM][Cl].

The prediction of the isobaric solid-liquid equilibria for[EMIM][Cl]+MIM is shown in Fig.3.The values for the heat of fusion were taken from [42] and the melting points for MIM and [EMIM][Cl]were taken from[39].Fig.3 represents the solubility of solid solute(MIM or [EMIM][Cl]) in a liquid phase comprising both compounds.The left side of the diagram represents the equilibria between the binary liquid mixture and solid EMIM,while the right side represents the equilibria between the liquid mixture and solid[EMIM][Cl].The model also predicts an eutectic point atT=263.6 K,where the three phases coexist and therefore the solid phase does not form a mixed crystal structure or in other words,the system exhibits total immiscibility in the solid phase.It should be pointed out that the only available experimental data reported for the systems presented in Figs.1 to 3 correspond to properties of pure compounds.It means that the results presented in this Section are essentially predictions and,therefore,some accuracy should be expected.The reliability of our results is based on accurate predictions for equivalent systems (i.e,[EMIM][Cl]+ethanol,[MIM]+ethanol,and ethanol+chloroethane)and the transferability of group-group interaction parameters that features the UNIFAC approach.

4.1.[EMIM][Cl] synthesis

With the optimal conditions obtained from the previous thermodynamic analysis,we proceed to carry out the reaction using a high-pressure reactor.In this way,we introduced 0.123 mol of MIM into the reactor and we suddenly decreased the temperature close to 273.15 K,then 0.1942 mol of gas chloroethane flowed in and condensed into the reactor.We adjusted the temperature to 349.15 K and a pressurization process occurred,allowing the inlet of argon into the system untilP=0.7 MPa was achieved.At this point the reaction began,and by keeping constant temperature and stirring,the reaction reached equilibrium after 48 h.The reactor was decompressed to release the excess of argon and chloroethane.The molten product (ionic liquid) was transferred to a dry box,where the product crystallizes at ambient conditions.

In order to characterize the products and possible traces of unconverted reactant,we started our analysis implementing FTIR to understand whether the ionic liquid was present in the products by identifying its characteristic peaks of the chemical groups.The phase transition of the products after the synthesis from reactor to ambient conditions is presented in Fig.4,and the corresponding IR spectrum of the samples is depicted in Fig.5.As is apparent from Fig.5,the IR spectrum exhibits peaks in the range (2800,2900)cm-1,which can be attributed to the ethyl group bonded to imidazole ring [34].The spectrum region accounting for the cation (imidazole) can be observed in (2913,2965) cm-1and peaks above 3000 cm-1correspond to the aromatic C-H stretching,whereas those below 3000 cm-1can be attributed to aliphatic stretching.

Fig.4.Phase transition of the product leaving the reactor from liquid to solid after the synthesis of [EMIM][Cl].

Fig.5.IR spectrum of synthesized[EMIM][Cl].Highlighted areas are spectrum regions with the stretching vibrations that features the ionic liquids of comprising imidazole rings.

As complementary information,the more relevant stretching vibrations and frequencies taken from the infrared spectral data(500,4000)cm-1for MIM and[EMIM][Cl]are presented in Table 2.

Table 2 Relevant stretching vibrations and frequencies from the infrared spectral data (500,4000) cm-1 of MIM and [EMIM][Cl]

Additionally,as can be seen from Fig.6,the formations of new bonds become apparent by comparing the corresponding spectra for [MIM] (depicted in red) and [EMIM][Cl] (depicted in blue).It is worth noting that absorption in 1170 cm-1corresponds to the strength of the bond C-H in the imidazole ring that features the interaction with the anion chloride [Cl]-.Regardless of some changes in the intensity for the C=C,C-N vibrations,the main difference between both spectra is the presence of the CH3CH2CH that features the presence of the ethyl group in the structure.These results allow us to conclude that the structure of the synthesized compound clearly is featured by the presence of a Cl-1-imidazole ring (IM) bonds as well as the interaction between the groups CH (aliphatic)-IM,which means the type of interactions featuring [EMIM][Cl].

Fig.6.FTIR spectra in (500,4000) cm-1 for [EMIM][Cl] (blue) and [MIM] (red).

When the alkyl substituent in the ionic liquid corresponds to short chain (i.e.,ethyl),the resulting compound tends to adopt a crystalline structure as it was presented for [EMIM][Cl] and depicted in Fig.4.In order to determine the density and conductivity,we followed the methodology detailed in Section 3.2 to measure the properties of melted ionic liquids at room temperature conditions.In Table 3,we present the measurements of density and conductivity for solutions of ionic liquid at 0.34 g of AlCl3in 1 ml of ionic liquid (in a saturated solution of AlCl3(50% (mol)).The main reason for using a saturated solution of AlCl3to dissolve[EMIM][Cl]and measure the conductivity is essentially its capability to decrease the melting point of the ionic liquid,keeping it as liquid at working conditions,without affecting the dielectric constant of the media.The dielectric constant of AlCl3is essentially low,with reported values of 2.68-3.42.This procedure has been implemented by Fanninet.al,[43] to measure density,electrical conductivity,and viscosity of 1,3-dialkylimidazolium chloridealuminum chloride ionic liquids.In addition,our measurements agreed the densities reported for[EMIM][Cl]in solution with AlCl3[44].The high values measured for the density envisage the complex molecular structure featuring the ionic liquid [EMIM][Cl].In addition,the conductivity values are a consequence of the role played by H-Cl bonds on the molecular structure.In this way,the measured conductivity values are greater than other values reported for imidazole-based ionic liquids with alkyl substituent of longer chain,such as 1-n-butyl-3-methylimidazolium tetrachloroaluminate evincing a value in the conductivity of 9.12 mS.cm-1at 398.15 K [45].It is important to point out that the use of Lewis acids to perform conductivity measurements undergoes a screening effect between cation and anion due to the effective electrostatic potential of the solvent.

Table 3 Experimental values of density and conductivity for [EMIM][Cl] at T=295.15 K

4.2.Rice husk pretreatment with [EMIM][Cl]

The composition of cellulose,hemicellulose,and lignin in samples of rice husk pretreated with ionic liquid and untreated are given in Table 4.In all cases,our experimental results agree the data reported in[17,46-48],which are in accordance with the corresponding composition of lignocellulosic biomass reported in[49].The experimental conditions,namely heating at 373.15 K for 10 h,were determined following the procedure reported by Angetal.[17,18].

Table 4 Composition of untreated and ionic liquids-pretreated rice husk

As far as residual materials for rice husk are concerned,differences in biomass composition are expected to affect the process.For example,the composition of rice husk reported in [17] comprises 53.18% (mass) cellulose,4.63% (mass) hemicellulose and 19.67%(mass)lignin.Here,the amount of dissolved cellulose varies depending on the solvent type (IL) implemented in the pretreatment,for instance: 0.37,0.31,and 0.16 g regenerated cellulose per 1 g rice husk under the same pretreatment conditions by using[EMIM][OAc],[BMIM][Cl],and [EMIM][DEP],respectively.Therefore,from Table 4,in our case,the samples of rice husk are featured by a composition of 41.8% cellulose,16.4% hemicellulose,and 28.0% lignin.

During the pretreatment,the biomass did not completely dissolve in the ionic liquid.Furthermore,all the mixtures exhibited a dark brown appearance after the dissolution process as is visible from Fig.7.In particular,different studies dealing with cellulose dilution and regeneration in ionic liquids have reported that the corresponding color change is related to the lignin content and distribution present in the samples affecting the dissolution process[50,51].This observation can be extended to different residual material rather than rice husk.Along these lines,Hasanovetal.[52] give an overview of the effect of lignin on the dissolution of lignocellulosic biomass in ionic liquids,allowing to establish that,among of variables affecting the dissolution process,the amount of lignin present in the initial concentration of biomass play a major role on partial dissolution of the corresponding material.Composition profile is necessary to test the effect of the pretreatment with [EMIM][Cl] in the native samples or rice husk.It is important to note that the small variation in the composition of the biomass components after the pretreatment suggests that the effect of the ionic liquid is mostly morphological and,therefore,only physical changes in the cellulose structure take place as we detailed in Section 4.4.Regarding to dissolution of cellulose,it was checked visually and during the test we could observe that part of the rice husk biomass was partially suspended on the top as supernatant (see Fig.7(a)),followed by a deep coloration of the solution (dark-brownish solutions).It suggests a possible degradation of the former material (see Fig.7(b) and (c)).[EMIM][Cl] showed very good dissolving properties as 13.21% of the pretreated material could be dissolved and the rest 86.79% is recovered.Based on this high yield of recovery,we can envisage a significant benefit for the use of [EMIM][Cl] in the pretreatment of rice husk.By contrast,the solutions tend to crystallize at room temperature as a consequence of the high melting point caused by [EMIM][Cl].

Fig.7.Pretreatment of rice husk with [EMIM][Cl]: (a) zero hours,initial time (b) four hours (c) twelve hours,final time.

The comparison of the solid recovery percentage is made with previous studies of biomass treatment using ionic liquids-based imidazolium.The solids recovery is the result of the ionic liquid solvation in the biomass,a high solid recovery represents a low solvation capability of ionic liquid[53].Trinchetal.[54]evaluated the temperature effect on the pretreatment of mixed softwood biomass using [BMIM][Cl];they found at 403.15 K and 15 h,a solid recovery of 83.6% and solvation capability of 16.3%,presenting the best performance in enzymatic digestibility.In another study,Trinchetal.[55] used [BMIM][Ac] for the same biomass,and the optimum point that presented the best fermentable sugar yield was at 373.15 K and 15 h;they obtained a solid recovery of 92%.However,there are other studies with shorter processing times;for example,Bahceguletal.[56],for the pretreatment of cotton stalks,used two ionic liquids: [EMIM][Cl] and [EMIM][Ac];they found a solid recovery of 81.7% and 75.9% at 423.15 K and 30 min,respectively,which results in a good performance of glucose yield.As it can be seen in the previous studies,the solid recovery is not only related to the solvation capability of ionic liquid towards the biomass,but it also favors the enzymatic digestibility performance on the pretreated material.Furthermore,it also depends on the temperature,processing time,and type of biomass.Then,we can be concluded that one of the factors that can favor enzymatic digestibility is a high solids recovery percentage due to low solubility of the pretreated material,as it is evidenced in our work.

The enzymatic hydrolysis performance after pretreatment with[EMIM][Cl] is presented in Fig.8.Here,we compare the reducing sugar amount for untreated and pretreated rice husk with[EMIM][Cl] at the same enzyme loading.Note that the samples of rice husk pretreated with [EMIM][Cl] give rise to a better digestibility of cellulose,which can be noticed from the amount of glucose recovered per 100 g of sample.In this way,the amount of sugars recovered from the pretreated material is 28.53 g per 100 g of rice husk,whereas for untreated material was just 1.43 g.These results can be contrasted with data shown in Table 1 and they allow to conclude that [EMIM][Cl] strongly disrupts the former structure of the cellulose chains present in the biomass of the native rice husk allowing to obtain an easier hydrolysable material.Considering the obtained results,as it can be noticed in Table 1,the performance of [EMIM][Cl] is better than the one reported for [EMIM][DEP] and very close or similar to the value obtained using[BMIM][Cl]in all cases.It is also important to highlight that although [EMIM][OAc] exhibit higher performance than[EMIM][Cl],the latter is less hygroscopic giving rise to considerable benefits in the dissolution of cellulose [32].

Fig.8.Performance comparison of the hydrolysis on rice husk samples pretreated in [EMIM][Cl] and untreated after 43 h.

Regarding the results presented in Table 4,variations in the composition of the constituent material are due to the fact that the interactions between of [EMIM][Cl] and biomass are mostly physical.There is not a loss of the chemical identity in each of the biomass constituents due their interaction with [EMIM][Cl].After the process,86.8% of the initial sample was recuperated.On average,85% of cellulose,95% of hemicellulose,and 92% of lignin were conserved in a pretreated material,while ash was 98%.Rice husk residues regenerated from the ionic liquid pretreatments have similar composition profile than untreated rice husk,but with carrying extent of surface disruption and swelling.It is worth noting that our measurements agree with the results reported in Ref.[17]for the cellulose dissolution of rice husk in[BMIM][OAc](69%)and[BMIM][Cl](59%)at 373.15 K and 10 h.The results for the pretreatment of rice husk were similar to[EMIM][Cl](28.9%)reported in Ref.[18].

4.3.Biomass crystallinity

The rice husk is a complex lignocellulosic material with crystalline and amorphous areas.The apparent crystallinity can be studied and its possibility to produce fermentable sugars by enzymatic hydrolysis can be evaluated.We evaluated the effect of ionic liquids-based pretreatment on CrI (See Table 5).

Table 5 Crystallinity index for untreated and pretreated material.

When the results presented in Fig.9 are compared with the corresponding peaks reported in the open literature,it can be seen that our diffractogram follows the angle of the main sharp peak,the shape,and width than the peaks reported for several samples of rice husk as follows.(i) extracted cellulose from rice husk[57]: 2θ≈17° and 2θ≈22°,(ii) powdered rice husk [58]: 2θ≈18°,2θ≈22°,(iii) cleaned rice husk [59]: 2θ≈16° and 2θ≈22°,and (iv)raw rice husk samples[60]: 2θ≈16° and 2θ≈22.5°.The mentioned works allow to establish that,for rice husk biomass,the XRD patterns exhibit a sharp peak near to 2θ≈(22°,23°) representing the crystalline part of the material and the other close to 2θ≈(16°,18°) representing the amorphous one.

Fig.9.XRD patterns of rice husk samples.Patterns for untreated material (blue line) and patterns for the pretreated samples in [EMIM][Cl] (black line).

Nevertheless,as mentioned in Section 3.4.1,we follow the Segal’s method to evaluate the crystallinity [9,38],which is essentially based on the intensity measured at two points in the diffractogram: the height of the main and sharp peak and the minimum between this and peaks 2θ=18°.We used the same procedure that[57-60] to obtain the ratio between the intensities of the crystalline and material’s amorphous part.After the pretreatment with ionic liquid the wide peak (2θ≈18°) disappeared and the main peak(2θ=22.8°) exhibited an appreciable lower intensity,indicating that the pretreatment modified the crystal planes for cellulose I,transforming it into cellulose II.Table 5 shows the effect of the pretreatment with [EMIM][Cl] on the crystallinity index.The pretreated rice husk exhibited a reduction of 22% in the cellulose crystallinity.

4.4.Biomass morphological analysis by SEM

In order to compare the scanning electron micrographs of the untreated and ionic liquid pretreated rice husk,we included Fig.10 with several magnification options.Note that SEM micrographs of untreated (Fig.10.I) and pretreated (Fig.10.II) rice husk are significantly different as the bundle of fiber in the latter show complete unpacking.This means that the former material is featured by a compact structure and smoothness surface as it is depicted in Fig.10.I.(b) and 10.I.(c),and it losses its structural order.Essentially,from Fig.10,we highlighted that the loss of the structural order is a consequence of the hydrogen bonds disruption in the cellulose chains,and the strength decreasing in the interactions linking the cellulose and lignin chains.Thus,there was no evidence of lignin accumulation.It means that the linkages between cellulose and lignin were disrupted as a consequence of the pretreatment in [EMIM][Cl] making weaker the dispersion forces.But,on the other hand,based on the amorphous structures depicted in Fig.10.II,it is apparent the hydrogen bonds disruption in the cellulose chains,which are essentially responsible for the rigid and highly compact structure.The shaped-seeds spots coming up from the surface elucidate the presence of silica in the material.Also,note in Fig.10.I.(a) and 10.II.(a) the change in the visual perception of raw samples and treated samples.

Fig.10.I.(a)Untreated material morphology,(b)SEM-×300(50 μm),(c)SEM-×1000(10 μm),and(d)SEM-×2500(10 μm).II.(a)Morphology of material treated with ionic liquids,(b) SEM-×500 (50 μm),(c) SEM-×1000 (10 μm),and (d) SEM-×2500 (10 μm).

5.Conclusions

Despite of the fact that [EMIM][Cl] exhibit a relatively simple molecular structure and it has been widely studied at theoretical level,there is a consensus neither in the path nor in the optimal conditions to carry out the synthesis.By implementing a consistent thermodynamic analysis our current work was devoted to find out the optimal procedure to synthesize[EMIM][Cl].FTIR characterization and values measured for the most representative properties validate the ionic liquid formation and the high yield achieved by the reaction following the proposed path.The synthesized[EMIM][Cl] was implemented,for the first time,in the pretreatment of rice husk.The results obtained here allow to conclude that the treated material clearly loss of the former cellulose crystalline structure.It was clearly visible from the results obtained in the structural (XRD) and morphological (SEM) analysis.The data suggest that pretreatment with [EMIM][Cl] improves the enzymatic hydrolysis of rice husk in terms of the amount of recovered reducing sugars.This result indicates that this ionic liquid is very effective in disrupting the biomass recalcitrance of the native rice husk allowing to obtain a significant amount of post-hydrolysis fermentable sugars.Regarding the interactions between synthesized [EMIM][Cl] and silica,we did not notice any sign of reaction nor influence on the bulk properties of the solutions.Finally,it should be noted that the implementation of [EMIM][Cl] for rice rusk pretreatment is advantageous and more beneficial than other reported ionic liquids due to the synthesis is simple,cheaper,and can be obtained from widely available materials.In addition,from the experimental measurements carried out in our current work,we have proofed that[EMIM][Cl]have equivalent or slightly better yields to dissolve rice husk cellulose than other and more complex ionic liquids.

Data Availability

All data supporting the experiments performed and the conclusions of the study are presented within this manuscript.However,if the reader considers that additional data are required,the corresponding authors (R.Vázquez-Medina: ruvazquez@ipn.mx and Felipe A.Perdomo: fperdomo@ed.ac.uk) can be contacted.For third-party data,we have indicated inside the manuscript the Internet address to access them.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors thank the research program of Universidad de Bogotá Jorge Tadeo Lozano 703-12-15 for financial support.