RSC Advances

PAPER

Role of ionic liqu

aDepartment of Physics, Banaras Hindu Un

rksingh_17@rediffmail.com; Fax: +91 542 2bDepartment of Physics and Astrophysics, Un

Cite this: RSC Adv., 2015, 5, 8263

Received 22nd October 2014Accepted 4th December 2014

DOI: 10.1039/c4ra12951b

www.rsc.org/advances

This journal is © The Royal Society of C

id [BMIMPF6] in modifying thecrystallization kinetics behavior of the polymerelectrolyte PEO-LiClO4

S. K. Chaurasia,ab Shalu,a A. K. Gupta,a Y. L. Verma,a V. K. Singh,a A. K. Tripathi,a

A. L. Saroja and R. K. Singh*a

We report on the modification in crystallization kinetics behavior of PEO + 10 wt% LiClO4 polymer

electrolyte by the addition of an ionic liquid 1-butyl-3-methylimidazolium hexafluorophosphate

(BMIMPF6). Three techniques have been used for studying crystallization kinetics, viz., (i) isothermal

crystallization technique using DSC, (ii) non-isothermal crystallization technique using DSC, and (iii) by

monitoring the growth of spherulites with time in the polymer electrolyte films using a polarizing optical

microscope (POM). Results from all the three techniques show that the presence of ionic liquid BMIMPF6suppresses the crystallization rate due to its plasticization effect. Isothermal crystallization data was well

described by the Avrami equation, and Avrami exponent n lies in the range of 1 to 2, which signifies 2D

crystal growth geometry occurring in these polymer electrolytes under the investigated temperature

range. However, the Avrami crystallization rate constant ‘K’ increases exponentially with crystallization

temperature and ionic liquid content as well. However, the non-isothermal crystallization kinetics of

these polymer electrolytes is discussed in terms of three different models (Jeziorny's, Ozawa's and Mo's

method), and it is found that Mo's method better explains the non-isothermal crystallization data. In

addition, crystalline morphology and spherulite growth were studied by POM, which shows the

suppression in crystallization in the presence of ionic liquid, as confirmed by spherulite growth rate (Gs)

analysis.

Introduction

The development of ion-conducting materials (such as poly-mers, gels, and glasses) has recently attracted global interest indeveloping a variety of electrochemical devices includingrechargeable batteries, fuel cells, supercapacitors, solar cells,and actuators.1–5 A straightforward strategy adopted for thispurpose is to integrate mobile ionic species into different solid/polymeric so matrices. Polymers have advantages over otherion conducting materials because they exhibit various favorableproperties such as ease of fabrication in thin lm form, and theyare mechanically, thermally and electrochemically more stable.6

One of the necessary conditions, in addition to thermal/mechanical/chemical stability, is to make a polymer backboneconducive for providing high ionic mobility for ionic movement.Earlier, many ion conducting polymer matrices or polymerelectrolytes were formed by complexing ionic salts (such as Li+,Na+, Mg2+, and Zn2+) with polar polymers (such as PEO, PMMA,PVA, and PVdF) and were much studied, but their roomtemperature ionic conductivity is very low (�10�6 to 10�7 S cm�1)

iversity, Varanasi-221005, India. E-mail:

368390; Tel: +91 542 2307308

iversity of Delhi, Delhi-110007, India

hemistry 2015

because of its high degree of crystallinity, which hinders themotion of the ions in the polymer network.7–11 Therefore, variousapproaches have been adopted to improve their ionic conduc-tivity, which includes (a) addition of low molecular weightplasticizers/organic carbonates,12,13 such as ethylene carbonate(EC), propylene carbonate (PC), diethyl carbonate (DEC), and (b)use of inorganic llers14–16 such as SiO2, Al2O3, CNT, TiO2. Theuse of conventional plasticizers in polymer electrolytes canenhance the ionic conductivity by lowering their glass transitiontemperature (Tg), which further increases the amorphous phaseof the polymer matrix but also cause poor mechanical andthermal stability and a narrower electrochemical potentialwindow due to their volatile nature, which limits their applica-tion in devices.17 Recently, a new approach has been adopted toenhance the ionic conductivity of the polymer electrolytemembranes by incorporating ionic liquids (ILs) into the polymermatrix.18–20 ILs have attracted considerable attention as anexcellent alternative to the conventional plasticizers due totheir distinct properties such as wide liquidus range,non-volatility, non-ammability, negligible vapour pressure,wide electrochemical stability window, high ionic conductivityand excellent thermal and chemical stability.21–23 ILs areentirely composed of bulky and asymmetric organic cationsand inorganic anions. Due to the unique properties of ILs,

RSC Adv., 2015, 5, 8263–8277 | 8263

Fig. 1 Heat Flow vs. time plots for PEO + 10 wt% LiClO4 + x wt%BMIMPF6 for (a) x ¼ 0, (b) x ¼ 5 and (c) x ¼ 10 during isothermalcrystallization at different Tc.

RSC Advances Paper

polymer electrolytes based on ionic liquid offer high conductivityalong with improved thermal and mechanical properties. Inpolymer electrolytes, ionic liquid plays the role of a plasticizer aswell as a supplier of free charge carriers for ion conduction.24

In most of the polymer electrolytes, the amorphous phaseis found to be more conducting than the crystalline phase.Accordingly, it is very important to learn about the crystalli-zation kinetics behavior of polymer as well as polymer elec-trolytes.25–28 Crystallization is a process of phasetransformation that involves the transformation of a disor-dered amorphous phase into a single or multi ordered phase.Polymer crystallization is a complex process that affects thenal properties of the materials.29 It has also been observedthat some polymers can crystallize, while some cannot.Among the polymers that crystallize, the degree of crystalli-zation, the structure of crystal and crystal size depend upon anumber of parameters, such as temperature, time, concen-tration of solution, and stress present during the crystalli-zation.30 Some studies are available on the modication incrystallization behaviour of polymers on the addition ofcomplexing salts31 or by changing its molecular weight,32

using inorganic llers,33 carbon nanotubes,34 and also inconned geometries.35 The polymer PEO is known to be a“semicrystalline” polymer that consists of both crystalline andamorphous phases, and its high capability in formingcomplexes with many salts, as well as its high chemicalstability, led to its emergence as a promising host matrix forthe preparation of polymer electrolytes. However, PEO tendsto crystallize due to its highly ordered chain structure, whichimpedes the ion transport in the polymeric matrix. This is ageneral observation for semicrystalline polymer electrolytesand is well documented in literature.36,37 Therefore, it wouldbe interesting to study the crystallization kinetics behavior ofsuch a semicrystalline polymer PEO including its crystalstructure and crystalline morphology. The degree of crystal-linity is of particular interest for the better understanding ofthe structure–property relationship to signicantly improvethe performances of the solid state devices containing PEO-based polymer electrolytes.

This paper reports the crystallization kinetics behavior of thepolymer electrolyte (PEO + 10 wt% LiClO4) with differentamounts of added ionic liquid (BMIMPF6). Crystallizationkinetics was studied by isothermal and non-isothermal crys-tallization methods using DSC, and the affirmative conrma-tion of the crystallization behavior was obtained by examiningthe expansion of spherulites by polarizing optical microscope(POM). It has been found that the addition of BMIMPF6 intoPEO + 10 wt% LiClO4 slows the crystallization rate due to theplasticization effect of the ionic liquid BMIMPF6.

Experimental sectionMaterials and method

The starting materials used for the preparation of polymerelectrolyte lms are poly(ethylene oxide), PEO of average mol. wt6 � 105 g mol�1, lithium salt LiClO4 and ionic liquid, 1-butyl-3-methylimidazolium hexauorophosphate BMIMPF6, obtained

8264 | RSC Adv., 2015, 5, 8263–8277 This journal is © The Royal Society of Chemistry 2015

Paper RSC Advances

from Sigma Aldrich. The ionic liquid BMIMPF6 was vacuumdried at 10�6 torr for 24 h before use.

Fig. 2 The Xt vs. t plot for PEO + 10 wt% LiClO4 + x wt% BMIMPF6 for(a) x ¼ 0 (b) x ¼ 5 and (c) x ¼ 10 at different Tc, viz., 44, 46, 48, 50 �C.

Fig. 3 The rate of crystallization (G ¼ 1/t1/2) vs. crystallizationtemperature plots for PEO + 10 wt% LiClO4 + x wt% BMIMPF6 for (a) x¼ 0 (b) x ¼ 5 and (c) x ¼ 10 at different crystallization temperatures 44,46, 48, 50 �C.

This journal is © The Royal Society of Chemistry 2015

Polymer electrolyte lms of PEO + 10 wt% LiClO4 + x wt%BMIMPF6 (for x ¼ 0, 5, 10, 15 and 20) were prepared by asolution casting method. The particular weight ratio of 10 wt%LiClO4 in PEO was chosen for incorporating varying amounts ofionic liquid BMIMPF6 to prepare PEO + 10 wt% LiClO4 +x wt%BMIMPF6 polymer electrolyte lms because of its excellentmechanical stability as well as reasonable ionic conductivitycompared to other high lithium salt-containing samples.Though the high conducting lms containing higher loading oflithium salt LiClO4 could be prepared without BMIMPF6, onloading of ionic liquid BMIMPF6 these lms tend to becomeunstable. In the solution casting method, polymer PEO wasdissolved in methanol with stirring at 40 �C and then requisiteamount of LiClO4 was added and stirred until it appeared as ahomogeneous solution. Subsequently, the required amount ofBMIMPF6 was added to the abovementioned solution and stir-red again for 2–4 h until a viscous solution was obtained. Theviscous solution so obtained was poured into polypropylenePetri dishes. Aer the complete evaporation of the solvent, PEO+ 10 wt% LiClO4 + x wt% BMIMPF6 polymer electrolyte lmscontaining different amounts of BMIMPF6 were obtained.These lms were vacuum dried before further use.

Three methods, viz., isothermal, non-isothermal and opticalmicroscopic study, were used for studying the crystallizationkinetics. The isothermal and non-isothermal crystallizationkinetics of the prepared samples were studied using a differ-ential scanning calorimeter (Mettler Toledo DSC1 system). Allthe DSC measurements were conducted under a nitrogenatmosphere. The detailed procedures of the isothermal andnon-isothermal method are given in their respective sectionswhere results are discussed.

For studying the crystalline morphology and spherulitegrowth rate, a Lietz DMR polarizing microscope was used. Allpolarizing optical microscopy (POM) studies were done at a

RSC Adv., 2015, 5, 8263–8277 | 8265

Fig. 4 Avrami plots using isothermal method for (a) PEO + 10 wt%LiClO4 (b) PEO + 10 wt% LiClO4 + 5 wt% BMIMPF6 and (c) PEO + 10wt% LiClO4 + 10 wt% BMIMPF6 at different crystallization temperatures(viz., Tc ¼ 44, 46, 48, 50 �C).

Table 1 Different crystallization parameters of PEO + 10 wt% LiClO4 +x wt% BMIMPF6 films obtained by Avrami plots using the isothermalcrystallization method

PEO + 10 wt%LiClO4 + x wt% BMIMPF6 Tc (�C) n K (min�n) t1/2 (min)

(a) x ¼ 0 44 1.48 0.70 0.9446 1.40 0.39 1.4548 1.65 0.07 3.8550 1.86 0.02 3.95

(b) x ¼ 5 44 1.55 0.66 1.0246 1.53 0.40 1.3548 1.63 0.10 4.0950 1.76 0.06 7.09

(c) x ¼ 10 44 1.34 0.33 1.5746 1.31 0.11 3.4948 1.70 0.02 10.6950 1.74 0.01 15.08

Fig. 5 ln K1/n vs. 1/Tc plot for PEO + 10 wt% LiClO4 + 10 wt% BMIMPF6

RSC Advances Paper

magnication of 50�. For carrying out the POM-studies, thesamples were rst heated above the melting temperature (Tm) ofthe polymer electrolyte lms and held there for some time until

8266 | RSC Adv., 2015, 5, 8263–8277

a completely isotropic amorphous phase was observed in POM.Then, the polymer lms were quickly quenched to the desiredtemperature of crystallization (these temperature are less thanTm of the polymer electrolytes). The size of spherulites as afunction of time elapsed aer their initial appearance wasmonitored by POM for crystallization study.

Crystallization kinetics by isothermal method

For carrying out the actual experiment, the samples were heatedto �75 �C (above the melting temperature of polymer Tm) andheld there for 5 min to erase the thermal history, and thenquickly cooled to the temperature at which crystallization wasdesired to be studied (note that these temperatures are less than(Tm)onset as determined by DSC thermograms). The sampleswere maintained at desired crystallization temperatures, viz., Tc¼ 44, 46, 48 and 50 �C, to crystallize, and DSC exothermic curvesfor heat ow vs. time were recorded. These DSC isothermalcurves obtained for different samples were used for studying thecrystallization kinetics by an isothermal method. The heat ow

during isothermal crystallization.

This journal is © The Royal Society of Chemistry 2015

Fig. 6 Heat flow vs. temperature plot for PEO + 10 wt% LiClO4 + xwt% BMIMPF6 films during non-isothermal crystallization for (a) x¼ 0 (b) x¼ 5(c) x ¼ 10 (d) x ¼ 15 and (e) x ¼ 20 at different cooling rates by DSC.

Paper RSC Advances

vs. time plot for PEO + 10 wt% LiClO4 + x wt% BMIMPF6 lms(for x¼ 0, 10 and 20) are shown in Fig. 1. It can be seen that withincreasing crystallization temperature, the crystallizationexothermic peak attens, possibly due to increase in the exi-bility of polymer PEO backbone, which suggests that thesamples take relatively longer time to crystallize as theyapproach their (Tm)onset.

The crystallization kinetics of polymers under isothermalconditions is described by the well-known Avrami equation38 interms of the dependence of relative crystallinity (Xt) on thecrystallization time (t) as follows:

Xt ¼ 1 � exp(�Ktn) (1)

where Xt is the relative crystallinity generated at any time t, n isthe Avrami exponent dependent on the nucleation and growthgeometry of the crystal, and K is the overall crystallization rateconstant associated with nucleation as well as growthcontributions.

This journal is © The Royal Society of Chemistry 2015

The relative crystallinity (Xt) generated at any time (t) can beobtained using the DSC exothermic isothermal curves, asillustrated in Fig. 1. The relative crystallinity (Xt), is dened asthe ratio of crystallinity generated at any time t to the crystal-linity when time approaches innity. Xt has been evaluatedusing the following relation:

Xt ¼ DHt

DHN

¼

ðt0

�dH

dt

�dt

ðN0

�dH

dt

�dt

(2)

where dH/dt is the rate of heat evolution, DHt is the total heatevolved at any time t and DHN is the heat evolved when timeapproaches innity (N).

Obviously, the values of Xt at a given crystallization time t canbe obtained by integrating the area of exothermic DSCisothermal curves between time t ¼ 0 to t, divided by the entirearea of the exothermic peaks.

RSC Adv., 2015, 5, 8263–8277 | 8267

Fig. 7 Xt vs. temperature plot during non-isothermal crystallization process of PEO+ 10wt% LiClO4 + xwt% BMIMPF6 for (a) x¼ 0 (b) x¼ 5 (c) x¼10 (d) x ¼ 15 and (e) x ¼ 20 at different cooling rates (viz., 5, 10, 15, 20 �C min�1).

RSC Advances Paper

Using eqn (2), the conversion curves of Xt (the crystallizedfraction) versus t for PEO + 10 wt% LiClO4 + x wt% BMIMPF6lms (for x¼ 0, 5 and 10) at various crystallization temperatures(Tc ¼ 44, 46, 48 and 50 �C) are shown in Fig. 2. All the curvesshow sigmoid characteristics with time and shi towards ahigher time regime as the crystallization temperature (Tc)increases. These features indicate that at higher Tc, the crys-tallization rates become slow and samples take more time tocrystallize because the system approaches (Tm)onset. Further, acomparison of the results of polymer electrolyte PEO + 10 wt%LiClO4 (see Fig. 2, curve ‘a’) with those polymer electrolyte lmsdoped with ionic liquid, BMIMPF6 (see Fig. 2, curves b and c)showed that BMIMPF6-containing samples take a longer time tocrystallize in comparison to pristine polymer electrolyte due tothe increase in amorphicity of the samples.39 For example, wecan see that the pristine polymer electrolyte PEO + 10 wt%LiClO4 took nearly 1.8 min to obtain complete crystallization atTc ¼ 44 �C, and the time required to nish crystallization for 5wt% BMIMPF6-containing polymer electrolyte slightly

8268 | RSC Adv., 2015, 5, 8263–8277

increased. It was found to be �2 min at the same crystallizationtemperature, whereas at higher Tc (e.g. 50 �C), it takes �15 minto complete crystallization (for detail see Fig. 2c).

These observations clearly indicate that with increasingcrystallization temperature or ionic liquid concentration, thesamples take longer time to crystallize because of the plastici-zation effect of ionic liquid.40–42 In our previous studies43,44 onthe crystallization kinetics of polymer PEO upon the inclusionof ionic liquid as well as lithium salt, it has been shown that thecrystallization rate of PEO changes more on addition ofBMIMPF6 alone rather than for (PEO + LiClO4) because theplasticization effect of ionic liquid is less experienced by thelatter, which has already been amorphosized by the addition ofLiClO4. This effect will become clearer while discussing theresults of crystallization half time (t1/2) in the subsequentsection.

The crystallization half time (t1/2), which is dened as thetime necessary to attain 50% of the nal crystallinity of thesamples, is an important parameter for discussing the

This journal is © The Royal Society of Chemistry 2015

Fig. 8 Xt vs. time plot during non-isothermal crystallization process of PEO+ 10 wt% LiClO4 + xwt% BMIMPF6 for (a) x¼ 0 (b) x¼ 5 (c) x¼ 10 (d) x¼ 15 and (e) x ¼ 20 at different cooling rates (viz., 5, 10, 15, 20 �C min�1).

Paper RSC Advances

crystallization kinetics.45 The values of t1/2 are directly obtainedfrom Fig. 2, from which the rate of crystallization (G¼ 1/t1/2) canbe calculated. The greater the value of t1/2, the lower is the rateof crystallization. Fig. 3 shows the variation of G for PEO + 10wt% LiClO4 + x wt% BMIMPF6 at various crystallizationtemperatures (Tc). As illustrated in Fig. 3, the values of Gdecrease with increasing Tc for all the samples, indicating thatthe overall crystallization rate becomes slow at higher Tcbecause the nucleation process is more difficult at highercrystallization temperatures.46 Furthermore, it may be notedhere that the values of G (¼1/t1/2) obtained for PEO + 10 wt%LiClO4 + x wt% BMIMPF6 lms are lower in comparison to thatof pristine polymer electrolyte PEO + 10 wt% LiClO4 at a givenTc, as shown in Fig. 3, which clearly indicates the slowing of thecrystallization rate by the incorporation of BMIMPF6. In addi-tion, the reduction in overall crystallization rate G of polymerelectrolytes with increasing BMIMPF6 concentration is attrib-uted to the fact that the presence of ionic liquid BMIMPF6 exertsa dilution effect for the crystallizable polymer PEO due to the

This journal is © The Royal Society of Chemistry 2015

plasticization effect of the ionic liquid. Our previous studies47–49

also showed of the similar situation in which the effects of ionicliquid on the properties of polymers and polymer electrolyteswere studied and it was found that the ionic liquid acts as anefficient plasticizer.

Avrami plots (i.e., double logarithmic graphic representa-tions of crystallization data) are generally used to calculate thevalues of n and K. They can be obtained by taking the doublelogarithmic of eqn (1) and can be written as follows:

log[�ln(1 � Xt)] ¼ log K + n log t (3)

According to eqn (3), the plot of log[�ln(1 � Xt)] vs. log tshould be a straight line. In the present study, the Avramiplots for PEO + 10 wt% LiClO4 lms containing differentamounts of BMIMPF6 give rise to a series of parallel straightlines at different crystallization temperatures (viz., Tc ¼ 44,46, 48 and 50 �C), as shown in Fig. 4. Aer linear tting ofthese plots by a straight line at different Tc, both the Avrami

RSC Adv., 2015, 5, 8263–8277 | 8269

RSC Advances Paper

exponent ‘n’ and crystallization rate constant ‘K’ can beobtained by the slope and intercept of the straight line,respectively. The various Avrami parameters obtained fromisothermal crystallization method for PEO + 10 wt% LiClO4 +x wt% BMIMPF6 at various crystallization temperatures aregiven in Table 1.

It is known that the values of Avrami exponent are theconsequences of the geometry of specic crystal dimension andhave been used to specify the dimension of growing crystals.The value of Avrami exponent n is assumed to lie in the rangebetween 1 and 4 and is related to the geometry characteristics ofthe growing crystals: n ¼ 1 is ascribed to the 1D structure, 2 isascribed to the 2D structure, and 3 or 4 ascribed to the 3Dstructure.50 In the present study, the values of ‘n’ for all thepolymer electrolyte PEO + 10 wt% LiClO4 + x wt% BMIMPF6lms lie between 1 and 2 at all crystallization temperaturesstudied. This indicates that the 2D crystal growth morphologydominates at all temperatures at which crystallization has beenstudied.

Fig. 9 Avrami plots using non-isothermal method of PEO + 10 wt% LiClOx ¼ 20 at different cooling rates.

8270 | RSC Adv., 2015, 5, 8263–8277

The effect of the incorporation of ionic liquid into the poly-mer membrane, apart from decreasing the rate of crystalliza-tion, is also expected to reect the values of activation energy forcrystallization. To check this, we determined this activationenergy for PEO + 10 wt% LiClO4 polymer electrolyte withBMIMPF6. The Avrami crystallization rate constant K can beassumed to be a consequence of a thermally activated processand has been used to determine the activation energy for crys-tallization.51 The crystallization rate constant K can beexpressed by the Arrhenius equation as follows:

K1/n ¼ Ko exp(�DE/RTc) (4)

Eqn (4) can also be written as

ln K1/n ¼ ln Ko � DE/RTc (5)

where Ko is the pre-exponential factor independent of temper-ature, DE is the isothermal crystallization activation energy, R isthe gas constant and Tc is the crystallization temperature.

4 + xwt% BMIMPF6 films for (a) x¼ 0 (b) x¼ 5 (c) x¼ 10 (d) x¼ 15 and (e)

This journal is © The Royal Society of Chemistry 2015

Paper RSC Advances

By plotting the ln K1/n versus 1/Tc, DE can be obtained by theslope of these curves. Typical plots for PEO + 10 wt% LiClO4 + 10wt% BMIMPF6 are shown in Fig. 5. The slope of this curve givesthe value of DE/R, from which the isothermal crystallizationactivation energy DE can be directly calculated. The value ofisothermal crystallization activation energy DE for PEO + 10wt% LiClO4 + 10 wt% BMIMPF6 polymer electrolyte was foundto be 97 kJ mol�1.

Crystallization kinetics by a non-isothermal method

Non-isothermal crystallization method using DSC is also one ofthe most convenient methods for studying the crystallizationkinetic behavior of polymers. In this method, samples wereheated above the melting temperature of the polymer PEO (�75�C), kept there for 5 min to erase the thermal history, and thencooled at different cooling rates (viz., 5, 10, 15 and 20 �Cmin�1).DSC exothermic curves for heat ow versus temperature wererecorded to analyze the non-isothermal crystallization data.Samples used for non-isothermal crystallization were the samefor which isothermal crystallization kinetic studies were carriedout. The DSC exothermal curves for PEO + 10 wt% LiClO4 + xwt% BMIMPF6 lms at different cooling rates 5, 10, 15 and 20 �Cmin�1 are shown in Fig. 6. It can be seen that the exothermiccrystallization peaks were shied to lower temperature side,which became broader with increasing cooling rates. To nd thenon-isothermal crystallization kinetics, relative crystallinity (Xt)versus temperature plots for these polymer electrolyte lms wereobtained using eqn (2) and are shown in Fig. 7 (the procedure issimilar to that used for the isothermal crystallization methoddescribed earlier). Fig. 7 shows the plots of Xt versus tempera-ture (T), which illustrate an anti-S-shaped characteristic. In non-isothermal crystallization process, temperature scale of Xt versus

Table 2 Different crystallization parameters of PEO + 10 wt% LiClO4 +crystallization method

PEO +10 wt% LiClO4 + x wt% BMIMPF6 Cooling rate f (�C min�1)

(a) x ¼ 0 5101520

(b) x ¼ 5 5101520

(c) x ¼ 10 5101520

(d) x ¼ 15 5101520

(e) x ¼ 20 5101520

This journal is © The Royal Society of Chemistry 2015

T plots could be converted into Xt versus t using the followingrelation:52

t ¼ To � T

f(6)

where t is the crystallization time, To is the onset temperature ofcrystallization (t ¼ 0), T is the crystallization temperature and f

is the cooling rate. Using eqn (6), the plot of relative crystallinityversus temperature (Xt vs. T) can be easily transformed to theplot of relative crystallinity versus time (Xt vs. t). The Xt vs. t plotsfor PEO + 10 wt% LiClO4 + x wt% BMIMPF6 lms (for x ¼ 0, 5,10, 15 and 20) are given in Fig. 8, showing that higher thecooling rate, shorter is the time required to complete crystalli-zation and vice versa.

Several approaches have been used for describing the crys-tallization process involved in the non-isothermal crystalliza-tion kinetics, which are based on various models, including themodied Avrami equation by Jeziorny,53 Ozawa analysis54 andMo's method.55 These approaches are discussed in the subse-quent sections.

Modied Avrami equation. The modied Avrami53 equationis frequently used to analyze the non-isothermal crystallizationprocess. Considering the non-isothermal crystallization char-acter of the process, the Avrami equation can be written asfollows:

Xt ¼ 1 � exp(�Zttn0) (7)

where n0 is a constant depending on the type of nucleation andcrystal growth dimension, also known as non-isothermalAvrami exponent, and Zt is the non-isothermal crystallizationrate constant and depends on nucleation and growth parame-ters. The values of n0 and Zt were obtained from the slope and

x wt% BMIMPF6 films obtained by Avrami plots using non-isothermal

n0 Zt (min�n0) Zc t 01/2 (min)

2.29 0.139 0.674 1.962.27 0.516 0.936 1.132.01 0.651 0.971 1.022.05 0.717 0.986 0.882.539 0.096 0.626 2.162.225 0.067 0.961 1.042.215 0.839 0.988 1.141.742 0.900 0.994 0.852.260 0.138 0.673 1.962.242 0.516 0.936 1.132.017 0.662 0.972 1.001.893 0.751 0.985 0.912.374 0.106 0.638 2.172.107 0.311 0.889 1.421.387 0.439 0.946 1.071.642 0.608 0.975 0.842.199 0.095 0.625 2.471.913 0.289 0.883 1.571.844 0.398 0.940 1.381.503 0.599 0974 1.13

RSC Adv., 2015, 5, 8263–8277 | 8271

RSC Advances Paper

intercept of straight regime of plots drawn between log[�ln(1 �Xt)] and log t. The log[�ln(1 � Xt)] vs. log t plots for PEO +10wt% LiClO4 + x wt% BMIMPF6 (for x ¼ 0, 5, 10, 15 and 20) atdifferent cooling rates (viz., 5, 10, 15 and 20 �C min�1) areshown in Fig. 9. From these plots, we can see that at initialstages, these curves are tted well by eqn (7) but deviate fromthe linear relation on higher crystallization. Therefore, theAvrami equation cannot be accurately used for describing theentire non-isothermal crystallization kinetics process.

Jeziorny53 modied the Avrami equation for non-isothermalcrystallization processes; it assumes that curves with a xedcooling rate represent a series of isothermal crystallizationprocesses, and it gave a modied non-isothermal crystallizationrate (Zc) in which the effect of cooling rate ‘f’ on the value of Ztwas included as follows:

log Zc ¼ log Zt

f(8)

The values of the Avrami constants n0 and Zc obtained by themodied Avrami equation in a non-isothermal crystallization

Fig. 10 Ozawa plots of log[�ln(1 � XT)] vs. log f during the non-isother(a) x ¼ 0 (b) x ¼ 5 (c) x ¼ 10 (d) x ¼ 15 and (e) x ¼ 20.

8272 | RSC Adv., 2015, 5, 8263–8277

method are given in Table 2. It may be noted here that thevalues of n0 and Zc obtained in non-isothermal crystallization donot have the same magnitude as in the isothermal method dueto different experimental conditions used in these twomethods.

Ozawa analysis. Ozawa54 extended the Avrami equation byconsidering that the non-isothermal crystallization processconsists of a large number of innitesimal isothermal crystal-lization steps. According to the Ozawa theory, the non-isothermal crystallization process can be described by thefollowing equation:

X 0T ¼ 1� exp

�� K*ðTÞ

fm

�(9)

where X 0T is the relative crystallinity at temperature T, f is the

cooling rate, K*(T) is the cooling function related to the overallcrystallization rate,m is the Ozawa exponent, which depends onthe dimension of the crystal growth. eqn (9) can also beexpressed as follows:

log[�ln(1 � X 0T)] ¼ log K*(T) � m log f (10)

mal crystallization of PEO + 10 wt% LiClO4 + x wt% BMIMPF6 films for

This journal is © The Royal Society of Chemistry 2015

Fig. 11 log f vs. log t plot of PEO + 10 wt% LiClO4 + x wt% BMIMPF6 films for (a) x ¼ 0 (b) x ¼ 5 (c) x ¼ 10 (d) x ¼ 15 and (e) x ¼ 20 at differentcrystallinities using the non-isothermal crystallization method.

Paper RSC Advances

If the Ozawa theory is fully applicable for describingthe non-isothermal crystallization process, then the plot oflog[�ln(1 � X 0

T)] vs. log f should be a straight line, whichwas not the case for our samples, as shown in Fig. 10.Therefore, the Ozawa equation is not be fully applicable fordescribing the non-isothermal crystallization process of PEO+ 10 wt% LiClO4 polymer electrolytes as well as PEO + 10 wt%LiClO4 + x wt% BMIMPF6 containing different amounts ofBMIMPF6.

Mo's method. The kinetic equation proposed by Mo andcoworkers55 based on the combination of the Ozawa and Avramiequations has frequently been used for describing the non-isothermal crystallization process. This method has beensuccessfully applied for describing the non-isothermal crystal-lization behavior of many polymeric systems.56,57 The generalform of this equation can be written as follows:

log f ¼ log F(T) � b log t (11)

This journal is © The Royal Society of Chemistry 2015

where log F(T) ¼ [K*(T)/Zt]1/m and b is the ratio between the

Avrami and Ozawa exponents n and m. Here, the function F(T)refers to the value of cooling rate required to reach a deneddegree of crystallinity at a certain temperature in unit crystal-lization time.58 The higher value of F(T) gives lower crystalliza-tion rate. The plot between log f and log t for PEO + 10 wt%LiClO4 + x wt% BMIMPF6 lms at different cooling rates andvarious crystallinity gives a straight line as shown in Fig. 11. Thevalues of log F(T) and b can be obtained for the samples fordifferent values of ‘x’ from the intercept and slope, respectively.These values are listed in Table 3. From this, we can see that fora given degree of crystallinity (e.g., 10%), the value of log F(T) forPEO + 10 wt% LiClO4 polymer electrolyte is 0.607, whichincreases to 0.629, 0.634, 0.688 and 0.714, respectively, for thepolymer electrolyte containing 5, 10, 15 and 20 wt% BMIMPF6.In general, the value of F(T) for all the polymeric lms is foundto increase with the increasing amount of BMIMPF6 in PEO + 10wt% LiClO4. The abovementioned observation clearly indicatesthat the incorporation of ionic liquid into PEO + 10 wt% LiClO4

RSC Adv., 2015, 5, 8263–8277 | 8273

Table 3 Non-isothermal crystallization parameters of PEO + 10 wt%LiClO4 + xwt% BMIMPF6 at different degrees of crystallinities obtainedby the Li and Mo method

PEO +10 wt% LiClO4 + x wt% BMIMPF6 X 0

T in % F(T) b

(a) x ¼ 0 10 0.607 1.56320 0.840 1.57930 0.983 1.66240 1.086 1.67750 1.171 1.67160 1.311 1.78470 1.504 1.50480 1.727 1.786

(b) x ¼ 5 10 0.629 1.26320 0.864 1.18030 0.957 1.24540 1.059 1.35650 1.172 1.17260 1.331 1.33170 1.570 1.57080 1.753 1.583

(c) x ¼ 10 10 0.634 1.48520 0.842 1.57330 0.973 1.63540 1.086 1.69550 1.204 1.77460 1.343 1.85070 1.529 1.52980 1.713 1.713

(d) x ¼ 15 10 0.688 1.42020 0.932 1.59430 1.078 1.71440 1.208 1.81550 1.334 1.93660 1.485 2.06070 1.672 2.14980 1.820 1.997

(d) x ¼ 20 10 0.714 1.07520 0.952 1.26730 1.113 1.47340 1.254 1.63550 1.397 1.81860 1.546 1.96770 1.722 2.12180 1.866 2.126

Fig. 12 Kissinger plot of ln(f/Tp2) vs. 1/Tp for PEO+ 10wt% LiClO4 + 10

RSC Advances Paper

polymer electrolyte decreases the crystallization rate due to theplasticization effect of an ionic liquid.59–61 The addition of ionicliquid BMIMPF6 into PEO + LiClO4 polymer electrolyte hindersthe crystallization of polymer and leads to slowing of the crystalgrowth rate, as conrmed by the morphological studies.

Kissinger analysis. It has been already shown by theisothermal crystallization kinetics study that the incorporationof BMIMPF6 in PEO + 10 wt% LiClO4 polymer electrolytechanges the activation energy of the crystallization. We havealso re-conrmed this assertion from the non-isothermalkinetic study. The crystallization activation energy for a non-isothermal crystallization process can be obtained by the Kis-singer method.62 In this method, the effect of various coolingrates (f) on the crystallization peak temperature (Tp) has beenstudied. The crystallization activation energy (DE0) for a non-

8274 | RSC Adv., 2015, 5, 8263–8277

isothermal process using Kissinger method can be expressedas follows:

dhln�f.Tp

2�i

d�1�Tp

� ¼ �DE0

R(12)

where R is the gas constant and f is the cooling rate. The slopeof the plot ln(f/Tp

2) vs. 1/Tp gives a straight line, from which byknowing the value of DE0/R, DE0 can directly be calculated.Typical Kissinger plots for PEO + 10 wt% LiClO4) + 10 wt%BMIMPF6 polymer electrolyte are shown in Fig. 12. The value ofDE0 obtained for PEO + 10 wt% LiClO4 + 10 wt% BMIMPF6polymer electrolyte was found to be 37 kJ mol�1. This value issomewhat lower than those obtained for the isothermal crys-tallization method previously found because different experi-mental conditions were used in the two methods. In the non-isothermal crystallization method, the temperature changesconstantly, which affects the rate of nuclei formation as well asspherulite growth.

Crystalline morphology and spherulite growth by polarizedoptical microscopy

When polymers are crystallized from the melt, the mostcommonly observed structures are the spherulites. Thesespherulites are in the form of spherical aggregates of lamellae,characteristics of polymers crystallized isothermally in theabsence of pronounced stress or ow.63–65 The typical dimen-sions of spherulites are of the order of microns, and sometimeseven millimeters. Therefore, they can be easily viewed under apolarizing optical microscope. The spherulites continue to growradially until they impinge upon one another. A measure oftheir growth until the time of impingement provides abundantinformation regarding the mechanism of crystallization in thepolymer, and has been the focus of several investigations.66–68

The number and size of spherulites growing in the polymermatrix controls the overall crystallinity of the polymer. There-fore, measuring the size of spherulites as a function of time canprovide an estimate of the rate of crystallization. We used this

wt% BMIMPF6.

This journal is © The Royal Society of Chemistry 2015

Paper RSC Advances

technique for measuring the rate of crystallization of our poly-mer electrolyte membranes. The size of spherulites wasmeasured by a polarizing optical microscope (POM). Typicalgrowth of such spherulites appearing in PEO + 10 wt% LiClO4

and PEO + 10 wt% LiClO4 + 10 wt% BMIMPF6 polymer elec-trolyte lms at different times, crystallized at 50 �C, are shownin Fig. 13. We can see that spherulite growth was very rapid forPEO + 10 wt% LiClO4 polymer electrolyte: within 10 seconds itacquired �100 mm diameter, and then its size increased to 310mm at the elapsed time of 120 s, as shown in Fig. 13 curve a (i)and (v). As BMIMPF6 was added to PEO +10 wt% LiClO4 polymerelectrolyte, the spherulite size was reduced and the number of

Fig. 13 Image of spherulite growth of (a) PEO + 10 wt% LiClO4 and (b)PEO + 10 wt% LiClO4 + 10 wt% BMIMPF6 films crystallized at 50 �C atdifferent crystallization times (i) 10 s, (ii) 30 s, (iii) 60 s, (iv) 90 s and (v)120 s.

Fig. 14 Spherulite diameter vs. crystallization time plot for (a) PEO +10 wt% LiClO4 (b) PEO + 10 wt% LiClO4 + 10 wt% BMIMPF6 filmscrystallized at 50 �C at different crystallization times.

This journal is © The Royal Society of Chemistry 2015

nucleating sites increased. This is evident when we compareFig. 13 curves a (i–v) with b (i–v).

In the present case, we can also see that the size of oneparticular spherulite at any particular snap increased with time,as indicated in rectangle of Fig. 13. Therefore, it is important todetermine the spherulitic growth rate (and therefore crystalli-zation rate) for deciding the effect of BMIMPF6 on the crystal-lization behavior of polymer electrolytes with and without ionicliquid. The spherulite size vs. time plot for PEO + 10 wt% LiClO4

polymer electrolyte and with added 10 wt% BMIMPF6 is shownin Fig. 14, whose slope gives the spherulites growth rate (Gs).The value of Gs for pure PEO + 10 wt% LiClO4 is found to be 1.77mm s�1, which decreases to 0.675 mm s�1 for PEO + 10 wt%LiClO4 + 10 wt% BMIMPF6. The abovementioned observationclearly indicates that the incorporation of BMIMPF6 in PEO +10wt% LiClO4 polymer electrolyte hinders the spherulites growthrate of polymer PEO. This supports our previous resultsobtained from isothermal and non-isothermal DSC technique,which indicate the suppression in the crystallization rate ofpolymer PEO due to the incorporation of an ionic liquid(BMIMPF6) owing to its plasticization effect, as discussed earlierin this paper.

Conclusions

The crystallization kinetics behavior of PEO + 10 wt% LiClO4 + xwt% BMIMPF6 polymer electrolytes has been studied usingthree techniques, viz., isothermal and non-isothermal crystal-lization methods using DSC as well as by monitoring spherulitegrowth using a polarizing optical microscope. The well-knownAvrami equation has been used to describe the isothermalcrystallization process. The obtained crystallization parametersfor isothermal crystallization, such as Avrami exponent (n),crystallization rate constant (K), crystallization half time (t1/2),crystallization rate (G) and isothermal crystallization activation

RSC Adv., 2015, 5, 8263–8277 | 8275

RSC Advances Paper

energy (DE) suggested that the incorporation of the ionic liquidBMIMPF6 into PEO + 10 wt% LiClO4 polymer electrolyte hindersthe crystallization of polymer PEO due to the plasticizationeffect. The value of Avrami exponent ‘n’ obtained for theisothermal crystallization process of PEO + 10 wt% LiClO4 + xwt% BMIMPF6 (for x ¼ 0, 5 and 10) was found to be less than 2,indicating the 2D crystal growth geometry. It was also foundthat the non-isothermal crystallization data cannot be fullydescribed by the modied Avrami equation (Jeziorny method)and Ozawa analysis. The method proposed by Mo andcoworkers was employed to accurately analyze the non-isothermal crystallization process. The POM study shows thatthe incorporation of BMIMPF6 in PEO + 10 wt% LiClO4 polymerelectrolyte slows down the spherulite growth rate, as observedby spherulite size vs. time plots. All the three techniques used byus to study the effect of ionic liquid BMIMPF6 on the crystalli-zation behavior of polymer electrolytes have led to the conclu-sion that the presence of BMIMPF6 suppresses thecrystallization rate. This will have an effect in decreasing theabsolute crystallinity of PEO with increasing BMIMPF6 content.

Acknowledgements

One of us (RKS) is grateful to BRNS-DAE Mumbai and DST NewDelhi, India, for nancial assistance for carrying out this workand SKC is thankful to UGC New Delhi, India, for providing theDr D. S. Kothari Postdoctoral research fellowship.

References

1 J. R. MacCallum and C. A. Vincent, Polymer ElectrolyteReviews, Elsevier Applied Science, London, 1987, vol. 1 & 2.

2 W. H. Meyer, Adv. Mater., 1998, 10, 439.3 Y. J. Wang, J. Qiao, R. Baker and J. Zhang, Chem. Soc. Rev.,2013, 42, 5768.

4 J. N. de Freitas, A. F. Nogueira and M. A. De Paoli, J. Mater.Chem., 2009, 19, 5279.

5 F. M. Gray, Polymer Electrolytes: Fundamental andTechnological Applications, VCH Publishers, New York, 1997.

6 W. Wieczorek, D. Raducha, A. Zalewska and J. R. Stevens, J.Phys. Chem. B, 1998, 102, 8725.

7 A. Karmakar and A. Ghosh, J. Appl. Phys., 2010, 107, 104113.8 E. Quartarone and P. Mustarelli, Chem. Soc. Rev., 2011, 40,2525.

9 S. Seki, K. Takei, H. Miyashiro and M. Watanabe, J.Electrochem. Soc., 2011, 158, A769.

10 J. J. Xu, H. Ye and J. Huang, Electrochem. Commun., 2005, 7,1309.

11 M. D. Slater, D. Kim, E. Lee and C. S. Johnson, Adv. Funct.Mater., 2013, 23, 947.

12 S. Rajendran, M. Sivakumar and R. Subadevi, Mater. Lett.,2004, 58, 641.

13 G. B. Appetecchi, F. Crore and B. Scrosati, Electrochim. Acta,1995, 40, 991.

14 F. Croce, G. B. Appetecchi, L. Persi and B. Scrosati, Nature,1998, 394, 456.

8276 | RSC Adv., 2015, 5, 8263–8277

15 R. Haggenmueller, J. E. Fischer and K. I. Winey,Macromolecules, 2006, 39, 2964.

16 J. Zhou and P. S. Fedkiw, Solid State Ionics, 2004, 166, 275.17 D. K. Pradhan, R. N. P. Choudhary, B. K. Samantaray,

N. K. Karan and R. S. Katiyar, Int. J. Electrochem. Sci., 2007,2, 861.

18 M. Armand, F. Endres, D. R. MacFarlane, H. Ohno andB. Scrosati, Nat. Mater., 2009, 621.

19 Y. S. Ye, J. Rick and B. J. Hwang, J. Mater. Chem. A, 2013, 1,2719.

20 J. Lu, F. Yana and J. Texter, Prog. Polym. Sci., 2009, 34, 431.21 H. Ohno, Electrochemical Aspects of Ionic Liquids, John Wiley

& Sons, Inc., Hoboken, New Jersey, 2005.22 C. A. Angell, Y. Ansari and Z. Zhao, Faraday Discuss., 2012,

154, 9.23 M. P. Singh, R. K. Singh and S. Chandra, Prog. Mater. Sci.,

2014, 64, 73.24 Shalu, S. K. Chaurasia, R. K. Singh and S. Chandra, J. Phys.

Chem. B, 2013, 117, 897–906.25 I. Tim, J. H. Lee, M. L. Shofner, K. Jacob and

R. Tannenbaum, Polymer, 2013, 53, 2402.26 A. J. Muller, M. L. Arnal, A. L. Spinelli, E. Canizales,

C. C. Puig, H. Wang and C. C. Han, Macromol. Chem. Phys.,2003, 204, 1497.

27 K. L. Singeld, R. A. Chisholm and T. L. King, J. Chem. Educ.,2012, 89, 159.

28 Shalu, S. K. Chaurasia and R. K. Singh, RSC Adv., 2014, 4,50914.

29 K. E. Strawhecker and E. Manias, Chem. Mater., 2003, 15,844.

30 A. Keller, G. G. Wood and M. Hikosaka, Faraday Discuss.,1993, 95, 109.

31 Y. Zhang, H. Huo, J. Li, Y. Shang, Y. Chen, S. S. Funari andS. Jiang, CrystEngComm, 2012, 14, 7972.

32 S. Acierno, N. Grizzuti and H. H. Winter, Macromolecules,2002, 35, 5043.

33 J. Jin, M. Song and F. Pan, Thermochim. Acta, 2007, 456, 25.34 V. Sencadas, P. Martins, A. Pitaes, M. Benelmekki,

R. L. G. Ribelles and S. Lanceros-Mendez, Langmuir, 2011,27, 7241.

35 H. Wang, J. K. Keum, A. Hiltner and E. Baer,Macromolecules,2010, 43, 3359.

36 D. Zhou, X. Mei and J. Ouyang, J. Phys. Chem. C, 2011, 115,16688.

37 K. Chirssopoulou, K. S. Andrikopoulos, S. Bollas,C. Karageorgaki, D. Chirstolos, G. A. Voyiatzis andS. H. Anastasiadis, Macromolecules, 2011, 44, 9710.

38 M. J. Avrami, J. Chem. Phys., 1939, 7, 1103.39 S. K. Chaurasia, R. K. Singh and S. Chandra, J. Polym. Sci.,

Part B: Polym. Phys., 2011, 49, 291.40 M. P. Scott, C. S. Brazel, M. G. Benton, J. W. Mays,

J. D. Holbrey and R. D. Rogers, Chem. Commun., 2002, 1370.41 N. Winterton, J. Mater. Chem., 2006, 16, 4281.42 S. K. Chaurasia, R. K. Singh and S. Chandra, J. Raman

Spectrosc., 2011, 42, 2168.43 S. K. Chaurasia, R. K. Singh and S. Chandra, CrystEngComm,

2013, 15, 6022.

This journal is © The Royal Society of Chemistry 2015

Paper RSC Advances

44 S. K. Chaurasia, R. K. Singh and S. Chandra, Solid StateIonics, 2014, 262, 790.

45 J. B. Zeng, F. Wu, C. L. Huang, Y. S. He and Y. Z. Wang, ACSMacro Lett., 2012, 1, 965.

46 H. Zhao, Y. Bian, M. Xu, C. Han, Y. Li, Q. Dong and L. Dong,CrystEngComm, 2014, 16, 3896.

47 S. K. Chaurasia, R. K. Singh and S. Chandra, Solid StateIonics, 2011, 183, 32.

48 Shalu, S. K. Chaurasia, R. K. Singh and S. Chandra, J. Appl.Polym. Sci., 2014, DOI: 10.1002/app.41456.

49 A. L. Saroj and R. K. Singh, J. Phys. Chem. Solids, 2012, 73,162.

50 J. Maiz, J. Martin and C. Mijangos, Langmuir, 2012, 28,12296.

51 J. Li, C. Zhou, G. Wang, Y. Tao, Q. Liu and Y. Li, Polym. Test.,2002, 21, 583.

52 J. B. Zeng, M. Srivanasan, S. L. Li, R. Narayan and Y. Z. Wang,Ind. Eng. Chem. Res., 2011, 50, 4471.

53 A. Jeziorny, Polymer, 1978, 19, 1142.54 T. Ozawa, Polymer, 1971, 12, 150.55 T. Liu, Z. Mo, S. Wang and H. Zhang, Polym. Eng. Sci., 1997,

37, 568.

This journal is © The Royal Society of Chemistry 2015

56 L. Wang, G. Li, Y. Chen and S. Li, Polym. Eng. Sci., 2012, 52,1621.

57 K. Cao, Q. Wang, Y. C. Zhou, J. M. Sun and Z. Yao, Ind. Eng.Chem. Res., 2012, 51, 5461.

58 M. L. Di Lorenzo and C. Silvestre, Prog. Polym. Sci., 1999, 24,917.

59 A. S. Fischer, M. B. Khalid, M. Widstrom and P. Konas, J.Power Sources, 2011, 196, 9767.

60 S. K. Chaurasia and R. K. Singh, Phase Trans., 2010, 83, 457.61 T. Wei, S. Pang, N. Xu, L. Pan, Z. Zhang, R. Xu, N. Ma and

Q. Lin, J. Appl. Polym. Sci., 2014, DOI: 10.1002/app.41308.62 H. E. Kissinger, Anal. Chem., 1957, 29, 1702.63 Y. Zheng, M. L. Bruening and G. L. Baker, J. Polym. Sci., Part

B: Polym. Phys., 2010, 48, 1955.64 A. G. Shtukenberg, Y. O. Punin, E. Gunn and B. Kahr, Chem.

Rev., 2012, 112, 1805.65 J. Xi, X. Qiu, S. Zheng and X. Tang, Polymer, 2005, 46, 5702.66 B. K. Choi, Solid State Ionics, 2004, 168, 123.67 D. F. Miranda, T. P. Russell and J. J. Watkins,

Macromolecules, 2010, 43, 10528.68 Z. B. Zeng, F. Wu, C. L. Huang, Y. S. He and Y. Z. Wang, ACS

Macro Lett., 2012, 1, 965.

RSC Adv., 2015, 5, 8263–8277 | 8277