8
16074 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 This journal is c the Owner Societies 2013 Cite this: Phys. Chem. Chem. Phys., 2013, 15, 16074 Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths Boxuan Li, ab Meng Qiu, ab Saran Long, ab Xuefei Wang, ab Qianjin Guo* ab and Andong Xia* ab The rotational dynamics of coumarin 153 (C153) have been investigated in a series of 1-alkyl-3- methylimidazolium hexafluorophosphate ionic liquids with different alkyl chain lengths (alkyl = butyl, pentyl, hexyl, heptyl, octyl) ([C n mim][PF 6 ], n = 4–8) to examine the alkyl chain length dependent local viscosity of the microenvironment surrounding the probe molecules. The excimer-to-monomer fluores- cence emission intensity ratio (I E /I M ) of a well-known microviscosity probe, 1,3-bis(1-pyrenyl)propane (BPP), is also employed to study the microviscosity of [C n mim][PF 6 ] as a complementary measurement. The rotational dynamics of C153 show that at a certain length of the alkyl chain there are incompact and compact domains within [C n mim][PF 6 ], resulting in fast and slow components of C153 rotational dynamics. The microviscosities in different structural domains of [C n mim][PF 6 ] with different alkyl chain lengths are investigated by studying the fluorescence anisotropy decay of probe molecules. The obtained average rotation time constants show that with an increase in the length of the alkyl chain, the microviscosity of [C n mim][PF 6 ] is obviously increased first and then slightly decreased. The steady state fluorescence measurements with the microviscosity probe of BPP further prove that the micro- viscosity is not increased as much as expected when ionic liquids [C n mim][PF 6 ] have a relatively long alkyl chain. The different heterogeneous structures of [C n mim][PF 6 ] with different lengths of the alkyl chain are proposed to interpret the unusual microviscosity behaviors. 1. Introduction Room temperature ionic liquids (RTILs), which are a class of novel compounds mainly containing organic cations and inorganic/ organic anions, have received a great deal of attention because of their interesting physicochemical properties such as negligible vapor pressure, high thermal and chemical stability, a broad electrochemical window, and good solubility of various organic and inorganic compounds. 1–5 These attractive properties have made RTILs potential environmentally benign solvents for a large number of reactions, catalysis, separation, and electrochemical studies. 1–5 Since many specific properties of RTILs can be tuned by applying minor changes to their chemical architecture, it is of great importance to understand the structure–property correlation of ionic liquids, so that novel ionic liquids having suitable properties can be designed and developed for many new applications. 6 To understand the structure-dependent intermolecular interactions and dynamics in ionic liquids, many experimental and theoretical studies have been carried out by several groups such as the Samanta group, 7–13 the Maroncelli group, 14–17 the Sarkar group, 18,19 the Bhattacharyya group, 20–22 the Quitevis group, 23–25 the Ernsting group, 26–29 and so on. 30–41 These studies revealed that the nature of RTILs is more complicated than conventional solvents due to the presence of structural hetero- geneity of the microenvironment in RTILs. 9–11,20,22,31,34,36,37,39 Previous studies suggested that, in some imidazolium-based ionic liquids with short alkyl chains less than C 4 , hydrogen bonds probably resulted in their structural heterogeneity related to an ordered microscopic network, instead of random inter- molecular interaction. 35,36 When the alkyl chain is butyl or longer in [C n mim] + , two types of structural domains are possibly involved in the heterogeneous structure of ionic liquids. 30–34 Nonpolar domains arising from aggregation of the alkyl chains are formed mainly through the essentially weak van der Waals interactions, and polar domains arising from charge ordering of the anions and the imidazolium ring of the cations are formed through the strong electrostatic interactions. The nonpolar and polar domains are interconnected and permeated into a a The State Key Laboratory of Molecular Reaction Dynamics, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China b Beijing National Laboratory for Molecular Sciences (BNLMS), Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China. E-mail: [email protected], [email protected] Received 29th June 2013, Accepted 1st August 2013 DOI: 10.1039/c3cp52724g www.rsc.org/pccp PCCP PAPER Published on 01 August 2013. Downloaded by Northeastern University on 29/10/2014 07:34:28. View Article Online View Journal | View Issue

Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

  • Upload
    andong

  • View
    213

  • Download
    1

Embed Size (px)

Citation preview

Page 1: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

16074 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 This journal is c the Owner Societies 2013

Cite this: Phys. Chem.Chem.Phys.,2013,15, 16074

Unraveling heterogeneous microviscosities of the1-alkyl-3-methylimidazolium hexafluorophosphateionic liquids with different chain lengths

Boxuan Li,ab Meng Qiu,ab Saran Long,ab Xuefei Wang,ab Qianjin Guo*ab andAndong Xia*ab

The rotational dynamics of coumarin 153 (C153) have been investigated in a series of 1-alkyl-3-

methylimidazolium hexafluorophosphate ionic liquids with different alkyl chain lengths (alkyl = butyl,

pentyl, hexyl, heptyl, octyl) ([Cnmim][PF6], n = 4–8) to examine the alkyl chain length dependent local

viscosity of the microenvironment surrounding the probe molecules. The excimer-to-monomer fluores-

cence emission intensity ratio (IE/IM) of a well-known microviscosity probe, 1,3-bis(1-pyrenyl)propane

(BPP), is also employed to study the microviscosity of [Cnmim][PF6] as a complementary measurement.

The rotational dynamics of C153 show that at a certain length of the alkyl chain there are incompact

and compact domains within [Cnmim][PF6], resulting in fast and slow components of C153 rotational

dynamics. The microviscosities in different structural domains of [Cnmim][PF6] with different alkyl chain

lengths are investigated by studying the fluorescence anisotropy decay of probe molecules. The

obtained average rotation time constants show that with an increase in the length of the alkyl chain,

the microviscosity of [Cnmim][PF6] is obviously increased first and then slightly decreased. The steady

state fluorescence measurements with the microviscosity probe of BPP further prove that the micro-

viscosity is not increased as much as expected when ionic liquids [Cnmim][PF6] have a relatively long

alkyl chain. The different heterogeneous structures of [Cnmim][PF6] with different lengths of the alkyl

chain are proposed to interpret the unusual microviscosity behaviors.

1. Introduction

Room temperature ionic liquids (RTILs), which are a class of novelcompounds mainly containing organic cations and inorganic/organic anions, have received a great deal of attention because oftheir interesting physicochemical properties such as negligiblevapor pressure, high thermal and chemical stability, a broadelectrochemical window, and good solubility of various organicand inorganic compounds.1–5 These attractive properties have madeRTILs potential environmentally benign solvents for a large numberof reactions, catalysis, separation, and electrochemical studies.1–5

Since many specific properties of RTILs can be tuned by applyingminor changes to their chemical architecture, it is of greatimportance to understand the structure–property correlation ofionic liquids, so that novel ionic liquids having suitable propertiescan be designed and developed for many new applications.6

To understand the structure-dependent intermolecularinteractions and dynamics in ionic liquids, many experimentaland theoretical studies have been carried out by several groupssuch as the Samanta group,7–13 the Maroncelli group,14–17 theSarkar group,18,19 the Bhattacharyya group,20–22 the Quitevisgroup,23–25 the Ernsting group,26–29 and so on.30–41 These studiesrevealed that the nature of RTILs is more complicated thanconventional solvents due to the presence of structural hetero-geneity of the microenvironment in RTILs.9–11,20,22,31,34,36,37,39

Previous studies suggested that, in some imidazolium-basedionic liquids with short alkyl chains less than C4, hydrogenbonds probably resulted in their structural heterogeneity relatedto an ordered microscopic network, instead of random inter-molecular interaction.35,36 When the alkyl chain is butyl orlonger in [Cnmim]+, two types of structural domains are possiblyinvolved in the heterogeneous structure of ionic liquids.30–34

Nonpolar domains arising from aggregation of the alkyl chainsare formed mainly through the essentially weak van der Waalsinteractions, and polar domains arising from charge ordering ofthe anions and the imidazolium ring of the cations are formedthrough the strong electrostatic interactions. The nonpolarand polar domains are interconnected and permeated into a

a The State Key Laboratory of Molecular Reaction Dynamics, Institute of Chemistry,

Chinese Academy of Sciences, Beijing 100190, P. R. Chinab Beijing National Laboratory for Molecular Sciences (BNLMS),

Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China.

E-mail: [email protected], [email protected]

Received 29th June 2013,Accepted 1st August 2013

DOI: 10.1039/c3cp52724g

www.rsc.org/pccp

PCCP

PAPER

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article OnlineView Journal | View Issue

Page 2: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 16075

nanostructurally organized network. Some studies suggestedthat the heterogeneity of [Cnmim]+-based ionic liquids withintermediate alkyl chain lengths is accounted for the physicalelongation of cations and does not require the aggregation ofthe alkyl chains.40,41 The crystal structure would be formed inionic liquids with longer alkyl chains such as [C12mim][PF6]through imidazolium rings and anions separated by interdigitatedalkyl chains.38 Therefore, with different alkyl chain lengths in[Cnmim]+, the heterogeneous local structures of ionic liquids vary.

Many specific features have been observed due to theheterogeneous nature of RTILs,10–13,20,22,36,37,42–48 includingthe significant difference between their bulk viscosity andmicroviscosity,12,42 the special solvation and rotational dynamicsand other processes of the solute in RTILs,10,11,13,20,22,36,37 thedistinct results for the viscosity upon increasing the alkyl chainlength of [Cnmim]+,43,44 their ability to dissolve both polar andnonpolar compounds,37,45–48 and so forth. Therefore, a deepinsight into the microenvironmental property and the localstructure of RTILs with heterogeneity is a key factor to under-stand the properties of RTILs and extend their applications.Microviscosity is closely related to many aspects of RTILs,49,50

and our previous studies showed that the microviscosity withstructural heterogeneity in RTILs is investigated through steady-state fluorescence spectroscopy and time-resolved fluorescenceanisotropy spectroscopy by employing fluorescence probes.36,37

Recently, some studies focused on the investigation of thephysicochemical properties of the imidazolium-based RTILswith different lengths of alkyl chains on the cationic or anionicmoiety.18,19,51–53 These studies showed that with an increase inthe length of the alkyl chains, the rotational dynamics of someprobe molecules are not as slow as expected from the increase inthe bulk viscosity.19,52 This anomalous trend probably reflectsthe difference between the microenvironment of ionic liquids andconventional solvents, which could be caused by the structuralheterogeneity in RTILs. In addition, some other studies based onN-alkyl-N-methylmorpholinium ionic liquids with different alkylchain lengths have found that the heterogeneous nature of thesemorpholinium media is more prominent in the case of long chaincontaining ionic liquids with more organized domain structure.12,13

In the present work, we try to reveal the microviscosity of aseries of 1-alkyl-3-methylimidazolium hexafluorophosphateionic liquids (alkyl = butyl, pentyl, hexyl, heptyl, octyl)([Cnmim][PF6], n = 4–8) by studying the time-resolved fluores-cence anisotropy spectroscopy of coumarin 153 (C153) and thesteady-state fluorescence emission spectra of 1,3-bis(1-pyrenyl)-propane (BPP). The structures of [Cnmim][PF6] and fluorescentprobes used in the present study are shown in Scheme 1. C153is considered to be an ideal probe because the dynamics probedby C153 have been found to be primarily related to nonspecificsolute–solvent interactions in many RTILs systems.16,54–56 BPPis a particularly sensitive probe of microviscosity especially in theheterogeneous environment such as ionic liquid systems.46–48,57–60

As a result, different microviscosity changing regimes in 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids ([Cnmim]-[PF6]) with an increase in the length of the alkyl chainswere observed by probing the microviscosity through the two

complementary experimental methods involved in two neutralmolecular probes. Both the methods with the rotational dynamicsof C153 and the excimer-to-monomer fluorescence emission inten-sity ratio of a well-known microviscosity probe BPP show similarresults of microviscosity within the [Cnmim][PF6] ionic liquids.

2. Materials and experimental methods2.1. Materials

Ionic liquids [Cnmim][PF6] (n = 4–8) (purity >99%, water contento500 ppm) were purchased from Lanzhou Greenchem ILS,LICP, CAS, China, and stored under a dry nitrogen atmosphere.C153 and BPP were from Aldrich and Invitrogen, respectively,and used as received.

2.2. Steady-state spectroscopy

Ultraviolet/visible (UV/vis) absorption and fluorescence emissionspectra were recorded on a spectrophotometer (UV1601, Shimadzu,Japan) and a fluorescence spectrometer (F4500, Hitachi, Japan),respectively. All the measurements were carried out at roomtemperature.

2.3. Time-resolved fluorescence anisotropy measurements

The time-resolved fluorescence anisotropy measurements werecarried out using a time-correlated single-photon counting(TCSPC) spectrometer (F900, Edinburgh Instrument). The sampleswere excited at 442 nm using an 80 ps laser diode (PicoQuantPDL 808). The instrument response function (IRF) of the detectionsystem is 200 ps. The fluorescence anisotropy was measured byusing the single-channel method. An expression for time-resolvedanisotropic decay r(t) is given by

rðtÞ ¼IkðtÞ � GI?ðtÞIkðtÞ þ 2GI?ðtÞ

(1)

where IJ(t) and I>(t) are the fluorescence decays polarizedparallel and perpendicular to the polarization of the excitationlight, respectively, when the excitation polarizer is mountedvertically. The factor G is the ratio of the sensitivity of thedetection system for vertically and horizontally polarized lightwhen the excitation polarizer is mounted horizontally.61

Scheme 1 Molecular structures of the ionic liquids [Cnmim][PF6] (n = 4–8) andprobe molecules C153, BPP.

Paper PCCP

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online

Page 3: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

16076 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 This journal is c the Owner Societies 2013

3. Results and discussion3.1. Steady-state behavior of C153

The absorption and fluorescence spectra of C153 (Scheme 1) in1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids([Cnmim][PF6]) with the alkyl chain increased from butyl to octylare shown in Fig. 1, and the extracted data are listed in Table 1.In order to reduce the possible influences of neat ionic liquidson the absorption and fluorescence spectra,9 the absorptionspectra of C153 in ionic liquids are measured within neat ionicliquids as a reference, and the fluorescence emissions of C153 inionic liquids are extracted from the emission of neat ionic liquidalthough the emissions of neat ionic liquids are very low in allcases of [Cnmim][PF6]. For instance, the emission obtained fromthe neat ionic liquid [C7mim][PF6] and C153 in [C7mim][PF6] areshown in the inset of Fig. 1.

The absorption peak of C153 is almost unaffected (between424 and 426 nm) in the ionic liquids [Cnmim][PF6] (n = 4–8)with the increase in length of the alkyl chain. The emissionmaxima of C153 in ionic liquids [Cnmim][PF6] are graduallyblue-shifted from 529 to 517 nm with the alkyl chain increasedfrom butyl to octyl. With the gradual increase of the alkyl chainof [Cnmim][PF6], the fluorescence Stokes shifts (Dl = lEm � lAbs)of C153 decrease from 105 to 92 nm.

The cation of [Cnmim][PF6] (n = 4–8) is composed of a polarheadgroup and a nonpolar alkyl chain. The polar imidazolium ringof cations and anions could connect each other to form the polardomain, while the nonpolar alkyl chains may come close to formthe nonpolar domain.31–34 It is expected that when the alkyl chainon the [Cnmim]+ is more and more longer, the nonpolar domain inionic liquids could be gradually increased. From the fluorescenceStokes shifts (Dl) of C153 in ionic liquids [Cnmim][PF6] (Table 1), itis found that with the alkyl chain increased from butyl (n = 4) tooctyl (n = 8), the value of Stokes shifts (Dl) decreases from 105 to92 nm, indicating a more nonpolar environment of [Cnmim][PF6]upon increasing the alkyl chain length. The observed behaviorsof ionic liquids [Cnmim][PF6] with different alkyl chains are inagreement with a similar system of ionic liquid.18

3.2. Rotational dynamics of C153

The rotational dynamics of excited probe molecules are verysensitive to the microstructure and the local viscosity of themicroenvironment around the probe molecule.62 In our previousstudies, the microviscosities of neat RTILs and their mixtures withother organic solvents within different structural domains due totheir heterogeneous nature were successfully investigated throughobserving the rotational dynamics of the excited probe.36,37 Todetermine the microviscosities in ionic liquids [Cnmim][PF6] withdifferent lengths of alkyl chains, which possibly cause differentstructural microenvironments, we further perform the time-resolvedfluorescence anisotropic measurements of the excited C153molecules by using the TCSPC techniques. All the anisotropydecay profiles of C153 are well fitted with a biexponentialfunction by eqn (2).

rðtÞ ¼X2i¼1

aie�t=ti (2)

where ai is the amplitude and ti is the time constant. The averagerotational relaxation time htri was calculated using eqn (3). Threetypical anisotropy decay profiles are shown in Fig. 2, and all thecorresponding fitted parameters are listed in Table 2.

trh i ¼X2i¼1

aiti

,X2i¼1

ai (3)

As shown in Fig. 2 and Table 2, a fast decay time constant(t1) and a slow decay time constant (t2) are obtained by fitting

Fig. 1 Steady-state absorption (left) and emission (right) spectra of C153 in1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids ([Cnmim][PF6],n = 4–8). The emission spectra are obtained under the excitation (lexc) at430 nm. All spectra are normalized at the corresponding peak maxima. The insetshows the fluorescence intensity of neat ionic liquid [C7mim][PF6] and C153 in[C7mim][PF6].

Table 1 Steady-state absorption and fluorescence properties of C153 in ionicliquids [Cnmim][PF6]

lAbsa (nm) lEm

b (nm) Dlc (nm)

[C4mim][PF6] 424 529 105[C5mim][PF6] 425 526 101[C6mim][PF6] 426 523 97[C7mim][PF6] 425 521 96[C8mim][PF6] 425 517 92

a The absorption peak wavelength. b The fluorescence emission peakwavelength. c The Stokes shift.

Fig. 2 Three typical decays of the time-resolved fluorescence anisotropy, r(t), ofC153 in [C4mim][PF6], [C6mim][PF6] and [C8mim][PF6], respectively. The solid linerepresents the fitting result.

PCCP Paper

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online

Page 4: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 16077

the decays of the time-resolved anisotropy of C153 in ionicliquids [Cnmim][PF6] (n = 4–8). Since there is no any specialfrictional coupling between the C153 probe molecules andionic liquids,16,54–56 the measured rotational relaxation timeis mainly correlated with the local viscosity of the microenviron-ment around C153. It is expected that there are two types ofdomains with different microviscosities related to the fast andslow rotational dynamics experienced by C153 due to the hetero-geneous nature of ionic liquid.31,34,37 Our previous study andsome other studies showed that there is a polar domaincomposed by the imidazolium ring of cations and anions, anda nonpolar domain composed by the alkyl chains within theionic liquids [Cnmim]+.31–34,37 It is not surprising that because ofthe strong electrostatic interactions among cations and anions of[Cnmim][PF6] the structure of polar domains is more closelypacked than that of nonpolar domains in which the interactionis from van der Waals force between nonpolar alkyl chainswithin [Cnmim][PF6]. Since C153 is a neutral molecule and thereis no specific solute–solvent interactions between C153 and ionicliquids, C153 could homogeneously disperse into incompactnonpolar and compact polar domains of [Cnmim][PF6] (n =4–8). The two different time constants observed from the rota-tional dynamics of C153 could correspond to such kinds of twotypical microenvironments. The fast component of C153 rota-tional dynamics is attributed to the incompact nonpolardomain, in which the microenvironment is built up by the alkylchains of [Cnmim][PF6] in a relatively loose microstructure. Theslow component of C153 rotational dynamics is accordinglycontributed from the compact polar domain with more closelypacked microstructure connected by imidazolium ring cationsand anions. The microviscosity around C153 in ionic liquidmainly affects the rotation behaviors and leads to the differentrotational relaxation times. The microviscosity in the incompactnonpolar domain is relatively low and the molecular probe C153is easier to rotate, while the microviscosity in the compact polardomain is relatively high and the rotation of C153 is moreconfined and becomes slow. Therefore, it is important to inves-tigate the microviscosities in different structural domains ofionic liquids [Cnmim][PF6] (n = 4–8) due to their heterogeneousnature by obtaining different rotation time constants from thefitted anisotropy decay of probe molecules and further estimatethe average rotational relaxation time for getting a generalinsight into the microviscosity.

Fig. 3 shows the fast component of C153 rotation timecorrelated with the length of the alkyl chain (n) in ionic liquids[Cnmim][PF6]. It is found that the value of the fast rotation timeconstant (t1) does not always monotonically change with anincrease in the length of alkyl chains of ionic liquids[Cnmim][PF6], which implies that the microviscosity of theincompact domains in ionic liquid is possibly not only influencedby the interactions among alkyl chains (with an increase in thelength of alkyl chains the van der Waals interactions are alwaysincreased), but also controlled by the different structural hetero-geneities of ionic liquids [Cnmim][PF6] in long alkyl chains. As aresult, the value of the fast rotation time constant (t1) in ionicliquids [Cnmim][PF6] significantly increases with the alkyl chainlength (n) between 4 and 6, but increases slowly with alkyl chainlength (n) between 6 and 7, and then turns to decrease when thevalue of n is 8. Since the fast decay of C153 rotational dynamics iscorrelated with the microviscosity of the incompact domainswithin ionic liquids [Cnmim][PF6], it is reasonable to concludethat the relatively loose microenvironment of the incompactdomains with low viscosity in ionic liquid with the length of thealkyl chain n = 4–5 becomes more denser when n = 6–7 withincreased microviscosity. With a further increase in the length ofthe alkyl chain up to n = 8, the viscosity turns to be less viscousbecause of the pronounced structural heterogeneities in longeralkyl chain ionic liquid.

Regarding the value of the slow time constant (t2) obtainedfrom the C153 rotational dynamics measurements, it is foundthat t2 changes little in all cases of ionic liquids [Cnmim][PF6]with different lengths of alkyl chains, as shown in Fig. 4. As theslow decay of C153 rotational dynamics is related to themicroviscosity of the compact domain of [Cnmim][PF6], it ispossible that the microviscosity and structure of the compactdomain are almost unchanged with an increase in the length ofalkyl chains of [Cnmim][PF6]. The compact domain is built by theimidazolium ring cations and anions of [Cnmim][PF6] throughthe strong electrostatic interactions, which could be possiblyalmost impervious when the incompact domain is changed withthe increase in the alkyl chain length through relatively weak vander Waals interactions. The value of the slow time constant (t2)of C153 rotational dynamics slightly changes in the range from4.90 to 5.29 ns in all cases of ionic liquids [Cnmim][PF6].

Table 2 Rotational relaxation parameters for C153 in ionic liquids [Cnmim][PF6]

Fitting results

A1 t1 (ns) A2 t2 (ns) tra (ns)

[C4mim][PF6] 0.49 0.32 0.51 4.90 2.66[C5mim][PF6] 0.44 0.39 0.56 4.96 2.95[C6mim][PF6] 0.44 0.47 0.56 5.29 3.17[C7mim][PF6] 0.45 0.48 0.55 5.26 3.11[C8mim][PF6] 0.49 0.44 0.51 5.10 2.82

a The average rotation time from biexponential fitting using eqn (3),

Ai ¼ ai

�P2i¼1

ai,P2i¼1

Ai ¼ 1.

Fig. 3 Fast component of C153 rotation time correlated with the length of alkylchains (n) in ionic liquids [Cnmim][PF6].

Paper PCCP

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online

Page 5: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

16078 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 This journal is c the Owner Societies 2013

With different alkyl chains on the [Cnmim]+, the structuralheterogeneity of ionic liquids could be changed due to the differentpacking and stacking effects of cations and anions.30–36,38 Whenthe alkyl chain is shorter than butyl (n o 4), the cations and anionsof ionic liquids could form an ordered compact network throughthe strong electrostatic interactions among ions.35,36 As the non-polar alkyl chains have a certain length, they can change their chainconformation to arrange themselves close to each other throughthe van der Waals interactions and form the nonpolar domain inionic liquids [Cnmim][PF6] (n Z 4), where in the compact region,most of the electrostatic charges are concentrated, the polar head-group imidazolium ring cations and anions connect each other toform the polar domain.30–34 The van der Waals interactions of thenonpolar alkyl chains and the electrostatic interactions betweenions are all involved in the main interactions within ionicliquids. When the alkyl chain is longer enough such as[C12mim][PF6] (n = 12), the alkyl chains can come close to eachother and interdigitate to form an ordered conformation. Theimidazolium rings and anions are separated by interdigitatedalkyl chains, and all the cations and anions stack each other to forman ordered crystal structure by the van der Waals interactions ofalkyl chains and electrostatic interactions of ions.38 Recently, somestudies suggested that in some ionic liquids with intermediate alkylchain length (4 r n r 10) based on [Cnmim]+, some morphologies

are formed through the long alkyl tails to the ordered crystalstructure,40,41 whereas many research groups showed that the alkylchains with intermediate length on [Cnmim]+ could aggregate toform the nonpolar domain without the obvious ordered structure.39

From our investigation of the microviscosities of ionic liquids[Cnmim][PF6] with intermediate alkyl chain length determined byC153, it is found that with an increase in the alkyl chain length frombutyl (n = 4) to octyl (n = 8), the microviscosity of the nonpolardomain is increased first and then decreased (Fig. 3), and themicroviscosity of the polar domain is almost unchanged (Fig. 4),which is possibly caused by the different heterogeneous structurescomposed of the different lengths of nonpolar alkyl chains andpolar imidazolium rings and anions. In this case, a reasonableschematic representation is drawn as shown in Fig. 5, which couldbe helpful to visualize the microstructure in ionic liquids[Cnmim][PF6] with different lengths of alkyl chains.

For a certain length of the alkyl chain on the imidazoliumring, e.g. like [C4mim][PF6], the van der Waals interactions ofaggregated alkyl chains in nonpolar domains could be the mostimportant aspect of the main interactions within the ionicliquid in addition to the interactions of polar domains. Theheterogeneous structure of ionic liquid [C4mim][PF6] iscomposed of both an incompact nonpolar domain and acompact polar domain as shown in Fig. 5(a). With an increasein the length of the alkyl chain from butyl (n = 4) to heptyl(n = 7), the incompact nonpolar domain in ionic liquids[Cnmim][PF6] gradually becomes more and more dense by thepacking and stacking effects of the longer alkyl chains, whichcauses increased microviscosity of incompact domains. Thiseffect is more pronounced when the alkyl chain increases frombutyl to hexyl (n = 6) (Fig. 3). The heterogeneous structure ofionic liquid [C6mim][PF6] as shown in Fig. 5(b) has almost thesame pattern as [C4mim][PF6] except for the longer alkyl chainaggregation. With the alkyl chain length on the imidazoliumring being increased to octyl (n = 8) as shown in Fig. 5(c), alkylchains have a larger capability to change their conformation,they could come close to each other in an interdigitated way orcoil themselves individually in the nonpolar domain of ionicliquid [C8mim][PF6]. Although the octyl chain is probably not

Fig. 4 Slow anisotropy time constants of C153 in ionic liquids [Cnmim][PF6] withdifferent lengths of alkyl chains.

Fig. 5 Schematic representation of the C153 probe molecules in structural heterogeneous ionic liquids [Cnmim][PF6] with different alkyl chain lengths,(a) [C4mim][PF6], (b) [C6mim][PF6], and (c) [C8mim][PF6].

PCCP Paper

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online

Page 6: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 16079

long enough to form the typical crystal structure like[C12mim][PF6], some local ordered structure can possibly beformed by the interdigitated intermediate length of the octylchain. The structure of the [C8mim][PF6] polar domaincomposed of the imidazolium rings and anions is almost thesame as in other ionic liquids [Cnmim][PF6] (4 r n r 7) (Fig. 5),but different from the heterogeneous structure of ionic liquids[Cnmim][PF6] (4 r n r 7), the nonpolar domain of[C8mim][PF6] is composed of local ordered and unordered alkylchains, as shown in Fig. 5(c). With some local ordered alkylchains appearing in [C8mim][PF6], the amounts of unorderedalkyl chains in nonpolar domain become less, and alkyl chainsmay coil themselves with an increase in their length to a certaindegree (Fig. 5(c)). This possibly makes the nonpolar domain of[C8mim][PF6] not as compact as expected and the microviscositythen becomes lower as shown in Fig. 3. In all cases of compactpolar domains of ionic liquids [Cnmim][PF6] (4 r n r 8), theCoulombic interactions between imidazolium rings and [PF6]�

anions dominate, and the structure of the polar domain showsno significant changes (Fig. 4 and 5).

To further understand the microviscosity changes in ionicliquids [Cnmim][PF6] with different alkyl chain lengths, theaverage rotational relaxation time (taverage) of C153 is furtherconsidered. Fig. 6 shows the average rotation time of C153 inionic liquids [Cnmim][PF6] (4 r n r 8). It is found that theaverage rotation time of C153 in the ionic liquids increasesfrom butyl to hexyl, and reaches a maximum from hexyl toheptyl, then turns to decrease when the alkyl chain is octyl, asshown in Fig. 6. Especially, the obtained average rotationalrelaxation time is 2.82 ns with the octyl chain in [Cnmim][PF6],which is unexpectedly faster than that in ionic liquids[Cnmim][PF6] (5 r n r 7), indicating some different localstructures emerged in ionic liquid [C8mim][PF6] by the longeralkyl chain interactions, leading to the decreased microviscositywhich makes the C153 probe molecules easier to rotate. Asshown in Fig. 6, with an increase in the length of alkyl chains,the structural heterogeneities of ionic liquids [Cnmim][PF6]could be changed (Fig. 5) and make the microviscosity increasedfirst and then unexpectedly decreased. It is possible that whenthe ionic liquid has a relatively long alkyl chain (n = 8), the octylchains form a local ordered structure in [C8mim][PF6] and coil

themselves (Fig. 5(c)), which makes some loose structure in[C8mim][PF6] somewhere with a lower overall microviscosity.

3.3. Fluorescence emission behavior of microviscosityprobe BPP

We have measured the microviscosity of ionic liquids[Cnmim][PF6] (4 r n r 8) by measuring the rotationaldynamics of C153, and in order to further confirm the micro-viscosity properties related to different heterogeneities of ionicliquids, as a complementary experiment, we perform the steady-state fluorescence emission experiments of a microviscosity probeBPP (1,3-bis(1-pyrenyl) propane), which is specifically sensitive inthe heterogeneous system.46–48,57–60 BPP (Scheme 1) is a well-established fluorescent probe, which can be employed to detectthe microviscosity by measuring the excimer emission bandlocated in about the 450–500 nm region in fluorescence emissionspectra. In a low-viscous solvent, the two pyrene units of BPPeasily fold together to form an intramolecular excimer, and theemission spectra of these compounds exhibit a usual structuredmonomer fluorescence band and a broad structureless excimerfluorescence band. As the microviscosity of the environmentincreases, the efficiency of the excimer formation decreases anda corresponding reduction in the intensity of the excimer bandis observed. From such a scenario, the excimer to monomeremission intensity ratio (IE/IM) of BPP can be used to measurethe microviscosity.

Fig. 7 shows the steady-state emission spectra of BPPdissolved in ionic liquids [Cnmim][PF6] (n = 4–6 and 8) andethanol. The corrected BPP steady-state emission spectra inionic liquids are subtracted from the neat ionic liquids fluores-cence background. It is found that unlike in ethanol, theexcimer peak of BPP fluorescence in ionic liquids is not obviousbut it can still be used to investigate the microviscosity changesas shown in the inset of Fig. 7. With the alkyl chain of ionicliquids [Cnmim][PF6] increased from butyl (n = 4) to octyl (n = 8),the relative intensity of excimer emission of BPP is slightlydecreased (inset of Fig. 7). In order to further determine themicroviscosity of ionic liquids [Cnmim][PF6], we investigate theexcimer-to-monomer emission intensity ratio (IE/IM) of BPP

Fig. 6 The average rotation time of C153 in ionic liquids [Cnmim][PF6] withdifferent alkyl chain lengths.

Fig. 7 Steady-state fluorescence emission spectra normalized to the peak of themonomer fluorescence band (IM(380nm) = 1.00) for BPP in ionic liquids[Cnmim][PF6] (n = 4–6 and 8) and ethanol; excitation wavelength: 330 nm. Theinset shows the excimer fluorescence band (IE(490nm)).

Paper PCCP

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online

Page 7: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

16080 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 This journal is c the Owner Societies 2013

under ambient conditions in ionic liquids [Cnmim][PF6] asshown in Fig. 8. The value of the excimer-to-monomer emissionintensity ratio (IE/IM) of BPP obviously decreases when the alkylchain of ionic liquids [Cnmim][PF6] increases from butyl (n = 4)to hexyl (n = 6) and slightly decreases when the alkyl chainincreases from hexyl (n = 6) to octyl (n = 8).

Since the value of BPP IE/IM is correlated with the micro-viscosity, it is found that the microviscosity of ionic liquids[Cnmim][PF6] is obviously increased when the alkyl chain isrelatively short (4 r n r 6) and slightly changed when the alkylchain is relatively long (6 r n r 8), as shown in Fig. 8. Such amicroviscosity changing effect with a relatively long alkyl chaindoes not increase as much as expected, possibly resulting fromthe different local structures of ionic liquids [Cnmim][PF6] withheterogeneous nature. When the alkyl chain of ionic liquids[Cnmim][PF6] is longer than hexyl (n = 6), the heterogeneousstructure could be probably changed by the relatively long alkylchain. Similar behavior is also seen from the average rotationalrelaxation time (taverage) of C153, as shown in Fig. 6, althoughwith a smaller molecular volume of C153 compared to BPP, themicroviscosity changing effect is more obviously detected by asmall-size C153 probe. With an increase in the alkyl chain ofionic liquids [Cnmim][PF6], the molecular interactions withincompact polar domains are almost unchanged, and in contrast,the van der Waals interactions of incompact nonpolar domainsare obviously increased, leading to an increase in bulk viscosityof [Cnmim][PF6].43,44,53 The microviscosity determined by probemolecules in our measurements is not only influenced by themolecular interactions but also sensitively depended on thelocal structure of ionic liquids [Cnmim][PF6]. From the resultsof measured microviscosity of ionic liquids [Cnmim][PF6] byIE/IM of BPP (Fig. 8) and average rotational relaxation time(taverage) of C153 rotational dynamics (Fig. 6), it is concludedthat the microviscosity is obviously increased when the lengthof alkyl chains of ionic liquids [Cnmim][PF6] is increased from4 to 6, and not increased as much as expected when the lengthof alkyl chains of ionic liquids [Cnmim][PF6] further increasesup to 8, indicating the different local structures formed by thepacking and stacking effects of different lengths of alkyl chainsof ionic liquids [Cnmim][PF6].

4. Conclusions

We have performed the measurements of the rotationaldynamics of C153 molecules in ionic liquids [Cnmim][PF6](4 r n r 8) to determine the local viscosity of the microenviron-ment surrounding the probe molecules. The microviscosity of[Cnmim][PF6] with different alkyl chain lengths is also investi-gated by a microviscosity probe BPP through steady-state fluores-cence spectroscopy. Due to the heterogeneous nature of ionicliquids, it is found that the C153 probe is experiencing two typesof domains with incompact and compact structures in[Cnmim][PF6]. With an increase in the length of alkyl chains,the microviscosity of the incompact domain in [Cnmim][PF6] isobviously increased first and then slightly decreased, while themicroviscosity of the compact domain is almost unchanged.With relatively long alkyl chains of [Cnmim][PF6], both the factsthat the excited C153 molecules show relatively fast rotationaldynamics determined by the average rotation time and theexcimer-to-monomer fluorescence emission intensity ratio(IE/IM) of BPP is higher than expected strongly suggest thatboth of the two probe molecules are facing a more incompactmicroenvironment.

Acknowledgements

This work was supported by NSFCs (21173235, 91233107,21127003), 973 Program (2013CB834604) and Chinese Academyof Sciences.

References

1 K. R. Seddon, Nat. Mater., 2003, 2, 363–365.2 T. Welton, Chem. Rev., 1999, 99, 2071–2083.3 J. Dupont, R. F. de Souza and P. A. Z. Suarez, Chem. Rev.,

2002, 102, 3667–3691.4 J. G. Huddleston, H. D. Willauer, R. P. Swatloski, A. E. Visser

and R. D. Rogers, Chem. Commun., 1998, 1765–1766.5 E. Dickinson, M. E. Williams, S. M. Hendrickson, H. Masui

and R. W. Murray, J. Am. Chem. Soc., 1999, 121, 613–616.6 M. Freemantle, Chem. Eng. News, 1998, 76, 32–37.7 R. Karmakar and A. Samanta, J. Phys. Chem. A, 2002, 106,

6670–6675.8 R. Karmakar and A. Samanta, J. Phys. Chem. A, 2002, 106,

4447–4452.9 A. Paul, P. K. Mandal and A. Samanta, J. Phys. Chem. B, 2005,

109, 9148–9153.10 A. Samanta, J. Phys. Chem. B, 2006, 110, 13704–13716.11 A. Samanta, J. Phys. Chem. Lett., 2010, 1, 1557–1562.12 D. C. Khara and A. Samanta, J. Phys. Chem. B, 2012, 116,

13430–13438.13 D. C. Khara, J. P. Kumar, N. Mondal and A. Samanta,

J. Phys. Chem. B, 2013, 117, 5156–5164.14 J. A. Ingram, R. S. Moog, N. Ito, R. Biswas and

M. Maroncelli, J. Phys. Chem. B, 2003, 107, 5926–5932.15 S. Arzhantsev, H. Jin, G. A. Baker and M. Maroncelli,

J. Phys. Chem. B, 2007, 111, 4978–4989.

Fig. 8 The excimer-to-monomer emission intensity ratio (IE/IM) of BPP in ionicliquids [Cnmim][PF6] with different alkyl chain lengths. All fluorescence emissionspectra of BPP in the mixtures were averaged from three time measurements.

PCCP Paper

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online

Page 8: Unraveling heterogeneous microviscosities of the 1-alkyl-3-methylimidazolium hexafluorophosphate ionic liquids with different chain lengths

This journal is c the Owner Societies 2013 Phys. Chem. Chem. Phys., 2013, 15, 16074--16081 16081

16 H. Jin, G. A. Baker, S. Arzhantsev, J. Dong andM. Maroncelli, J. Phys. Chem. B, 2007, 111, 7291–7302.

17 M. Maroncelli, X. X. Zhang, M. Liang, D. Roy andN. P. Ernsting, Faraday Discuss., 2012, 154, 409–424.

18 S. K. Das and M. Sarkar, ChemPhysChem, 2012, 13, 2761–2768.19 S. K. Das and M. Sarkar, J. Phys. Chem. B, 2012, 116,

194–202.20 A. Adhikari, K. Sahu, S. Dey, S. Ghosh, U. Mandal and

K. Bhattacharyya, J. Phys. Chem. B, 2007, 111, 12809–12816.21 K. Bhattacharyya, J. Phys. Chem. Lett., 2010, 1, 3254–3255.22 D. K. Sasmal, A. K. Mandal, T. Mondal and

K. Bhattacharyya, J. Phys. Chem. B, 2011, 115, 7781–7787.23 B. R. Hyun, S. V. Dzyuba, R. A. Bartsch and E. L. Quitevis,

J. Phys. Chem. A, 2002, 106, 7579–7585.24 D. Xiao, J. R. Rajian, A. Cady, S. Li, R. A. Bartsch and

E. L. Quitevis, J. Phys. Chem. B, 2007, 111, 4669–4677.25 D. Xiao, L. G. Hines, Jr., R. A. Bartsch and E. L. Quitevis,

J. Phys. Chem. B, 2009, 113, 4544–4548.26 M. Maroncelli, X.-X. Zhang, M. Liang, D. Roy and

N. P. Ernsting, Faraday Discuss., 2012, 154, 409–424.27 X. X. Zhang, C. Schroeder and N. P. Ernsting, J. Chem. Phys.,

2013, 138, 111102.28 X.-X. Zhang, M. Liang, N. P. Ernsting and M. Maroncelli,

J. Phys. Chem. B, 2013, 117, 4291–4304.29 X.-X. Zhang, M. Liang, N. P. Ernsting and M. Maroncelli,

J. Phys. Chem. Lett., 2013, 4, 1205–1210.30 S. M. Urahata and M. C. C. Ribeiro, J. Chem. Phys., 2004, 120,

1855–1863.31 Y. T. Wang and G. A. Voth, J. Am. Chem. Soc., 2005, 127,

12192–12193.32 J. Lopes and A. A. H. Padua, J. Phys. Chem. B, 2006, 110,

3330–3335.33 J. N. C. Lopes, M. F. C. Gomes and A. A. H. Padua, J. Phys.

Chem. B, 2006, 110, 16816–16818.34 Y. T. Wang and G. A. Voth, J. Phys. Chem. B, 2006, 110,

18601–18608.35 K. Dong, S. J. Zhang, D. X. Wang and X. Q. Yao, J. Phys.

Chem. A, 2006, 110, 9775–9782.36 X. N. Ma, L. Y. Yan, X. F. Wang, Q. J. Guo and A. D. Xia,

J. Phys. Chem. A, 2011, 115, 7937–7947.37 B. X. Li, Y. Y. Wang, X. F. Wang, S. Vdovic, Q. J. Guo and

A. D. Xia, J. Phys. Chem. B, 2012, 116, 13272–13281.38 C. M. Gordon, J. D. Holbrey, A. R. Kennedy and

K. R. Seddon, J. Mater. Chem., 1998, 8, 2627–2636.39 O. Russina, A. Triolo, L. Gontrani and R. Caminiti, J. Phys.

Chem. Lett., 2012, 3, 27–33.

40 H. V. R. Annapureddy, H. K. Kashyap, P. M. De Biase andC. J. Margulis, J. Phys. Chem. B, 2010, 114, 16838–16846.

41 C. Hardacre, J. D. Holbrey, C. L. Mullan, T. G. A. Youngs andD. T. Bowron, J. Chem. Phys., 2010, 133, 074510.

42 A. Paul and A. Samanta, J. Phys. Chem. B, 2008, 112,16626–16632.

43 H. Tokuda, K. Hayamizu, K. Ishii, M. Susan andM. Watanabe, J. Phys. Chem. B, 2005, 109, 6103–6110.

44 A. A. H. Padua, M. F. Gomes and J. Lopes, Acc. Chem. Res.,2007, 40, 1087–1096.

45 A. Paul and A. Samanta, J. Phys. Chem. B, 2008, 112,947–953.

46 A. Sarkar, S. Trivedi, G. A. Baker and S. Pandey, J. Phys.Chem. B, 2008, 112, 14927–14936.

47 A. Sarkar, S. Trivedi and S. Pandey, J. Phys. Chem. B, 2009,113, 7606–7614.

48 S. Trivedi and S. Pandey, J. Phys. Chem. B, 2011, 115,7405–7416.

49 B. Bhattacharya and A. Samanta, J. Phys. Chem. B, 2008, 112,10101–10106.

50 S. Tiwari, N. Khupse and A. Kumar, J. Org. Chem., 2008, 73,9075–9083.

51 K. Fruchey and M. D. Fayer, J. Phys. Chem. B, 2010, 114,2840–2845.

52 G. B. Dutt, J. Phys. Chem. B, 2010, 114, 8971–8977.53 A. B. Pereiro, J. L. Legido and A. Rodriguez, J. Chem.

Thermodyn., 2007, 39, 1168–1175.54 D. Chakrabarty, A. Chakraborty, D. Seth, P. Hazra and

N. Sarkar, Chem. Phys. Lett., 2004, 397, 469–474.55 N. Ito, S. Arzhantsev and M. Maroncelli, Chem. Phys. Lett.,

2004, 396, 83–91.56 D. Chakrabarty, A. Chakraborty, D. Seth and N. Sarkar,

J. Phys. Chem. A, 2005, 109, 1764–1769.57 G. A. Baker, S. N. Baker and T. M. McCleskey, Chem.

Commun., 2003, 2932–2933.58 K. A. Fletcher and S. Pandey, J. Phys. Chem. B, 2003, 107,

13532–13539.59 R. Karmakar and A. Samanta, Chem. Phys. Lett., 2003, 376,

638–645.60 S. Pandey, K. A. Fletcher, S. N. Baker and G. A. Baker,

Analyst, 2004, 129, 569–573.61 J. R. Lakowicz, Principles of Fluorescence Spectroscopy,

Springer, New York, 3rd edn, 2006, ch. 4, vol. 10,pp. 361–363.

62 T. A. Smith, L. M. Bajada and D. E. Dunstan, Macromolecules,2002, 35, 2736–2742.

Paper PCCP

Publ

ishe

d on

01

Aug

ust 2

013.

Dow

nloa

ded

by N

orth

east

ern

Uni

vers

ity o

n 29

/10/

2014

07:

34:2

8.

View Article Online