- Email: [email protected]
15th anniversary of polymerised ionic liquids
Polymer xxx (2014) 1e9
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
15th anniversary of polymerised ionic liquids Naomi Nishimura a, b, Hiroyuki Ohno a, b, * a b
Department of Biotechnology, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan Functional Ionic Liquid Laboratories, Graduate School of Technology, Tokyo University of Agriculture and Technology, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 January 2014 Received in revised form 10 February 2014 Accepted 11 February 2014 Available online xxx
Polymerised ionic liquids (PILs) have unique properties such as low glass transition temperature (Tg) in spite of very high charge density. Due to these advanced points, PILs have been prepared and initially evaluated as ion conductive polymers. Progress of low-Tg polyelectrolytes has been previously demonstrated with polyethers having charged end(s) as a kind of PILs. Then, imidazolium-type ionic liquids (ILs) were polymerised after introducing vinyl groups onto the imidazolium cation rings. It is reasonable that the ionic conductivity of thus prepared PILs decreased due to elevation of Tg and decrease of the number of mobile small ions. Efforts were then paid to suppress drop of ionic conductivity after polymerisation. Variety of PILs has been improved to show excellent ionic conductivity, selective ion transport, and other properties. With the progress of functional ILs, some functions were also added to PILs which cannot be realised with ordinary charged polymers. In the present mini-review, we briefly introduce history of a variety of polymerised ILs and some applications of these PILs. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Polymerisation Glass transition temperature Polyelectrolytes
1. Introduction Ionic liquids (ILs) are liquids composed of only both cations and anions at temperature below 100 C. Just 100 years ago, Paul Walden has reported that melting point of ethylammonium nitrate (CH3CH2NH3NO3) is 12 C [1]. This should be the first report on the ILs. On the other hand, aluminium chloride type molten salts have also a long history in this research field. After a few types of researches, air and water stable ILs were reported in 1992 [2]. At this stage there was however not many responses on this memorial paper. After this, there are explosively increasing papers on ILs after 2000 because of their interesting properties and charm of the possibility of a variety of designs on ions. We have reported polymerised ionic liquids (PILs) in 1998 for the first time in the world [3]. Our approach had not been accepted by general scientists due to inverse approach to the development of the lowering of melting point of salts. Although all of polyelectrolytes were brittle and stiff materials to deal with because of their high charge density, PILs were obtained as flexible films with low glass transition temperature (Tg). These advanced properties were gradually accepted by polymer scientists after more than five years of publication. After that, number of papers on ILs increased year by year and now more
* Corresponding author. Department of Biotechnology, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan. E-mail address: [email protected] (H. Ohno).
than 1000 papers on PILs have been reported. These materials have different attractive points from those of ILs, and there are many attempts to seek the possibilities to apply PILs to many different research fields. In this mini-review, we desire to summarise research results on PIL mainly obtained in our laboratory. Since polymers prepared from ionic liquids are widely called as polymerised ionic liquids (PILs), we also use this abbreviation in this mini-review. In the initial stage, PILs were prepared by the direct polymerisation of IL monomers. PILs were mainly prepared from imidazolium-cation type ILs. By the progress of researches on ILs, many kinds of polymerisable cations and anions have been synthesised and used to prepare PILs. It should be important to choose cations or anions to prepare PILs for particular purpose. Recently, there are increasing researches to prepare PILs with controlled molecular weight and sequences by precise polymerisation. Accordingly, there are many studies on the different PILs having novel structure and functions to date. Those PILs have been contributed to spread the possibility to use these PILs in a variety of scientific fields. In order to add functions onto ILs, it is a common way to introduce functional groups onto component ions of the ILs. Since it is also possible to put functions onto ILs by the structural control, we should care about the dimension of the ILs. To lower the dimension of the three-dimensionally isotropic ILs, there should be additional advancement reflecting the lowered dimension. Dimension control of the ILs is another very interesting subject, and
http://dx.doi.org/10.1016/j.polymer.2014.02.042 0032-3861/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
2
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
there are two major way to lower the dimension of the ILs; the use of liquid crystalline phase and polymerisation. It is of great interesting approach to use the liquid crystalline characteristics for the functionalisation of ILs [4,5]. For examples, anisotropic ionconductive materials were designed by lowering dimension of isotropic ILs. Although we have reported more than 30 papers on this subject, we do not want to discuss this subject here because of limited pages. Among these, there are a few papers on the polymerisation of ILs in the liquid crystalline state [6e8]. The use of liquid crystalline properties onto the functional design of ILs will be reviewed in the near future. The present review treats mainly the results on the polymerised ILs. 2. PEO/salt hybrids Before introducing IL-based PILs, we have to mention about the pre-study on the low Tg polyelectrolytes. We have been studying ion conductive polymers for over thirty years. Since polyether structure especially poly(ethylene oxide) (PEO) or poly(propylene oxide) (PPO) have very low Tg, ion conduction path have been made widely with them. In the late 20 century, it was a normal approach to mix PEOs with inorganic salts to design polymer electrolytes. Sufficiently long PEO chains and salts were mainly used to prepare solid polymer electrolytes. Thus prepared films are effective to transport ions due to low Tg of the PEO chains. We have recognised that low molecular weight PEO chains, so-called oligo(ethylene oxide)s, were liquid state at room temperature. Then, we tried to put charge(s) onto the end(s) of oligo(ethylene oxide) to prepare liquid salts. According to this idea, we have started the investigation on PEO/salt hybrid in 1993. Fig. 1 shows that the state of PEO/salt hybrid depending on the PEO molecular weight. The hybrids of salts and low molecular weight PEOs are solid because properties of salt are dominant factor for the hybrids. On the other hand, high molecular weight PEOs gave the hybrids as wax like solids. PEOs with molecular weight from 350 to 1000 are found to be effective to design liquid state hybrids. The hybrids composed of PEOs with moderate molecular weight are considered to be a kind of ionic liquid. PEO/salt hybrids which have hitherto been synthesised are shown in Fig. 2. The most important characteristic of these hybrids is the single ion conductive property due to low mobility of PEO chains with charged end(s) rather than that of free counter microions. In order to obtain cation conductive materials, PEOs
Crystal
Liquid
500
Wax, Crystal
1000 Mw of PEO
Fig. 1. Schematic illustration of PEO/salt hybrids and their state as the function of PEO molecular weight.
Fig. 2. Terminal structure of PEO/salt hybrids.
having anion terminal(s) and free cations are synthesised [9e23]. Sulfonate residue, carboxylate residue, and sulfonamide residue are selected as typical negative charges to fix on the PEO terminal(s). These hybrids are obtained as amorphous salts when PEO with molecular weight ranging from 1000 to 350. The ionic conductivity of these hybrids was around 104 S cm1 at room temperature [11]. This is an excellent ionic conductivity as a single ion conductor. Since charge density of the PEO/salt hybrids is the function of PEO molecular weight, the hybrids composed of shorter PEO chains are expected to show high ionic conductivity. For higher ionic conductivity in the bulk, dissociation degree of the terminal salt is important. As expected, PEO/salt hybrid with sulfonamide terminal showed higher ionic conductivity because of higher degree of dissociation [13]. Similarly, higher ionic conductivity was found when cations with larger ion radius were used, namely potassium salts show higher ionic conductivity than sodium salts or lithium salts [15]. There is another approach to get cation conductive polymer electrolyte. Organoboron polymers are prepared since boron acts as an anion receptor through strong interaction of anion with vacant p-orbitals of boron atom (Fig. 3, A) [24e27]. Since anions are trapped by boron atoms, added salts were promoted to dissociate to generate more amounts of cations. This leads to an increase in lithium cation transference number (tþ Li). When lithium triflate was added, the tþ Li of more than 0.8 was obtained. These ion conductive properties of organoboron polymer/salt mixtures are better than simple mixture of PEO and salts [24]. Mechanism of ion conduction in this organoboron polymer/salt mixture was confirmed to depend on the segmental motion of the component polymers which was analysed with VogeleFulchere Tamman (VFT) equation [27]. The ionic conductivity in this mixture was not so high because of low career ion concentration attributed to low solubility of the salts in this organoboron polymer. Relatively low polarity and high Tg of the organoboron polymers are the remained agendas. It is effective to introduce polyether domains into organoboron polymers to improve both points(Fig. 3, B), but cations were trapped by the polyether segments as mentioned above, and the ionic conductivity was not so high as expected. PEO/salt hybrids having cation(s) on the PEO chain end(s) should be useful to design anion conductive materials [28e30]. Imidazolium, ammonium, and pyridinium cation units are introduced at the PEO terminal(s). These are also obtained as amorphous
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
3
Fig. 4. Structure of poly(N-vinyl-3-methylimidazolium [Tf2N]).
A
B Fig. 3. Structure of PIL containing borate anions on the main chain.
salts when PEO molecular weight of from 350 to 1000. Anion transport number should also be larger than 0.5 but it was not measured electrochemically because there was no active electrodes for corresponding anions, and diffusion coefficient of these anions was not measured with NMR at that time. In any case, one of potential preparation methods to prepare polymer electrolytes to transport target ions predominantly was established in this period. 3. Polymerisation of IL as polymer electrolytes When we were playing games with the above mentioned PEO/ salt hybrids, there is the other stream on the potential ion conductive material; ionic liquids composed of organic ions! Wilkes et al. reported air and water stable ILs using bis(trifluoromethansulfonyl)imide (Tf2N) anion in 1992 [2], and number of papers on the ILs slowly increased after this paper [31e34]. After 2000, the number of the papers on the ILs steeply increased with several years’ delay. Some ILs have been potentially studied as candidates of added salts for preparing polymer electrolytes. There are two major ways to prepare polymer electrolytes with relatively low-molecular weight ILs. One is gel-type polymer electrolyte composed of cross-linked polymer matrix and ILs and the other is the polymerised ILs. Former is proud of very high ionic conductivity that is comparable to that of pure ILs. However, the weak point is mobile ion species. In the gel-type polymer electrolytes, mobile ions are all component ions for the added ILs. Generally larger ions are used for the preparation of ILs, these component ions are not so useful for the electrochemical devices. The only exception was the capacitors or super-capacitors, where ion species are not the main issue but the ionic conductivity of the electrolyte layer is important. In spite of high ionic conductivity, there are a few drawbacks of these gel-type polymer electrolytes such as long-term durability and limit of ion species. Since there are many papers and reviews on the gel-type polymer electrolytes containing ILs [35,36], we do not go into details on this subject. Here we focused on the latter ones, so-called polymerised ILs. We first reported polymerised ILs (PILs) in 1998 [3]. N-Vinyl-3ethylimidazolium [Tf2N] was polymerised with radical initiators as shown in Fig. 4. This is a quite easy way to prepare PILs because
monomer was easily prepared by the quaternisation of N-vinylimidazole with alkyl halide, then anion species was converted into Tf2N. Since Tf2N salts are hardly soluble in water, it is rather easy to purify the monomers with water washing. As expected, Nvinylimidazolium [Tf2N] turned to solid state after polymerisation. The ionic conductivity also dropped lower than one-hundredth after polymerisation as shown in Fig. 5 [3]. The reason for this drop in the ionic conductivity was mainly the drop of mobility of the ions in the medium. This is easily comprehensible that solution turned solid after polymerisation. The number of mobile ions fell to one-half because all cations were fixed onto the polymer main chains. Then salts were added to improve the ionic conductivity. Although equimolar amount of Li[Tf2N] was added to this PIL, the ionic conductivity was not improved as expected. New strategy to improve the ionic conductivity of thus prepared PILs will be mentioned later. There is another point to be mentioned here, neutralisation of amines with acid to prepare ILs. Tertiary amines were neutralised (or protonated) by the addition of the equimolar amount of acids to prepare ILs [37]. This method was called neutralisation method, and it was very effective and easy way to prepare ILs and monomeric ILs [37,38]. In some cases of the mixing of amines with weak acids, it should be better to add acids a little excess and then the formed ILs were separated to get purified ILs. Thus prepared ILs composed of protonated amines and acids were called protic ILs and were expected to be proton conductive ILs [39]. There are many reviews on the PILs, and particularly, we have first reported the review on the initial stage of these PILs [40]. The PILs are classified into several cases by the polymerisation manner as shown in Fig. 6. Here these cases are introduced with their characteristics.
Fig. 5. Arrhenius plot of the ionic conductivity for monomer (C), polymer (B), and an equimolar mixture of polymer and LiTf2N (:).
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
4
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
+ polymer
Polycation
Zwitterionic-type polymer
Ionic liquids
3.2. Polyanion type ILs
Polyanion
Polymer blend
due to the trapping of cations in spite of lower Tg of the oligoether chains [44]. By the contribution of oligoethylene chain as an effective spacer between the IL unit and the polymer main chain, drop of ionic conductivity was suppressed to one tenth of the ionic conductivity of the corresponding monomer [45]. It is also important to select suitable cation structure for high ion conduction. Eight different cation structures for monomeric ILs were selected and compared their ionic conductivity after polymerisation. Imidazolium-type cation was confirmed to be better cation structure due to lower Tg of the corresponding monomers and polymers [46].
Copolymer Fig. 6. The schematic illustration of the strategy to prepare PILs with different manners.
3.1. Polycation type ILs Polymerisable cations such as N-viny-3-alkylimidazolium cations were synthesised and used to prepare polymerisable ILs [3,41e 46]. Polymerisation of N-viny-3-alkylimidazolium salts was easily carried out as mentioned above. Since cations were fixed onto the polymer chains, these PILs were classified as anion-conductive polymers. As mentioned before, drawback of the polycation-type PILs is the drop of ionic conductivity. Addition of salts to these PILs improved the ionic conductivity. For example equimolar amount of LiTf2N was added to N-viny-3-alkylimidazolium Tf2N salt to considerably improve the ionic conductivity due to increase in the number of mobile ions [43]. To improve the segmental motion of IL units that were fixed on the polymer main chain, spacers were introduced in between the IL units and the polymer main chain. Fig. 7 shows typical example of the polymerisable IL with oligoether chain or oligoethylene chain spacers. Higher ionic conductivity of the PILs was found when oligoethylene chains were introduced in them than the case of oligoether chains were used as spacers. When oligoether chain was used as a spacer, there are considerable interaction between the oligoether spacer chains and cations. This ionedipole interaction lowered the ionic conductivity
Compared with cationic monomers, there are many anionic monomers which can form ILs. Much wider variety of structure of anions is the reason why many anions have chance to form ILs. In other words, it is possible to find suitable cations having adequate properties to form anionic monomers. Polymerisation of ILs composed of anion monomers and cations gave polyanion type ILs in which only cations were mobile. Similar to previous study, simple polymerisation of ILs composed of anion monomers and cations was quite easy to get solid polymers, but their ionic conductivity dropped considerably as compared with that of monomeric ILs. Spacer chains were also introduced between polymerisable group and IL ion pair to improve the ionic conductivity after polymerisation. As expected, the ionic conductivity of polymers was improved 20 times higher than that of the polymers without any spacers [47]. Some ILs were added to polyelectrolytes lithium salts to form lithium ion conductive polymer gels [48]. This is a convenient method to prepare lithium ion conductive films. A key issue is the compatibility between polyelectrolyte matrix and the added ILs. Many ILs were compatible with polyanions, and homogeneously formed gels showed ionic conductivity depending on the amount of the ILs. Only 5 wt% polyanions were found to be required to form gels with ILs. Arrhenius plots of ionic conductivity were depicted on a curved line strongly suggesting that the ion conduction was governed by the diffusion of ions in a viscous matrix. As mentioned in the section of PEO/salt hybrid, organoboron polymers are prepared to fix anions on the chains. These are a kind of polyanion (Fig. 3) [24e27,49] and accordingly behave as cationconductive polymer electrolytes. Some of these mixtures of organoboron polymers and ILs showed very high lithium cation transference number of 0.87. Again, it should be mentioned here that it is quite difficult to achieve both high cation transference number and high ionic conductivity keeping sufficient mechanical strength of polymer films. For these, it should be important to design strong polymeric scaffolds containing fractions with low Tg and high affinity with ILs especially with anions. 3.3. Copolymer type ILs
Fig. 7. Structure of typical polymerisable ILs having spacers.
Copolymerisation of ILs composed of both cationic monomers and anionic monomers has also been examined. Simple copolymerisation of ILs composed of only both cationic monomers and anionic monomers mostly gave charged polymer solid that was insoluble in most solvents. The copolymers were obtained as flexible and transparent solid when IL composed of monomeric cation and typical anion and IL composed of monomeric anion and typical cation were copolymerised. For example, 3-(6-acryloyloxyhexyl)-1ethylimidazolium [Tf2N] salt ([C6eim][Tf2N]) was copolymerised with methacryloyloxy-octa(ethylene oxide)-2-sulfobenzoate lithium salt (PE8) [50]. By increasing fraction of PE8 in the copolymers, Tg was found to increase and accordingly the ionic
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
conductivity decreased. However, since PE8 contains mobile lithium cation, lithium ion transference number was found to reach higher than 0.5 by increasing PE8 fraction. It is therefore quite difficult to increase both ionic conductivity and lithium ion transference number by the copolymerisation of ILs. Alternative copolymer of imidazolium cation and borate anion was synthesised by the direct coupling reaction followed by the anion exchange from bromide anion to Tf2N anion as shown in Fig. 8 [51]. The ionic conductivity of this copolymer was about 106 S cm1 at 50 C. This conductivity is high considering relatively high Tg of this polymer (6 C).
5
synthesised and polymerised. As easily expected, ionic conductivity of pure polymers of these polymerisable zwitterions was quite low clearly showing that there are no mobile ions. However, addition of Li[Tf2N], for example, improved their ionic conductivity like in the case of low molecular weight zwitterion/Li[Tf2N] mixtures. In spite of improved ionic conductivity of the polymerised zwitterions after addition of salts, polymerised zwitterions are not suitable to design ion conductive polymers. There are other applications for these polymerised zwitterions such as selective gas permeation membranes. 3.5. Cross-linked PIL
3.4. Polyzwitterion type ILs Zwitterion is composed of tethered cation and anion. There were a few challenges to prepare zwitterions from component ions of ILs [52]. When ILs that does not migrate under potential gradient were synthesised, these ILs should be useful for electrochemical reaction media. Of course all ions are known to migrate along with the potential gradient, but once cation and anion were tethered, they should not migrate under the same condition. Zwitterionic liquids were considered to be attractive materials reflecting this idea. More than 20 different zwitterions composed of imidazolium cation and sulfonate or carboxylate anion have initially been synthesised, and their physico-chemical properties have been analysed [53]. However, their melting point was above 100 C, and all of them were solid at room temperature. It was clarified that the tethering cation and anion increased the melting point 120 C higher than that of corresponding ILs. In spite of poor properties as ILs, these zwitterions have been used as catalysts or effective additives for many reactions. We however do not go into details on the application of zwitterions here. It should be noted here that drastic increase in the ionic conductivity of zwitterions was found when Li[Tf2N] was added equimolarly to the zwitterion [52]. This increase was comprehended to the formation of IL-like low Tg environment by the coupling of imidazolium cation on the zwitterion and [Tf2N] anion. Transport of lithium cation was suggested in this mixture because couple of [Tf2N] anion and zwitterion was much less mobile than that for small lithium cation. There were also a few trials to synthesise polymerised zwitterions [52]. Zwitterions containing polymerisable vinyl groups were
Polymerised IL gels are also interesting materials. Oligo(ethylene oxide) diacrylates have frequently been used as crosslinker without elevating Tg of the polymers. Oligo(ethylene oxide) chains were used as spacer between polymerisable group and IL salt as well as cross-linker, expecting low Tg of the polymers. For example, tetra(ethylene oxide) diacrylates was added to the imidazolium-type polymerisable IL monomers, and the mixture was radical polymerised to prepare flexible and transparent crosslinked PIL film (Fig. 9) [54]. They showed high ionic conductivity of about 104 S cm1 at room temperature due to low Tg. Cross-linkers with IL units were also designed to prepare thermally stable cross-linked PILs [55]. Acryloyl groups were introduced into IL unit containing two imidazolium cations (so called Geminitype ILs), and these were added as cross-linker to the imidazolium type IL monomers. For example, 1,4-bis{3-[2-(acryloyloxy)ethyl] imidazolium-1-yl}butane bis(trifluoromethanesulfonyl)imide was used as this kind of IL cross-linker. The polymerisation gave crosslinked PILs with high ionic conductivity. The ionic conductivity decreased by increasing the amount of IL-cross-linker in the PILs due to lowering the diffusion coefficient of mobile ions in the polymer matrix. The most notable point of this study is considerably improved decomposition temperature of the PILs. Ten percent weight loss was found above 400 C when this series of cross-linked PILs were analysed. Simple polymerisation of Gemini-type IL crosslinker gave ion conductive solid with high decomposition temperature [55]. IL cross-linker was found here to be effective cross-linker to keep thermal stability of the ILs. This stabilisation cannot be realised by the polyether type cross-linkers.
Fig. 8. Alternative copolymer of imidazolium cation and borate anion.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
6
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
Fig. 9. Picture of cross-linked flexible PIL film.
On the other hand, cross-linker is not always required to prepare IL gels. There are some papers on the gel formation of ILs, for example, some amino acid ILs were effective to form gels by mixing them with some zwitterions [56]. Even mixing of some different amino acid ILs was effective to form IL gels [57]. The ionic conductivity of these ILs was maintained even after gel formation. The only drawback of these IL gels is that the mobile ions are limited to IL component ions. These gels are not suitable to transport target ions such as lithium cation. However, this is an easy-to-use method to prepare conductive gels having excellent advantage of ILs such as negligibly small vapour pressure at wide temperature range.
3.6. PILs prepared with natural polymers There are a few approaches to prepare PILs with natural polymers. Natural rubber is a representative elastic material. These rubbers are widely used in our life. There is however almost no examples to make conductive materials with these rubbers because they are typical insulator. It is difficult to add inorganic salts in natural rubber due to low polarity of the rubbers, and little increase in the ionic conductivity was expected due to little degree of dissociation of the salts if these inorganic salts were compatible with the rubbers. Sometimes this relatively low compatibility works good to prepare successive ion conduction pathways. Relatively less polar ILs and/or hydrophobic lithium salts were examined to make composites with natural rubbers [58e60]. Most ILs were found to show clear and macroscopic phase separation with these rubbers, but some ILs showed micro-phase separation with rubbers which is effective to construct successive ion conduction pathways. Macroscopically homogeneous composite with two Tg values, attributed to rubbers and additive ILs, is the proof of this kind of micro-phase separation. The PEO/salt hybrids as mentioned above were also added to natural rubbers to prepare ion conductive elastic films [58]. Nitrile rubber, reference material for natural rubber, was also mixed with ILs to improve ionic conductivity, but their affinity was not the same as the case of natural rubber [59]. Then, imidazolium-type zwitterion was added to nitrile rubber to enhance micro-phase separation and accordingly higher ionic conductivity was observed [60]. In spite of increasing demand on the conductive rubbers, there are not potential methods to achieve high conductivity. ILs and their polymers should be effective to improve conductivity of natural and synthetic rubbers.
Cellulose, typical natural polysaccharide, was also used as base polymer to prepare ion conductive polymers [61]. After designing ILs as solvents for cellulose [62e64], there are some approaches to chemical modification of cellulose directly in the ILs. Novel ion-gels bearing lithium borate were synthesised via condensation between cellulose and pentafluorophenyl boric acid in polar IL. Ion-gels obtained showed very high ionic conductivity (103 S cm1 at 90 C) comparable to the added IL itself [49]. We have also made PILs with deoxyribonucleic acid (DNA). DNA is known as a molecular memory at the genetic level. Nucleic acid bases are heteroaromatic rings, and it is easy to associate them with starting amines such as imidazole or pyridine rings which are frequently-used components for preparing ILs. Then, we have tried to make ILs with nucleic acid bases as a preliminary study. Since there are several mobile protons on these nucleic acid bases, quaternisation with alkyl halide was concluded not to be a suitable method to make cations from these nucleic acid bases. Then the nucleic acid bases were neutralised with acids [65]. Neutralisation has already been introduced to be an easy but potential method to prepare ILs [37], and this neutralisation method was again used to prepare nucleic acid base salts. After neutralisation with HTf2N, both adenine and cytosine salts were obtained as liquids. Since both guanine and thymine were weak acids, they did not form salts with strong acid. Compare with cytosine, adenine was weaker against acid treatment, further studies were carried out with cytosine. Cytosine was neutralised with some different acids, and melting point of the formed salts was found to be the function of pKa of the acids. Based on the above-mentioned reference studies, we have neutralised DNAs with HBF4 or HTf2N in an aqueous medium. DNA, from salmon milt, was dissolved in aqueous solution, and acids were added, stirred gently and dried under reduced pressure. Since the content of both adenine and cytosine in any DNA should be 50%, they would provide successive IL-like moieties along with the DNA chains. The obtained neutralised DNA was white powder. However double strand helix structure of the DNA was not maintained anymore because of breaking complementary hydrogen bonds and charge repulsion among IL-like moieties. Accordingly, low ionic conductivity was found with these neutralised DNAs. Addition of cytosine [Tf2N] salt to these neutralised DNAs enhanced the ionic conductivity only a little, because the inherent ionic conductivity of cytosine [Tf2N] salt was not so high (6.9 105 S cm1 at 50 C). Of course, addition of typical ILs to these neutralised DNAs greatly enhanced the ionic conductivity. These were obtained as transparent and flexible films to show moderate ionic conductivity. An ionic liquid domain was successfully prepared outside double-stranded DNA by fixing 1-alkyl-3-methylimidazolium cations on the phosphate anions of DNA. The counter Naþ cations of the phosphate groups of DNA were exchanged with the imidazolium cations. The resulting IL-robed DNA was soluble in ordinary organic solvents such as methanol or ethanol. Ionic conductivity of the IL-robed DNA was low, because the ion density was insufficient to form continuous IL domains around the DNA strands [66]. Addition of some ILs to this IL-robed DNA certainly improved the ionic conductivity. DNAs were confirmed to be useful scaffold materials to prepare ion conductive films. 4. Phase behaviour of PIL/water mixtures General PILs are soluble in water and show typical polyelectrolyte behaviour. By increasing hydrophobicity of the PILs, they turned insoluble in water and absorb small amount of water to form hydrated polyelectrolyte gels. What kind of phase behaviour was seen when PILs with intermediate
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
Cations
7
monomers used to design this kind of polyelectrolytes. Among many pairs, following six-different polymerisable ILs ([P4,4,4,4][SS], [P4,4,4,6][C3S], [P4,4,4,8][C3S], [P4,4,4,VB][C5S], [P4,4,4,VB][C3S], and [P4,4,4,VB][AC3S]) were confirmed to show the LCST-type phase behaviour in water. Polymerisation of these polymerisable ILs also provided polyelectrolytes showing LCST-type phase change in water [69]. Fig. 11 shows an example of the phase behaviour of the mixture of water and poly([P4,4,4,6][C3S]) having adequate hydrophobicity/hydrophilicity balance. This is the first paper on the preparation of polyelectrolytes showing LCST-type phase change in water without using well-known key monomers such as isopropylacrylamide [68]. Transmittance of the aqueous solution of this PIL showed sharp change in response to temperature [69]. From this steep change in transmittance, it is easy to determine the phase transition temperature (Tc). The Tc of these PILs was not the same as that of corresponding monomers, and further it was the function of both polymer concentration and salt concentration as seen in Fig. 12. The transition temperature was also found to be the function of degree of polymerisation. Since the transition temperature was affected by the hydrophobicity/hydrophilicity balance of monomers, it is easy to control the Tc by the composition of comonomers. The phase change is very sensitive to the temperature, and this kind of polymers should be used in many applications.
5. Other application
Anions Fig. 10. Structure of composed ions for ionic liquid monomers to show LCST type phase behaviour after mixing them with water.
hydrophobicity were mixed with water? In 2007, we found ILs to show unique phase change after mixing them with water [67]. Partly hydrophobic amino acid-derived ILs were found to be soluble with water, but they turned to be insoluble by heating. This phase change, namely lowering solubility by heating, is generally known as lower critical solution temperature (LCST)-type phase change. After analysing the required physicochemical properties of ILs to show this kind of LCST-type phase behaviour after mixing them with water, balance of hydrophilicity and hydrophobicity of ILs was found to be the dominant factor [68]. After this study, the hydrophobicity/hydrophilicity of ILs was semi-empirically controlled by selecting suitable ion pairs [69]. Mixture of water and polymerisable ILs also showed this kind of LCST-type phase change only when the hydrophobicity/hydrophilicity balances of ILs was in the proper range. Fig. 10 shows typical
Area of PIL science has been spread wider than we initially expected. Here we briefly introduce other interesting and noteworthy results on PILs reported by other scientists. In1999, [C4mim]PF6 was found to dissolve large amount of CO2 gas [70]. After this report, it is generally known that CO2 gas was considerably dissolved in many ILs as compared with other gas molecules. Then PILs were also prepared as membranes expecting selective adsorption and/or separation of CO2 gas [71e75]. This CO2 gas separation is one of big subjects of PILs now. Study on the catalytic activity of ILs was also applied to design catalytic PILs. Since polymeric catalysts are easy to treat, catalytic PILs are of great interests for recycling of catalysts as well as separation of products from catalysts. For example, PILs were reported as a catalyst for nucleophilic substitution reactions including fluorinations [76]. There are a few reports that synthesis of carbonates via cycloaddition reaction of CO2 with epoxides was catalysed by PIL [77e79]. Heavy metal ions, expected to act as a catalyst, can be fixed on PIL as counter cations. Also, counter anions of imidazolium-type PIL were exchanged with Pd complex to catalyse Sonogashira coupling reactions [80]. Other imidazolium type IL was fixed on resins to be used to catalyse Michael reaction. The reaction was carried out without solvent to get more than 90% yield with excellent enantioselectivity up to 99%. It is known that ILs are used as extraction of many target materials from aqueous media. When these ILs were
Fig. 11. Polymerisation of [P4,4,4,6][C3S] (a, b, and c) and LCST behaviour (c and d) of poly([P4,4,4,6][C3S]) after mixing it with water.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
8
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
been supported by the Grant-in-Aid for Scientific Researches from the Japan Society for the Promotion of Science (No. 21225007, 17205020, 17073005, 14205136, 11555250, and 08246101). References [1] [2] [3] [4] [5] [6] [7] [8]
Fig. 12. Temperature-dependent change in transmittance of poly([P4446][C3S]) in pure water upon heating. Change in transmittance of poly([P4446][C3S]) after mixing with pure water. The polymer concentration was 2.5 wt% (a), 5 wt% (b), 10 wt% (c), and 20 wt% (d) respectively.
polymerised, PILs were expected to be used as adsorbents. Since these PILs are insoluble in a medium, separation of target materials turns easier than that with liquid extractants. For example, PIL was used to adsorb heavy metal ions from waste water. Imidazolium-type and pyridinium-type ILs were fixed onto bcyclodextrin, and they were polymerised to adsorb heavy metal ions [81]. The p-nitrophenol, 2,4,6-trichlorophenol, and Cr6 ion were successfully extracted from an aqueous mixture. Poly(triethylenetetramine acrylate salt) was prepared to collect petroleum from petroleum/water mixture [82]. This will relate to the recovery of valuable or target materials from waste water or polluted seawater. There should be many different approaches to collect, extract, or recover useful or target materials from mixtures with suitably-designed PILs. There are many excellent reviews on PIL researches, and readers are strongly recommended to refer them, for examples, those by Mecerreyes [83] and Yuan [84].
[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38]
6. Conclusion Development of polymerised ionic liquids in these 15 years has been mentioned in this mini-review. At the initial stage, poly(ethylene oxide) chains were tethered to ions to lower the melting point of the salts, and these were analysed as components of polymerised ionic liquids. Report on the ionic liquids with low melting point pushed us to introduce imidazolium type ionic liquid unit on the end of poly(ethylene oxide), and at the initial stage of polymerised ionic liquids, we then realised the potential possibility of these ionic liquids on the design of ion conductive polymers. After this stage, we made efforts to design many polymerised ionic liquids. During these 15 years, number of papers on the polymerised ionic liquids increased year by year, and their applications spread widely. Polymer materials are good in designing functional materials because of their unique properties to make three dimensional domains. Use of these polymerised ionic liquids will be common in many fields in the near future. Acknowledgement The present mini-review is based on the results mainly obtained in our lab for these 15 years. Most part of these researches have
[39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64]
Walden P. Bull Acad Sci; 1914:405e22. St. Petersburg. Wilkes JS, Zaworotko MJ. J Chem Soc Chem Commun; 1992:965e7. Ohno H, Ito K. Chem Lett 1998;27:751e2. Mukai T, Yoshio M, Kato T, Ohno H. Future Materials 2006;6(2):2e5 [in Japanese]. Ohno H, editor. Electrochemical aspects of ionic liquids. John Wiley & Sons; 2011. Kishimoto K, Yoshio M, Mukai T, Yoshizawa M, Ohno H, Kato T. J Am Chem Soc 2003;125:3196e7. Hoshino K, Yoshio M, Mukai T, Kishimoto K, Ohno H, Kato T. J Polym Sci Part A Polym Chem 2003;41:3486e92. Kishimoto K, Suzawa T, Yokota T, Mukai T, Ohno H, Kato T. J Am Chem Soc 2005;127:15618e23. Ohno H, Ito K. Polymer 1995;36:891e3. Ito K, Ohno H. Solid State Ionics 1995;79:300e5. Ito K, Nishina N, Tominaga Y, Ohno H. Solid State Ionics 1996;86-88:325e8. Ito K, Tominaga Y, Ohno H. Electrochim Acta 1997;42:1561e70. Tominaga Y, Ito K, Ohno H. Polymer 1997;38:1949e51. Ito K, Nishina N, Ohno H. J Mater Chem 1997;7:1357e62. Ito K, Ohno H. Electrochim Acta 1998;43:1247e52. Tominaga Y, Ohno H. Chem Lett 1998;27:955e6. Tominaga Y, Ohno H. Solid State Ionics 1999;124:323e9. Tominaga Y, Takizawa N, Ohno H. Electrochim Acta 2000;45:1285e9. Furukawa T, Yoneya K, Takahashi Y, Ito K, Ohno H. Electrochim Acta 2000;45: 1443e8. Kato K, Ito K, Ohno H. J Solid State Electrochem 2000;4:141e5. Tominaga Y, Ohno H. Electrochim Acta 2000;45:3081e6. Tominaga Y, Mizumo T, Ohno H. Polym Adv Technol 2000;11:524e8. Mizumo T, Ohno H. Polymer 2004;45:861e4. Matsumi N, Sugai K, Ohno H. Macromolecules 2002;35:5731e3. Matsumi N, Mizumo T, Ohno H. Chem Lett 2004;33:372e3. Mizumo T, Sakamoto K, Matsumi N, Ohno H. Chem Lett 2004;33:396e7. Matsumi N, Sugai K, Sakamoto K, Mizumo T, Ohno H. Macromolecules 2005;38:4951e4. Ito K, Ohno H. Polymer 1997;38:921e6. Ohno H, Nakai Y, Ito K. Chem Lett 1998;27:15e6. Nakai Y, Ito K, Ohno H. Solid State Ionics 1998;113e115:0199e204. Fuller J, Breda AC, Carlin RT. J Electrochem Soc 1994;144:L67e70. Bonhôte P, Ana-Paula Dias, Papageorgiou N, Kalyanasundaram K, Grätzel M. Inorg Chem 1996;35:1168e78. Ohno H. Bull Chem Soc Jpn 2006;79:1665e80. Armand M, Endres F, MacFarlane DR, Ohno H, Scrosati B. Nat Mater 2009;8: 621e9. Ueki T, Watanabe M. Macromolecules 2008;41:3739e49. Lu J, Yan F, Texter J. Prog Polym Sci 2009;34:431e48. Hirao M, Sugimoto H, Ohno H. J Electrochem Soc 2000;147:4168e72. Yoshizawa M, Ogihara W, Ohno H. Electrochem Solid State Lett 2001;4:E25e 7. Belieres J, Angell CA. J Phys Chem B 2007;111:4926e37. Ohno H. Electrochim Acta 2001;46:1407e11. Yoshizawa M, Ohno H. Chem Lett 1999;28:889e90. Hirao M, Ito K, Ohno H. Electrochim Acta 2000;45:1291e4. Hirao M, Ito-Akita K, Ohno H. Polym Adv Technol 2000;11:534e8. Ohno H, Yoshizawa M, Ogihara W. Electrochim Acta 2004;50:255e61. Yoshizawa M, Ohno H. Electrochim Acta 2001;46:1723e8. Ogihara W, Washiro S, Nakajima H, Ohno H. Electrochim Acta 2006;51:2614e 9. Yoshizawa M, Ogihara W, Ohno H. Polym Adv Technol 2002;113:589e94. Ogihara W, Sun J, Forsyth M, MacFarlane DR, Yoshizawa M, Ohno H. Electrochim Acta 2004;49:1797e801. Matsumi N, Nakamura Y, Aoi K, Watanabe T, Mizumo T, Ohno H. Polym J 2009;41:437e41. Ogihara W, Suzuki N, Nakamura N, Ohno H. Polym J 2006;38:117e21. Narita A, Shibayama W, Matsumi N, Ohno H. Polym Bull 2006;57:109e14. Yoshizawa M, Hirao M, Ito-Akita K, Ohno H. J Mater Chem 2001;11:1057e62. Yoshizawa M, Narita A, Ohno H. Aust J Chem 2004;57:139e44. Washiro S, Yoshizawa M, Nakajima H, Ohno H. Polymer 2004;45:1577e82. Nakajima H, Ohno H. Polymer 2005;46:11499e504. Taguchi S, Ichikawa T, Kato T, Ohno H. RSC Adv 2013;3:23222e7. Kagimoto J, Nakamura N, Kato T, Ohno H. Chem Commun; 2009:2405e7. Yoshizawa M, Marwanta E, Ohno H. Polymer 2000;41:9049e53. Marwanta E, Mizumo T, Nakamura N, Ohno H. Polymer 2005;46:3795e800. Marwanta E, Mizumo T, Ohno H. Solid State Ionics 2007;178:227e32. Fukaya Y, Sugimoto A, Ohno H. Biomacromolecules 2006;7:3295e7. Fukaya Y, Hayashi K, Wada M, Ohno H. Green Chem 2008;10:44e6. Ohno H, Fukaya Y. Chem Lett 2009;38:2e7. Abe M, Fukaya Y, Ohno H. Green Chem 2010;12:1274e80.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9 [65] Nishimura N, Ohno H. J Mater Chem 2002;12:2299e304. [66] Nishimura N, Nomura Y, Nakamura N, Ohno H. Biomaterials 2005;26:5558e 63. [67] Fukumoto K, Ohno H. Angew Chem Int Ed 2007;46:1852e5. [68] Kohno Y, Ohno H. Aust J Chem 2012;65:9194e7. [69] Kohno Y, Deguchi Y, Ohno H. Chem Commun 2012;48:11883e5. [70] Blanchard LA, Hancu D, Beckman EJ, Brennecke JF. Nature 1999;399:28e9. [71] Tang J, Tang H, Sun W, Radosz M, Shen Y. Polymer 2005;46:12460e7. [72] Bara JE, Noble RD, Gin DL. Ind Eng Chem Res 2009;48:4607e10. [73] Carlisle TK, Bara JE, Lafrate AL, Gin DL, Noble RD. J Membr Sci 2010;359:37e 43. [74] Hudiono YC, Carlisle TK, Lafrate AL, Gin DL, Noble RD. J Membr Sci 2011;370: 141e8. [75] Carlisle TK, Nicodemus GD, Gin DL, Noble RD. J Membr Sci 2012;397e398:24e37.
9
[76] Kim DW, Chi DY. Angew Chem Int Ed 2004;43:483e5. [77] Xie Y, Zhang Z, Jiang T, He J, Han B, Wu T, et al. Angew Chem Int Ed 2007;46: 7255e8. [78] Sun J, Cheng W, Fan W, Wang Y, Meng Z, Zhang S. Catal Today 2009;148:361e 7. [79] Yu J, Choi H, Selvaraj M, Park D. Reac Kinet Mech Cat 2011;102:353e65. [80] Wang Y, Liu J, Xia C. Tetrahedron Lett 2011;52:1587e91. [81] Mahlambi MM, Malefetse TJ, Mamba BB, Krause RW. J Poly Res 2010;17:589e 600. [82] Asadov ZH, Rahimov RA, Poladova TA, Nasibova SM, Guliyev AD, Asadova AZ. J Mol Liq 2012;166:44e8. [83] Mecerreyes D. Prog Polym Sci 2011;36:1629e48. [84] Yuan J, Mecerreyes D, Antonietti M. Prog Polym Sci 2013;38:1009e36.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
Contents lists available at ScienceDirect
Polymer journal homepage: www.elsevier.com/locate/polymer
15th anniversary of polymerised ionic liquids Naomi Nishimura a, b, Hiroyuki Ohno a, b, * a b
Department of Biotechnology, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan Functional Ionic Liquid Laboratories, Graduate School of Technology, Tokyo University of Agriculture and Technology, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 7 January 2014 Received in revised form 10 February 2014 Accepted 11 February 2014 Available online xxx
Polymerised ionic liquids (PILs) have unique properties such as low glass transition temperature (Tg) in spite of very high charge density. Due to these advanced points, PILs have been prepared and initially evaluated as ion conductive polymers. Progress of low-Tg polyelectrolytes has been previously demonstrated with polyethers having charged end(s) as a kind of PILs. Then, imidazolium-type ionic liquids (ILs) were polymerised after introducing vinyl groups onto the imidazolium cation rings. It is reasonable that the ionic conductivity of thus prepared PILs decreased due to elevation of Tg and decrease of the number of mobile small ions. Efforts were then paid to suppress drop of ionic conductivity after polymerisation. Variety of PILs has been improved to show excellent ionic conductivity, selective ion transport, and other properties. With the progress of functional ILs, some functions were also added to PILs which cannot be realised with ordinary charged polymers. In the present mini-review, we briefly introduce history of a variety of polymerised ILs and some applications of these PILs. Ó 2014 Elsevier Ltd. All rights reserved.
Keywords: Polymerisation Glass transition temperature Polyelectrolytes
1. Introduction Ionic liquids (ILs) are liquids composed of only both cations and anions at temperature below 100 C. Just 100 years ago, Paul Walden has reported that melting point of ethylammonium nitrate (CH3CH2NH3NO3) is 12 C [1]. This should be the first report on the ILs. On the other hand, aluminium chloride type molten salts have also a long history in this research field. After a few types of researches, air and water stable ILs were reported in 1992 [2]. At this stage there was however not many responses on this memorial paper. After this, there are explosively increasing papers on ILs after 2000 because of their interesting properties and charm of the possibility of a variety of designs on ions. We have reported polymerised ionic liquids (PILs) in 1998 for the first time in the world [3]. Our approach had not been accepted by general scientists due to inverse approach to the development of the lowering of melting point of salts. Although all of polyelectrolytes were brittle and stiff materials to deal with because of their high charge density, PILs were obtained as flexible films with low glass transition temperature (Tg). These advanced properties were gradually accepted by polymer scientists after more than five years of publication. After that, number of papers on ILs increased year by year and now more
* Corresponding author. Department of Biotechnology, Tokyo University of Agriculture and Technology, Koganei, Tokyo 184-8588, Japan. E-mail address: [email protected] (H. Ohno).
than 1000 papers on PILs have been reported. These materials have different attractive points from those of ILs, and there are many attempts to seek the possibilities to apply PILs to many different research fields. In this mini-review, we desire to summarise research results on PIL mainly obtained in our laboratory. Since polymers prepared from ionic liquids are widely called as polymerised ionic liquids (PILs), we also use this abbreviation in this mini-review. In the initial stage, PILs were prepared by the direct polymerisation of IL monomers. PILs were mainly prepared from imidazolium-cation type ILs. By the progress of researches on ILs, many kinds of polymerisable cations and anions have been synthesised and used to prepare PILs. It should be important to choose cations or anions to prepare PILs for particular purpose. Recently, there are increasing researches to prepare PILs with controlled molecular weight and sequences by precise polymerisation. Accordingly, there are many studies on the different PILs having novel structure and functions to date. Those PILs have been contributed to spread the possibility to use these PILs in a variety of scientific fields. In order to add functions onto ILs, it is a common way to introduce functional groups onto component ions of the ILs. Since it is also possible to put functions onto ILs by the structural control, we should care about the dimension of the ILs. To lower the dimension of the three-dimensionally isotropic ILs, there should be additional advancement reflecting the lowered dimension. Dimension control of the ILs is another very interesting subject, and
http://dx.doi.org/10.1016/j.polymer.2014.02.042 0032-3861/Ó 2014 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
2
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
there are two major way to lower the dimension of the ILs; the use of liquid crystalline phase and polymerisation. It is of great interesting approach to use the liquid crystalline characteristics for the functionalisation of ILs [4,5]. For examples, anisotropic ionconductive materials were designed by lowering dimension of isotropic ILs. Although we have reported more than 30 papers on this subject, we do not want to discuss this subject here because of limited pages. Among these, there are a few papers on the polymerisation of ILs in the liquid crystalline state [6e8]. The use of liquid crystalline properties onto the functional design of ILs will be reviewed in the near future. The present review treats mainly the results on the polymerised ILs. 2. PEO/salt hybrids Before introducing IL-based PILs, we have to mention about the pre-study on the low Tg polyelectrolytes. We have been studying ion conductive polymers for over thirty years. Since polyether structure especially poly(ethylene oxide) (PEO) or poly(propylene oxide) (PPO) have very low Tg, ion conduction path have been made widely with them. In the late 20 century, it was a normal approach to mix PEOs with inorganic salts to design polymer electrolytes. Sufficiently long PEO chains and salts were mainly used to prepare solid polymer electrolytes. Thus prepared films are effective to transport ions due to low Tg of the PEO chains. We have recognised that low molecular weight PEO chains, so-called oligo(ethylene oxide)s, were liquid state at room temperature. Then, we tried to put charge(s) onto the end(s) of oligo(ethylene oxide) to prepare liquid salts. According to this idea, we have started the investigation on PEO/salt hybrid in 1993. Fig. 1 shows that the state of PEO/salt hybrid depending on the PEO molecular weight. The hybrids of salts and low molecular weight PEOs are solid because properties of salt are dominant factor for the hybrids. On the other hand, high molecular weight PEOs gave the hybrids as wax like solids. PEOs with molecular weight from 350 to 1000 are found to be effective to design liquid state hybrids. The hybrids composed of PEOs with moderate molecular weight are considered to be a kind of ionic liquid. PEO/salt hybrids which have hitherto been synthesised are shown in Fig. 2. The most important characteristic of these hybrids is the single ion conductive property due to low mobility of PEO chains with charged end(s) rather than that of free counter microions. In order to obtain cation conductive materials, PEOs
Crystal
Liquid
500
Wax, Crystal
1000 Mw of PEO
Fig. 1. Schematic illustration of PEO/salt hybrids and their state as the function of PEO molecular weight.
Fig. 2. Terminal structure of PEO/salt hybrids.
having anion terminal(s) and free cations are synthesised [9e23]. Sulfonate residue, carboxylate residue, and sulfonamide residue are selected as typical negative charges to fix on the PEO terminal(s). These hybrids are obtained as amorphous salts when PEO with molecular weight ranging from 1000 to 350. The ionic conductivity of these hybrids was around 104 S cm1 at room temperature [11]. This is an excellent ionic conductivity as a single ion conductor. Since charge density of the PEO/salt hybrids is the function of PEO molecular weight, the hybrids composed of shorter PEO chains are expected to show high ionic conductivity. For higher ionic conductivity in the bulk, dissociation degree of the terminal salt is important. As expected, PEO/salt hybrid with sulfonamide terminal showed higher ionic conductivity because of higher degree of dissociation [13]. Similarly, higher ionic conductivity was found when cations with larger ion radius were used, namely potassium salts show higher ionic conductivity than sodium salts or lithium salts [15]. There is another approach to get cation conductive polymer electrolyte. Organoboron polymers are prepared since boron acts as an anion receptor through strong interaction of anion with vacant p-orbitals of boron atom (Fig. 3, A) [24e27]. Since anions are trapped by boron atoms, added salts were promoted to dissociate to generate more amounts of cations. This leads to an increase in lithium cation transference number (tþ Li). When lithium triflate was added, the tþ Li of more than 0.8 was obtained. These ion conductive properties of organoboron polymer/salt mixtures are better than simple mixture of PEO and salts [24]. Mechanism of ion conduction in this organoboron polymer/salt mixture was confirmed to depend on the segmental motion of the component polymers which was analysed with VogeleFulchere Tamman (VFT) equation [27]. The ionic conductivity in this mixture was not so high because of low career ion concentration attributed to low solubility of the salts in this organoboron polymer. Relatively low polarity and high Tg of the organoboron polymers are the remained agendas. It is effective to introduce polyether domains into organoboron polymers to improve both points(Fig. 3, B), but cations were trapped by the polyether segments as mentioned above, and the ionic conductivity was not so high as expected. PEO/salt hybrids having cation(s) on the PEO chain end(s) should be useful to design anion conductive materials [28e30]. Imidazolium, ammonium, and pyridinium cation units are introduced at the PEO terminal(s). These are also obtained as amorphous
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
3
Fig. 4. Structure of poly(N-vinyl-3-methylimidazolium [Tf2N]).
A
B Fig. 3. Structure of PIL containing borate anions on the main chain.
salts when PEO molecular weight of from 350 to 1000. Anion transport number should also be larger than 0.5 but it was not measured electrochemically because there was no active electrodes for corresponding anions, and diffusion coefficient of these anions was not measured with NMR at that time. In any case, one of potential preparation methods to prepare polymer electrolytes to transport target ions predominantly was established in this period. 3. Polymerisation of IL as polymer electrolytes When we were playing games with the above mentioned PEO/ salt hybrids, there is the other stream on the potential ion conductive material; ionic liquids composed of organic ions! Wilkes et al. reported air and water stable ILs using bis(trifluoromethansulfonyl)imide (Tf2N) anion in 1992 [2], and number of papers on the ILs slowly increased after this paper [31e34]. After 2000, the number of the papers on the ILs steeply increased with several years’ delay. Some ILs have been potentially studied as candidates of added salts for preparing polymer electrolytes. There are two major ways to prepare polymer electrolytes with relatively low-molecular weight ILs. One is gel-type polymer electrolyte composed of cross-linked polymer matrix and ILs and the other is the polymerised ILs. Former is proud of very high ionic conductivity that is comparable to that of pure ILs. However, the weak point is mobile ion species. In the gel-type polymer electrolytes, mobile ions are all component ions for the added ILs. Generally larger ions are used for the preparation of ILs, these component ions are not so useful for the electrochemical devices. The only exception was the capacitors or super-capacitors, where ion species are not the main issue but the ionic conductivity of the electrolyte layer is important. In spite of high ionic conductivity, there are a few drawbacks of these gel-type polymer electrolytes such as long-term durability and limit of ion species. Since there are many papers and reviews on the gel-type polymer electrolytes containing ILs [35,36], we do not go into details on this subject. Here we focused on the latter ones, so-called polymerised ILs. We first reported polymerised ILs (PILs) in 1998 [3]. N-Vinyl-3ethylimidazolium [Tf2N] was polymerised with radical initiators as shown in Fig. 4. This is a quite easy way to prepare PILs because
monomer was easily prepared by the quaternisation of N-vinylimidazole with alkyl halide, then anion species was converted into Tf2N. Since Tf2N salts are hardly soluble in water, it is rather easy to purify the monomers with water washing. As expected, Nvinylimidazolium [Tf2N] turned to solid state after polymerisation. The ionic conductivity also dropped lower than one-hundredth after polymerisation as shown in Fig. 5 [3]. The reason for this drop in the ionic conductivity was mainly the drop of mobility of the ions in the medium. This is easily comprehensible that solution turned solid after polymerisation. The number of mobile ions fell to one-half because all cations were fixed onto the polymer main chains. Then salts were added to improve the ionic conductivity. Although equimolar amount of Li[Tf2N] was added to this PIL, the ionic conductivity was not improved as expected. New strategy to improve the ionic conductivity of thus prepared PILs will be mentioned later. There is another point to be mentioned here, neutralisation of amines with acid to prepare ILs. Tertiary amines were neutralised (or protonated) by the addition of the equimolar amount of acids to prepare ILs [37]. This method was called neutralisation method, and it was very effective and easy way to prepare ILs and monomeric ILs [37,38]. In some cases of the mixing of amines with weak acids, it should be better to add acids a little excess and then the formed ILs were separated to get purified ILs. Thus prepared ILs composed of protonated amines and acids were called protic ILs and were expected to be proton conductive ILs [39]. There are many reviews on the PILs, and particularly, we have first reported the review on the initial stage of these PILs [40]. The PILs are classified into several cases by the polymerisation manner as shown in Fig. 6. Here these cases are introduced with their characteristics.
Fig. 5. Arrhenius plot of the ionic conductivity for monomer (C), polymer (B), and an equimolar mixture of polymer and LiTf2N (:).
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
4
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
+ polymer
Polycation
Zwitterionic-type polymer
Ionic liquids
3.2. Polyanion type ILs
Polyanion
Polymer blend
due to the trapping of cations in spite of lower Tg of the oligoether chains [44]. By the contribution of oligoethylene chain as an effective spacer between the IL unit and the polymer main chain, drop of ionic conductivity was suppressed to one tenth of the ionic conductivity of the corresponding monomer [45]. It is also important to select suitable cation structure for high ion conduction. Eight different cation structures for monomeric ILs were selected and compared their ionic conductivity after polymerisation. Imidazolium-type cation was confirmed to be better cation structure due to lower Tg of the corresponding monomers and polymers [46].
Copolymer Fig. 6. The schematic illustration of the strategy to prepare PILs with different manners.
3.1. Polycation type ILs Polymerisable cations such as N-viny-3-alkylimidazolium cations were synthesised and used to prepare polymerisable ILs [3,41e 46]. Polymerisation of N-viny-3-alkylimidazolium salts was easily carried out as mentioned above. Since cations were fixed onto the polymer chains, these PILs were classified as anion-conductive polymers. As mentioned before, drawback of the polycation-type PILs is the drop of ionic conductivity. Addition of salts to these PILs improved the ionic conductivity. For example equimolar amount of LiTf2N was added to N-viny-3-alkylimidazolium Tf2N salt to considerably improve the ionic conductivity due to increase in the number of mobile ions [43]. To improve the segmental motion of IL units that were fixed on the polymer main chain, spacers were introduced in between the IL units and the polymer main chain. Fig. 7 shows typical example of the polymerisable IL with oligoether chain or oligoethylene chain spacers. Higher ionic conductivity of the PILs was found when oligoethylene chains were introduced in them than the case of oligoether chains were used as spacers. When oligoether chain was used as a spacer, there are considerable interaction between the oligoether spacer chains and cations. This ionedipole interaction lowered the ionic conductivity
Compared with cationic monomers, there are many anionic monomers which can form ILs. Much wider variety of structure of anions is the reason why many anions have chance to form ILs. In other words, it is possible to find suitable cations having adequate properties to form anionic monomers. Polymerisation of ILs composed of anion monomers and cations gave polyanion type ILs in which only cations were mobile. Similar to previous study, simple polymerisation of ILs composed of anion monomers and cations was quite easy to get solid polymers, but their ionic conductivity dropped considerably as compared with that of monomeric ILs. Spacer chains were also introduced between polymerisable group and IL ion pair to improve the ionic conductivity after polymerisation. As expected, the ionic conductivity of polymers was improved 20 times higher than that of the polymers without any spacers [47]. Some ILs were added to polyelectrolytes lithium salts to form lithium ion conductive polymer gels [48]. This is a convenient method to prepare lithium ion conductive films. A key issue is the compatibility between polyelectrolyte matrix and the added ILs. Many ILs were compatible with polyanions, and homogeneously formed gels showed ionic conductivity depending on the amount of the ILs. Only 5 wt% polyanions were found to be required to form gels with ILs. Arrhenius plots of ionic conductivity were depicted on a curved line strongly suggesting that the ion conduction was governed by the diffusion of ions in a viscous matrix. As mentioned in the section of PEO/salt hybrid, organoboron polymers are prepared to fix anions on the chains. These are a kind of polyanion (Fig. 3) [24e27,49] and accordingly behave as cationconductive polymer electrolytes. Some of these mixtures of organoboron polymers and ILs showed very high lithium cation transference number of 0.87. Again, it should be mentioned here that it is quite difficult to achieve both high cation transference number and high ionic conductivity keeping sufficient mechanical strength of polymer films. For these, it should be important to design strong polymeric scaffolds containing fractions with low Tg and high affinity with ILs especially with anions. 3.3. Copolymer type ILs
Fig. 7. Structure of typical polymerisable ILs having spacers.
Copolymerisation of ILs composed of both cationic monomers and anionic monomers has also been examined. Simple copolymerisation of ILs composed of only both cationic monomers and anionic monomers mostly gave charged polymer solid that was insoluble in most solvents. The copolymers were obtained as flexible and transparent solid when IL composed of monomeric cation and typical anion and IL composed of monomeric anion and typical cation were copolymerised. For example, 3-(6-acryloyloxyhexyl)-1ethylimidazolium [Tf2N] salt ([C6eim][Tf2N]) was copolymerised with methacryloyloxy-octa(ethylene oxide)-2-sulfobenzoate lithium salt (PE8) [50]. By increasing fraction of PE8 in the copolymers, Tg was found to increase and accordingly the ionic
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
conductivity decreased. However, since PE8 contains mobile lithium cation, lithium ion transference number was found to reach higher than 0.5 by increasing PE8 fraction. It is therefore quite difficult to increase both ionic conductivity and lithium ion transference number by the copolymerisation of ILs. Alternative copolymer of imidazolium cation and borate anion was synthesised by the direct coupling reaction followed by the anion exchange from bromide anion to Tf2N anion as shown in Fig. 8 [51]. The ionic conductivity of this copolymer was about 106 S cm1 at 50 C. This conductivity is high considering relatively high Tg of this polymer (6 C).
5
synthesised and polymerised. As easily expected, ionic conductivity of pure polymers of these polymerisable zwitterions was quite low clearly showing that there are no mobile ions. However, addition of Li[Tf2N], for example, improved their ionic conductivity like in the case of low molecular weight zwitterion/Li[Tf2N] mixtures. In spite of improved ionic conductivity of the polymerised zwitterions after addition of salts, polymerised zwitterions are not suitable to design ion conductive polymers. There are other applications for these polymerised zwitterions such as selective gas permeation membranes. 3.5. Cross-linked PIL
3.4. Polyzwitterion type ILs Zwitterion is composed of tethered cation and anion. There were a few challenges to prepare zwitterions from component ions of ILs [52]. When ILs that does not migrate under potential gradient were synthesised, these ILs should be useful for electrochemical reaction media. Of course all ions are known to migrate along with the potential gradient, but once cation and anion were tethered, they should not migrate under the same condition. Zwitterionic liquids were considered to be attractive materials reflecting this idea. More than 20 different zwitterions composed of imidazolium cation and sulfonate or carboxylate anion have initially been synthesised, and their physico-chemical properties have been analysed [53]. However, their melting point was above 100 C, and all of them were solid at room temperature. It was clarified that the tethering cation and anion increased the melting point 120 C higher than that of corresponding ILs. In spite of poor properties as ILs, these zwitterions have been used as catalysts or effective additives for many reactions. We however do not go into details on the application of zwitterions here. It should be noted here that drastic increase in the ionic conductivity of zwitterions was found when Li[Tf2N] was added equimolarly to the zwitterion [52]. This increase was comprehended to the formation of IL-like low Tg environment by the coupling of imidazolium cation on the zwitterion and [Tf2N] anion. Transport of lithium cation was suggested in this mixture because couple of [Tf2N] anion and zwitterion was much less mobile than that for small lithium cation. There were also a few trials to synthesise polymerised zwitterions [52]. Zwitterions containing polymerisable vinyl groups were
Polymerised IL gels are also interesting materials. Oligo(ethylene oxide) diacrylates have frequently been used as crosslinker without elevating Tg of the polymers. Oligo(ethylene oxide) chains were used as spacer between polymerisable group and IL salt as well as cross-linker, expecting low Tg of the polymers. For example, tetra(ethylene oxide) diacrylates was added to the imidazolium-type polymerisable IL monomers, and the mixture was radical polymerised to prepare flexible and transparent crosslinked PIL film (Fig. 9) [54]. They showed high ionic conductivity of about 104 S cm1 at room temperature due to low Tg. Cross-linkers with IL units were also designed to prepare thermally stable cross-linked PILs [55]. Acryloyl groups were introduced into IL unit containing two imidazolium cations (so called Geminitype ILs), and these were added as cross-linker to the imidazolium type IL monomers. For example, 1,4-bis{3-[2-(acryloyloxy)ethyl] imidazolium-1-yl}butane bis(trifluoromethanesulfonyl)imide was used as this kind of IL cross-linker. The polymerisation gave crosslinked PILs with high ionic conductivity. The ionic conductivity decreased by increasing the amount of IL-cross-linker in the PILs due to lowering the diffusion coefficient of mobile ions in the polymer matrix. The most notable point of this study is considerably improved decomposition temperature of the PILs. Ten percent weight loss was found above 400 C when this series of cross-linked PILs were analysed. Simple polymerisation of Gemini-type IL crosslinker gave ion conductive solid with high decomposition temperature [55]. IL cross-linker was found here to be effective cross-linker to keep thermal stability of the ILs. This stabilisation cannot be realised by the polyether type cross-linkers.
Fig. 8. Alternative copolymer of imidazolium cation and borate anion.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
6
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
Fig. 9. Picture of cross-linked flexible PIL film.
On the other hand, cross-linker is not always required to prepare IL gels. There are some papers on the gel formation of ILs, for example, some amino acid ILs were effective to form gels by mixing them with some zwitterions [56]. Even mixing of some different amino acid ILs was effective to form IL gels [57]. The ionic conductivity of these ILs was maintained even after gel formation. The only drawback of these IL gels is that the mobile ions are limited to IL component ions. These gels are not suitable to transport target ions such as lithium cation. However, this is an easy-to-use method to prepare conductive gels having excellent advantage of ILs such as negligibly small vapour pressure at wide temperature range.
3.6. PILs prepared with natural polymers There are a few approaches to prepare PILs with natural polymers. Natural rubber is a representative elastic material. These rubbers are widely used in our life. There is however almost no examples to make conductive materials with these rubbers because they are typical insulator. It is difficult to add inorganic salts in natural rubber due to low polarity of the rubbers, and little increase in the ionic conductivity was expected due to little degree of dissociation of the salts if these inorganic salts were compatible with the rubbers. Sometimes this relatively low compatibility works good to prepare successive ion conduction pathways. Relatively less polar ILs and/or hydrophobic lithium salts were examined to make composites with natural rubbers [58e60]. Most ILs were found to show clear and macroscopic phase separation with these rubbers, but some ILs showed micro-phase separation with rubbers which is effective to construct successive ion conduction pathways. Macroscopically homogeneous composite with two Tg values, attributed to rubbers and additive ILs, is the proof of this kind of micro-phase separation. The PEO/salt hybrids as mentioned above were also added to natural rubbers to prepare ion conductive elastic films [58]. Nitrile rubber, reference material for natural rubber, was also mixed with ILs to improve ionic conductivity, but their affinity was not the same as the case of natural rubber [59]. Then, imidazolium-type zwitterion was added to nitrile rubber to enhance micro-phase separation and accordingly higher ionic conductivity was observed [60]. In spite of increasing demand on the conductive rubbers, there are not potential methods to achieve high conductivity. ILs and their polymers should be effective to improve conductivity of natural and synthetic rubbers.
Cellulose, typical natural polysaccharide, was also used as base polymer to prepare ion conductive polymers [61]. After designing ILs as solvents for cellulose [62e64], there are some approaches to chemical modification of cellulose directly in the ILs. Novel ion-gels bearing lithium borate were synthesised via condensation between cellulose and pentafluorophenyl boric acid in polar IL. Ion-gels obtained showed very high ionic conductivity (103 S cm1 at 90 C) comparable to the added IL itself [49]. We have also made PILs with deoxyribonucleic acid (DNA). DNA is known as a molecular memory at the genetic level. Nucleic acid bases are heteroaromatic rings, and it is easy to associate them with starting amines such as imidazole or pyridine rings which are frequently-used components for preparing ILs. Then, we have tried to make ILs with nucleic acid bases as a preliminary study. Since there are several mobile protons on these nucleic acid bases, quaternisation with alkyl halide was concluded not to be a suitable method to make cations from these nucleic acid bases. Then the nucleic acid bases were neutralised with acids [65]. Neutralisation has already been introduced to be an easy but potential method to prepare ILs [37], and this neutralisation method was again used to prepare nucleic acid base salts. After neutralisation with HTf2N, both adenine and cytosine salts were obtained as liquids. Since both guanine and thymine were weak acids, they did not form salts with strong acid. Compare with cytosine, adenine was weaker against acid treatment, further studies were carried out with cytosine. Cytosine was neutralised with some different acids, and melting point of the formed salts was found to be the function of pKa of the acids. Based on the above-mentioned reference studies, we have neutralised DNAs with HBF4 or HTf2N in an aqueous medium. DNA, from salmon milt, was dissolved in aqueous solution, and acids were added, stirred gently and dried under reduced pressure. Since the content of both adenine and cytosine in any DNA should be 50%, they would provide successive IL-like moieties along with the DNA chains. The obtained neutralised DNA was white powder. However double strand helix structure of the DNA was not maintained anymore because of breaking complementary hydrogen bonds and charge repulsion among IL-like moieties. Accordingly, low ionic conductivity was found with these neutralised DNAs. Addition of cytosine [Tf2N] salt to these neutralised DNAs enhanced the ionic conductivity only a little, because the inherent ionic conductivity of cytosine [Tf2N] salt was not so high (6.9 105 S cm1 at 50 C). Of course, addition of typical ILs to these neutralised DNAs greatly enhanced the ionic conductivity. These were obtained as transparent and flexible films to show moderate ionic conductivity. An ionic liquid domain was successfully prepared outside double-stranded DNA by fixing 1-alkyl-3-methylimidazolium cations on the phosphate anions of DNA. The counter Naþ cations of the phosphate groups of DNA were exchanged with the imidazolium cations. The resulting IL-robed DNA was soluble in ordinary organic solvents such as methanol or ethanol. Ionic conductivity of the IL-robed DNA was low, because the ion density was insufficient to form continuous IL domains around the DNA strands [66]. Addition of some ILs to this IL-robed DNA certainly improved the ionic conductivity. DNAs were confirmed to be useful scaffold materials to prepare ion conductive films. 4. Phase behaviour of PIL/water mixtures General PILs are soluble in water and show typical polyelectrolyte behaviour. By increasing hydrophobicity of the PILs, they turned insoluble in water and absorb small amount of water to form hydrated polyelectrolyte gels. What kind of phase behaviour was seen when PILs with intermediate
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
Cations
7
monomers used to design this kind of polyelectrolytes. Among many pairs, following six-different polymerisable ILs ([P4,4,4,4][SS], [P4,4,4,6][C3S], [P4,4,4,8][C3S], [P4,4,4,VB][C5S], [P4,4,4,VB][C3S], and [P4,4,4,VB][AC3S]) were confirmed to show the LCST-type phase behaviour in water. Polymerisation of these polymerisable ILs also provided polyelectrolytes showing LCST-type phase change in water [69]. Fig. 11 shows an example of the phase behaviour of the mixture of water and poly([P4,4,4,6][C3S]) having adequate hydrophobicity/hydrophilicity balance. This is the first paper on the preparation of polyelectrolytes showing LCST-type phase change in water without using well-known key monomers such as isopropylacrylamide [68]. Transmittance of the aqueous solution of this PIL showed sharp change in response to temperature [69]. From this steep change in transmittance, it is easy to determine the phase transition temperature (Tc). The Tc of these PILs was not the same as that of corresponding monomers, and further it was the function of both polymer concentration and salt concentration as seen in Fig. 12. The transition temperature was also found to be the function of degree of polymerisation. Since the transition temperature was affected by the hydrophobicity/hydrophilicity balance of monomers, it is easy to control the Tc by the composition of comonomers. The phase change is very sensitive to the temperature, and this kind of polymers should be used in many applications.
5. Other application
Anions Fig. 10. Structure of composed ions for ionic liquid monomers to show LCST type phase behaviour after mixing them with water.
hydrophobicity were mixed with water? In 2007, we found ILs to show unique phase change after mixing them with water [67]. Partly hydrophobic amino acid-derived ILs were found to be soluble with water, but they turned to be insoluble by heating. This phase change, namely lowering solubility by heating, is generally known as lower critical solution temperature (LCST)-type phase change. After analysing the required physicochemical properties of ILs to show this kind of LCST-type phase behaviour after mixing them with water, balance of hydrophilicity and hydrophobicity of ILs was found to be the dominant factor [68]. After this study, the hydrophobicity/hydrophilicity of ILs was semi-empirically controlled by selecting suitable ion pairs [69]. Mixture of water and polymerisable ILs also showed this kind of LCST-type phase change only when the hydrophobicity/hydrophilicity balances of ILs was in the proper range. Fig. 10 shows typical
Area of PIL science has been spread wider than we initially expected. Here we briefly introduce other interesting and noteworthy results on PILs reported by other scientists. In1999, [C4mim]PF6 was found to dissolve large amount of CO2 gas [70]. After this report, it is generally known that CO2 gas was considerably dissolved in many ILs as compared with other gas molecules. Then PILs were also prepared as membranes expecting selective adsorption and/or separation of CO2 gas [71e75]. This CO2 gas separation is one of big subjects of PILs now. Study on the catalytic activity of ILs was also applied to design catalytic PILs. Since polymeric catalysts are easy to treat, catalytic PILs are of great interests for recycling of catalysts as well as separation of products from catalysts. For example, PILs were reported as a catalyst for nucleophilic substitution reactions including fluorinations [76]. There are a few reports that synthesis of carbonates via cycloaddition reaction of CO2 with epoxides was catalysed by PIL [77e79]. Heavy metal ions, expected to act as a catalyst, can be fixed on PIL as counter cations. Also, counter anions of imidazolium-type PIL were exchanged with Pd complex to catalyse Sonogashira coupling reactions [80]. Other imidazolium type IL was fixed on resins to be used to catalyse Michael reaction. The reaction was carried out without solvent to get more than 90% yield with excellent enantioselectivity up to 99%. It is known that ILs are used as extraction of many target materials from aqueous media. When these ILs were
Fig. 11. Polymerisation of [P4,4,4,6][C3S] (a, b, and c) and LCST behaviour (c and d) of poly([P4,4,4,6][C3S]) after mixing it with water.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
8
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9
been supported by the Grant-in-Aid for Scientific Researches from the Japan Society for the Promotion of Science (No. 21225007, 17205020, 17073005, 14205136, 11555250, and 08246101). References [1] [2] [3] [4] [5] [6] [7] [8]
Fig. 12. Temperature-dependent change in transmittance of poly([P4446][C3S]) in pure water upon heating. Change in transmittance of poly([P4446][C3S]) after mixing with pure water. The polymer concentration was 2.5 wt% (a), 5 wt% (b), 10 wt% (c), and 20 wt% (d) respectively.
polymerised, PILs were expected to be used as adsorbents. Since these PILs are insoluble in a medium, separation of target materials turns easier than that with liquid extractants. For example, PIL was used to adsorb heavy metal ions from waste water. Imidazolium-type and pyridinium-type ILs were fixed onto bcyclodextrin, and they were polymerised to adsorb heavy metal ions [81]. The p-nitrophenol, 2,4,6-trichlorophenol, and Cr6 ion were successfully extracted from an aqueous mixture. Poly(triethylenetetramine acrylate salt) was prepared to collect petroleum from petroleum/water mixture [82]. This will relate to the recovery of valuable or target materials from waste water or polluted seawater. There should be many different approaches to collect, extract, or recover useful or target materials from mixtures with suitably-designed PILs. There are many excellent reviews on PIL researches, and readers are strongly recommended to refer them, for examples, those by Mecerreyes [83] and Yuan [84].
[9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38]
6. Conclusion Development of polymerised ionic liquids in these 15 years has been mentioned in this mini-review. At the initial stage, poly(ethylene oxide) chains were tethered to ions to lower the melting point of the salts, and these were analysed as components of polymerised ionic liquids. Report on the ionic liquids with low melting point pushed us to introduce imidazolium type ionic liquid unit on the end of poly(ethylene oxide), and at the initial stage of polymerised ionic liquids, we then realised the potential possibility of these ionic liquids on the design of ion conductive polymers. After this stage, we made efforts to design many polymerised ionic liquids. During these 15 years, number of papers on the polymerised ionic liquids increased year by year, and their applications spread widely. Polymer materials are good in designing functional materials because of their unique properties to make three dimensional domains. Use of these polymerised ionic liquids will be common in many fields in the near future. Acknowledgement The present mini-review is based on the results mainly obtained in our lab for these 15 years. Most part of these researches have
[39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64]
Walden P. Bull Acad Sci; 1914:405e22. St. Petersburg. Wilkes JS, Zaworotko MJ. J Chem Soc Chem Commun; 1992:965e7. Ohno H, Ito K. Chem Lett 1998;27:751e2. Mukai T, Yoshio M, Kato T, Ohno H. Future Materials 2006;6(2):2e5 [in Japanese]. Ohno H, editor. Electrochemical aspects of ionic liquids. John Wiley & Sons; 2011. Kishimoto K, Yoshio M, Mukai T, Yoshizawa M, Ohno H, Kato T. J Am Chem Soc 2003;125:3196e7. Hoshino K, Yoshio M, Mukai T, Kishimoto K, Ohno H, Kato T. J Polym Sci Part A Polym Chem 2003;41:3486e92. Kishimoto K, Suzawa T, Yokota T, Mukai T, Ohno H, Kato T. J Am Chem Soc 2005;127:15618e23. Ohno H, Ito K. Polymer 1995;36:891e3. Ito K, Ohno H. Solid State Ionics 1995;79:300e5. Ito K, Nishina N, Tominaga Y, Ohno H. Solid State Ionics 1996;86-88:325e8. Ito K, Tominaga Y, Ohno H. Electrochim Acta 1997;42:1561e70. Tominaga Y, Ito K, Ohno H. Polymer 1997;38:1949e51. Ito K, Nishina N, Ohno H. J Mater Chem 1997;7:1357e62. Ito K, Ohno H. Electrochim Acta 1998;43:1247e52. Tominaga Y, Ohno H. Chem Lett 1998;27:955e6. Tominaga Y, Ohno H. Solid State Ionics 1999;124:323e9. Tominaga Y, Takizawa N, Ohno H. Electrochim Acta 2000;45:1285e9. Furukawa T, Yoneya K, Takahashi Y, Ito K, Ohno H. Electrochim Acta 2000;45: 1443e8. Kato K, Ito K, Ohno H. J Solid State Electrochem 2000;4:141e5. Tominaga Y, Ohno H. Electrochim Acta 2000;45:3081e6. Tominaga Y, Mizumo T, Ohno H. Polym Adv Technol 2000;11:524e8. Mizumo T, Ohno H. Polymer 2004;45:861e4. Matsumi N, Sugai K, Ohno H. Macromolecules 2002;35:5731e3. Matsumi N, Mizumo T, Ohno H. Chem Lett 2004;33:372e3. Mizumo T, Sakamoto K, Matsumi N, Ohno H. Chem Lett 2004;33:396e7. Matsumi N, Sugai K, Sakamoto K, Mizumo T, Ohno H. Macromolecules 2005;38:4951e4. Ito K, Ohno H. Polymer 1997;38:921e6. Ohno H, Nakai Y, Ito K. Chem Lett 1998;27:15e6. Nakai Y, Ito K, Ohno H. Solid State Ionics 1998;113e115:0199e204. Fuller J, Breda AC, Carlin RT. J Electrochem Soc 1994;144:L67e70. Bonhôte P, Ana-Paula Dias, Papageorgiou N, Kalyanasundaram K, Grätzel M. Inorg Chem 1996;35:1168e78. Ohno H. Bull Chem Soc Jpn 2006;79:1665e80. Armand M, Endres F, MacFarlane DR, Ohno H, Scrosati B. Nat Mater 2009;8: 621e9. Ueki T, Watanabe M. Macromolecules 2008;41:3739e49. Lu J, Yan F, Texter J. Prog Polym Sci 2009;34:431e48. Hirao M, Sugimoto H, Ohno H. J Electrochem Soc 2000;147:4168e72. Yoshizawa M, Ogihara W, Ohno H. Electrochem Solid State Lett 2001;4:E25e 7. Belieres J, Angell CA. J Phys Chem B 2007;111:4926e37. Ohno H. Electrochim Acta 2001;46:1407e11. Yoshizawa M, Ohno H. Chem Lett 1999;28:889e90. Hirao M, Ito K, Ohno H. Electrochim Acta 2000;45:1291e4. Hirao M, Ito-Akita K, Ohno H. Polym Adv Technol 2000;11:534e8. Ohno H, Yoshizawa M, Ogihara W. Electrochim Acta 2004;50:255e61. Yoshizawa M, Ohno H. Electrochim Acta 2001;46:1723e8. Ogihara W, Washiro S, Nakajima H, Ohno H. Electrochim Acta 2006;51:2614e 9. Yoshizawa M, Ogihara W, Ohno H. Polym Adv Technol 2002;113:589e94. Ogihara W, Sun J, Forsyth M, MacFarlane DR, Yoshizawa M, Ohno H. Electrochim Acta 2004;49:1797e801. Matsumi N, Nakamura Y, Aoi K, Watanabe T, Mizumo T, Ohno H. Polym J 2009;41:437e41. Ogihara W, Suzuki N, Nakamura N, Ohno H. Polym J 2006;38:117e21. Narita A, Shibayama W, Matsumi N, Ohno H. Polym Bull 2006;57:109e14. Yoshizawa M, Hirao M, Ito-Akita K, Ohno H. J Mater Chem 2001;11:1057e62. Yoshizawa M, Narita A, Ohno H. Aust J Chem 2004;57:139e44. Washiro S, Yoshizawa M, Nakajima H, Ohno H. Polymer 2004;45:1577e82. Nakajima H, Ohno H. Polymer 2005;46:11499e504. Taguchi S, Ichikawa T, Kato T, Ohno H. RSC Adv 2013;3:23222e7. Kagimoto J, Nakamura N, Kato T, Ohno H. Chem Commun; 2009:2405e7. Yoshizawa M, Marwanta E, Ohno H. Polymer 2000;41:9049e53. Marwanta E, Mizumo T, Nakamura N, Ohno H. Polymer 2005;46:3795e800. Marwanta E, Mizumo T, Ohno H. Solid State Ionics 2007;178:227e32. Fukaya Y, Sugimoto A, Ohno H. Biomacromolecules 2006;7:3295e7. Fukaya Y, Hayashi K, Wada M, Ohno H. Green Chem 2008;10:44e6. Ohno H, Fukaya Y. Chem Lett 2009;38:2e7. Abe M, Fukaya Y, Ohno H. Green Chem 2010;12:1274e80.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042
N. Nishimura, H. Ohno / Polymer xxx (2014) 1e9 [65] Nishimura N, Ohno H. J Mater Chem 2002;12:2299e304. [66] Nishimura N, Nomura Y, Nakamura N, Ohno H. Biomaterials 2005;26:5558e 63. [67] Fukumoto K, Ohno H. Angew Chem Int Ed 2007;46:1852e5. [68] Kohno Y, Ohno H. Aust J Chem 2012;65:9194e7. [69] Kohno Y, Deguchi Y, Ohno H. Chem Commun 2012;48:11883e5. [70] Blanchard LA, Hancu D, Beckman EJ, Brennecke JF. Nature 1999;399:28e9. [71] Tang J, Tang H, Sun W, Radosz M, Shen Y. Polymer 2005;46:12460e7. [72] Bara JE, Noble RD, Gin DL. Ind Eng Chem Res 2009;48:4607e10. [73] Carlisle TK, Bara JE, Lafrate AL, Gin DL, Noble RD. J Membr Sci 2010;359:37e 43. [74] Hudiono YC, Carlisle TK, Lafrate AL, Gin DL, Noble RD. J Membr Sci 2011;370: 141e8. [75] Carlisle TK, Nicodemus GD, Gin DL, Noble RD. J Membr Sci 2012;397e398:24e37.
9
[76] Kim DW, Chi DY. Angew Chem Int Ed 2004;43:483e5. [77] Xie Y, Zhang Z, Jiang T, He J, Han B, Wu T, et al. Angew Chem Int Ed 2007;46: 7255e8. [78] Sun J, Cheng W, Fan W, Wang Y, Meng Z, Zhang S. Catal Today 2009;148:361e 7. [79] Yu J, Choi H, Selvaraj M, Park D. Reac Kinet Mech Cat 2011;102:353e65. [80] Wang Y, Liu J, Xia C. Tetrahedron Lett 2011;52:1587e91. [81] Mahlambi MM, Malefetse TJ, Mamba BB, Krause RW. J Poly Res 2010;17:589e 600. [82] Asadov ZH, Rahimov RA, Poladova TA, Nasibova SM, Guliyev AD, Asadova AZ. J Mol Liq 2012;166:44e8. [83] Mecerreyes D. Prog Polym Sci 2011;36:1629e48. [84] Yuan J, Mecerreyes D, Antonietti M. Prog Polym Sci 2013;38:1009e36.
Please cite this article in press as: Nishimura N, Ohno H, 15th anniversary of polymerised ionic liquids, Polymer (2014), http://dx.doi.org/ 10.1016/j.polymer.2014.02.042