J. Phys. Chem. B 2009, 113, 2725–2733
2725
Molecular Mechanism for Formation of Polyaniline Lamella from a Lyotropic Liquid Crystal: An NMR Study Li Shi, Xiaodong Wu,* Lude Lu, Xujie Yang, and Xin Wang* Key Laboratory for Soft Chemistry and Functional Materials, Nanjing UniVersity of Science and Technology, Ministry of Education, Nanjing 210094, People’s Republic of China ReceiVed: January 12, 2009
Polyaniline (PANI) microlamellas with an average interlamellar distance of 2.6 nm were prepared from a nematic lyotropic liquid crystal system composed by sodium dodecyl sulfate (SDS) aqueous solution. To reveal the formation mechanism of these lamellas, a series of NMR studies have been performed. At first, variable-temperature (VT) 13C NMR experiments have suggested that, prior to polymerization, anilines are predominantly located in the vicinity of the SDS polar head region with a limited mobility at low temperature, whereas they become more mobile and penetrate into the SDS hydrophobic domain at elevated temperature. Subsequent in situ 13C NMR measurements at 310 K have indicated that the overall polymerization can be taken place in two stages. In the beginning, the reaction sites are within the SDS micelles, resulting in the formation of oligomeric PANI species with benzenoid and quinoid structures. Interestingly, these oligomeric species fall off from the micellar hydrophobic domains and reorganize into layered structures with the support of SDS. In the second stage, further polymerization can be continued within the interlayers. This paper provides a good example in studying the roles of surfactants at the nucleation stage qualitatively during the synthesis of morphology-specific polymers with the application of NMR techniques, a period difficult to be examined by other approaches currently. Introduction Over the past few decades, the synthesis and application of inorganic nanomaterials has aroused intense research interest worldwide because of their significantly different properties relative to their bulk counterparts.1 For the same reason, studies on size-dependent polymers with well-defined morphology have become another hot topic recently.2 For example, PANIs with morphologies such as nanotubes,3 nanowires,4 and nanorods5 have been prepared successfully. One most general procedure for the synthesis of these materials involves the application of surfactant molecules which build up a special reaction environment due to their capability to aggregate into a large family of mesophases in aqueous solutions, ranging from micelles at low concentration to lyotropic liquid crystalline (LLC) states at a higher one.6,7 However one awkward situation in contemporary research is the poverty in understanding the underlying mechanism for the formation of each individual morphology, as a result most current synthesis still staying at a “trial and error” stage or, at best, a combinatorial multidimensional parameter optimization. Up to now, a widely accepted proposal for the synthesis of nanomaterials involves three stages: nucleation, crystal growth, and precipitation, following the sequence of events.8,9 From the viewpoints of thermodynamics and kinetics, the first period is speculated to be the most important one since it defines the forthcoming evolution pathways. However nearly no work on this stage has been reported, even though some primary results have already been obtained for the latter two.10,11 Major reasons for such discrepancy can be attributed to the multiplicity and complexity of this stage and also to the shortage of the appropriate techniques to characterize this multicomponent * Towhomcorrespondenceshouldbeaddressed.E-mail:
[email protected] (X.Wang);
[email protected].
system. Specifically, techniques such as various microscopes, suitable for studying the later stages, cannot be applicable to the first one when the overall system is still in the molecular level; no microscopically observable aggregates have been produced. On the other hand, although spectroscopic approaches can be employed, severe signal overlapping and the complicated multibody interactions render the spectral interpretation, in most cases, extremely difficult. Recently we have found that NMR is an appropriate technique to investigate the mechanism of the nucleation stage, in particular the organic nanomaterials in LLC systems, on the basis of the following considerations: (1) Most importantly, the liquid crystalline state is an anisotropic solution, which will not average out the chemical shift tensor due to the molecular motions, so generally three chemical shift tensors σxx, σyy, and σzz can be observed. If the system further becomes more solidlike, a characteristic powder pattern can be detected, that is, the peak width widened broadly, and also the line shapes of the peaks depend strongly on their orientation, in particular their 13 C-1H dipolar interaction tensors and 13C chemical shift tensors. Accordingly the chemical shift anisotropy (CSA) provides a means to evaluate the shape and orientation of the corresponding liquid crystal system.12 (2) One interesting feature for the liquid crystalline state is that when any molecule dissolves in this system, it is inclined to be orientated, as a result; a careful analysis of the chemical shift anisotropy of this molecule provides information on its orientation in liquid crystal solutions.13-15 (3) Another essential feature for this paper is to select 13C NMR technique instead of 1H NMR. The reason for this is that the strong dipolar couplings between the protons prevent obtaining a well-resolved spectrum, in contrast taking into account that 13C is a rare nuclei and their dipolar interactions will be dropped several magnitudes relative to proton interactions, so that well-resolved 13C NMR can be obtainable. (4) It
10.1021/jp9002824 CCC: $40.75 2009 American Chemical Society Published on Web 02/11/2009
2726 J. Phys. Chem. B, Vol. 113, No. 9, 2009 is worth noting that protons are decoupled in all experiments in order to improve the 13C NMR sensitivity via nuclear Overhauser effect; unfortunately one drawback for such experiments is the inability to provide quantitative information on each reactive species. (5) NMR technique is a powerful approach in studying molecular structures, molecular mobilities, and intermolecular interactions. A great variety of one- (1D), two- (2D), and even high-dimensional NMR approaches have been developed, which not only simplify the spectral interpretation but also provide the possibility to investigate a system from different aspects of eyesight. (6) Being a kind of nondestructive method, NMR experiment can be used in situ to monitor the whole process whenever kinetically allowed. We have found that anilines are inserted into sodium dodecyl sulfate (SDS) micellar at some time, and consequently protons in SDS can improve the sensitivity of the 13C nuclei in aniline through heteronuclear Overhauser effect significantly. It is only for this reason that the utilization of in situ 13C NMR to investigate the nucleation mechanism for organic nanospecies in the LLC system becomes not only sensitive but also informative. Recent state of art methods such as multinuclear NMR experiments involving 2H NMR,16 23Na NMR,17 and 129Xe NMR,18 etc., and multiple dynamic NMR experiments ranging from spin-spin and spin-lattice relaxations19 to self-diffusion experiment20 have been carried on the aggregates of the surfactant molecules. Furthermore novel solid-state NMR techniques have also been applied to this system at either magic angle21 or other manipulated angles22 have been performed. Harvested information about the orientations, molecular interactions, and the mobilities of the surfactant molecules or their aggregates has been obtained. Unfortunately all these methods currently are only focused on the surfactant molecules; nearly no attention has been paid to both surfactants and the growing nanospecies at the same time. In this paper, we first present the synthesis of PANIs with lamella morphology from a nematic discotic liquid crystalline solution (self-organized by SDS in aqueous solution). Then, a series of VT 13C NMR and in situ 13C NMR experiments have been performed to reveal the formation mechanism for these morphology-specific PANIs after combining the information from both the surfactant molecules and the growing nanospecies. Finally a plausible mechanism is proposed, and the application of our results to other similar systems is also discussed. Experimental Section Materials. Sodium n-dodecyl sulfate (99%) was purchased from Alfa Aesar and used as received. 1-Decanol and ammonium persulfate (APS) were obtained from Sinopharm Chemical Reagent Co. Ltd. Sodium sulfate and aniline were provided by Nanjing Chemical Reagent Co. Ltd. and Shanghai Lingfeng Chemical Reagent Co. Ltd., respectively. Sodium sulfate was heated in an oven at 350 °C for 1 day to remove hydrogenated water, and APS was recrystallized with water. Aniline and 1-decanol were distilled under reduced pressure. Deuterium oxide was purchased from Beijing Chemical Reagents Co. Ltd. Triply distilled water was used throughout the sample preparation. Methods. Nematic discotic (ND) lyotropic liquid crystalline was prepared by mixing SDS (3.5 g), sodium sulfate (0.29 g), 1-decanol (0.68 g), and water (5.53 g) in a 100 mL three-neck round-bottom flask with pennyhead stoppers at 25 °C under continuous magnetic stirring for 48 h. Subsequently, the freshly distilled aniline (0.3 g) was added dropwise under nitrogen atmosphere. The system was kept stirring slowly for an
Shi et al. overnight to allow complete dissolution of aniline in nematic lyotropics. Prior to polymerization, the solution was left standing still for about 2 h. Then 0.74 g of APS was introduced into the above system by slowly stirring in 2 h, followed by slow magnetic stirring for another 48 h. Finally the product was washed by distilled water ten times and then by ethanol and acetone three times separately to remove small molecular species, and dried under vacuum for 12 h to obtain a dark blue powder. The X-ray diffraction (XRD) patterns were recorded on a Bruker D8 advanced X-ray diffractometer, using Cu KR radiation (λ ) 0.1542 nm) with the range of the diffraction angle of 2θ ) 2.0-50°. Scanning electron microscopic (SEM) images were obtained on a JSM-6380 scanning electron microscope (JEOL). Elemental analysis was performed on a CHN-O-Rapid Elemental Analyzer (Heraeus). NMR Measurements. All NMR experiments were carried out on a Bruker DRX 300 spectrometer (300.13 MHz for 1H and 75.47 MHz for 13C) with a 5 mm dual probehead equipped with variable-temperature accessories. All spectra were required with a free induction decay (FID) composed of 32K data points with a power gated decoupling pulse sequence. The spectral width for 13C is 220 ppm, with a 90° tilted angle of 4.3 µs, and a relaxation delay of 3 s. A WALTZ16 broad-band decoupling method was applied on proton channel, and the transient scans for all samples at varied temperature are 2 K. During the sample preparation period, a total volume of 0.5 mL of pure nematic SDS lyotropics with a composition of 35% (w/w) SDS, 6.8% (w/w) 1-decanol, 2.9% (w/w) Na2SO4, and 55.3% (w/w) D2O was first filled in an NMR tube and mixed well at 25 °C. To investigate the effect of temperature on the structure of SDS lyotropics, a VT 13C NMR in the range of 295-315 K was performed, with an increment of 5 K at each step. Subsequently the interaction of aniline and SDS as a function of temperature with a relative concentration of 1, 3, and 30% (aniline/SDS (w/w)) was evaluated by similar approaches. As a contrast, a VT experiment was carried out for 1-decanol also. During all VT experiments, the solution was equilibrated for 30 min at each desired temperature before data acquisition. To avoid the radio frequency mismatch induced by the temperature variation, the probe was tuned and matched before each experiment. Next, an in situ experiment was designed at which the uniformly dispersed aniline in SDS lyotropics was polymerized in an NMR tube by addition of APS, and a series of 13C NMR spectra were collected as a function of the reaction time. The in situ monitoring was carried out at 295 and 310 K, respectively. During the in situ experiment, the initial period (the first six spectra), the transients were 256 in an effort to collect all necessary information in the nucleation period, whereas the transients were increased to 2 K due to limited sensitivity at later stages. The experimental parameters are identical for in situ experiment at 295 K as those in 310 K except the transients are 512 at each temperature. To determine the spatial correlation between protons in SDS and those in aniline, a standard phase-sensitive NOESY pulse sequence was applied; the 90° tilt angle for this experiment is 7.9 µs, with a mixing time of 500 ms, using a watergate pulse sequence to suppress the water signal, recorded at 305 K for a mixture of aniline and a discotic nematic LLC with a relative concentration of 3% (w/w). Results and Discussion Morphology of the Prepared PANI. Figure 1 illustrates the XRD pattern of the prepared PANI. One sharp reflection at about
PANI Lamella from a Lyotropic Liquid Crystal
J. Phys. Chem. B, Vol. 113, No. 9, 2009 2727 SCHEME 1: Labeling of Carbons in SDS
Figure 1. XRD pattern of PANI prepared from SDS discotic nematic LLC.
3.4° and a wide broad one centered at about 20° can be observed. Sharp signals at small angles in XRD pattern generally refer to lamellar or porous structures, and therefore SEM was employed to differentiate the exact one. As seen clearly in Figure 2a, the product is in lamella morphology, with an average length of 6 µm and a width of 4 µm. All layers are aligned in parallel to each other. The interlamellar distance was calculated to be 2.6 nm, nearly twice the length of one SDS molecule, according to the small diffraction angle in the XRD pattern. An energy dispersive X-ray analysis (EDX) technique based on the SEM image was used to characterize the as-prepared sample. Figure 2b displays the EDX spectrum, and signals from C, H, N, and S were observed. The EDX spectrum recorded at a larger sample area provides information on the average elemental composition of the product. The result shows that the average weight ratios of C, H, N, and S are 65.84, 22.27, 10.16, and 1.27%, respectively. To have a more accurate result, the elemental analysis was performed and the compositions for C, H, N, and S are 62.91, 5.68, 8.21, and 4.94%, accordingly. Combining all information obtained right now, we speculate that SDS might have been involved in producing PANI lamellas. To have a clear understanding of the roles of SDS in forming these morphology-specific PANI, a systematic VT 13C NMR and in situ 13C NMR experiments were carried out. It should be noted that identical PANI lamellas in doped states (with camphor sulfuric acid as the dopant) have also been prepared in the same way; however, herein aniline was selected as the probe molecule to investigate the overall process in order to prevent the strong signal overlapping between camphor sulfuric acid and SDS in 13C NMR experiment. Aggregation Behavior of SDS LLC Solution. As the first step, we studied the features of the reaction environment (SDS lytropic system). The discotic nematic SDS solution was prepared according to Labes and co-workers’ results.23 To evaluate the effect of temperature on the aggregation behavior of SDS, a VT 13C NMR experiment, in the range of 295-315 K, was carried out. The labeling of each carbon in SDS and the recorded 13C NMR spectra are shown in Scheme 1 and Figure 3, respectively. The assignments of each carbon signal are straightforward, and the resonances centered at about 70, 32, 30, 26, 23, and 15 ppm can be assigned to Ca, Cb, Cc, Cd, Ce, and Cf, respectively (all assignments are based on Figure 3a). It can be seen that all peaks for SDS in discotic nematic state at low temperatures (
