pubs.acs.org/Langmuir © 2010 American Chemical Society
Evidence of Solvent-Gelator Interaction in Sugar-Based Organogel Studied by Field-Cycling NMR Relaxometry Michal Bielejewski and Jadwiga Tritt-Goc* Institute of Molecular Physics, Polish Academy of Sciences, ul. M. Smoluchowskiego 17, 60-179 Pozna n, Poland Received July 6, 2010. Revised Manuscript Received September 21, 2010 The dynamics of bulk toluene and toluene confined in the 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose gel was studied using 1H field-cycling nuclear magnetic resonance relaxometry. The proton spin-lattice relaxation time T1 was measured as a function of the magnetic field strength and temperature. The observed dispersion in the frequency range 104-106 Hz for the relaxation rate of toluene in the gel system give evidence of the interaction between the toluene and the gelator aggregates. The data were interpreted in terms of the two-fraction fast-exchange model. Additionally it was also shown that a cooling rate during gel preperation process influences the gel microstructure and leads to different gelator-solvent interactions as reflected in a different behavior of the proton spin-lattice relaxation rate of toluene within the gel observed at the low frequency range.
Introduction Physical gels made by low molecular weight organogelators (LMOG) are unique systems which consist of two phases: a solid one which basically is the gelator and liquid one which can be any sort of a solvent. The only forces which are responsible for gel formation are noncovalent interactions such as hydrogen bonding, aromatic π-π stacking, van der Waals forces, ionic or organometallic coordination bonding, or a combination of these. Any of these interactions between the gelator molecules lead to their aggregation promoting one-dimensional growth which in turn leads to the formation of self-assembled fibrillar networks and immobilizes the liquid molecules.1-5 Most physical gels formed by LMOGs are thermally reversible. The organogels represent an important class of functional materials with potential applications in template materials, biomimetics, as viscosity modifiers in applications such as paints, coatings, oil recovery, in controlled drug release and in a variety of pharmaceutical and hygienic applications.1,3,4 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose is a representative of physical sugar-based gelators. It is one of the simplest, smallest and the most efficient gelator which possesses excellent gelator properties over a broad spectrum of organic solvents.5-7 Thanks to the hydroxyl groups attached to the furanose ring and hydrophilic part the gelator molecule has the ability to form onedimensional intramolecular hydrogen-bond networks in the solid state. The gels formed by 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose with benzene, toluene, chlorobenzene, and nitrobenzene *To whom correspondence should be addressed. Phone: þ48-61 8695-226. E-mail:
[email protected]. (1) Weiss, R. G.; Terech, P. Molecular Gels, Materials with Self-Assembled Fibrillar Network; Springer: Dordrecht, The Netherlands, 2006. (2) Geiger, C.; Stanescu, M.; Chen, L.; Whitten, D. G. Langmuir 1999, 15, 2241– 2245. (3) Terech, P.; Weiss, R. G. Chem. Rev. 1997, 97, 3133–3159. (4) Gronwald, O.; Snip., E.; Shinkai, S. Curr. Opin. Colloid Interface Sci. 2002, 7, 148–156. (5) Luboradzki, R.; Pakulski, Z; Sartowska, B. Tetrahedron 2005, 61, 10122– 10128. (6) Luboradzki, R.; Pakulski, Z. Tetrahedron 2004, 60, 4613–4616. (7) Luboradzki, R.; Gronwald, O.; Ikeda, M.; Shinkai, S.; Reinhoudt, D. N. Tetrahedron 2000, 56, 9595–9599. (8) Tritt-Goc, J.; Bielejewski, M.; Luboradzki, R.; Łapinski, A. Langmuir 2008, 24, 534–540.
Langmuir 2010, 26(22), 17459–17464
were the subject of our studies by different methods.8-11 The FT-IR spectroscopy revealed that the creation of hydrogen bonds between gelator molecules is the main driving force for the selfaggregation process and the gel formation. In our work we were particularly interested in the role of the solvent in organogel formation. The nature of the interaction between the solvent and the gelator is one of the most interesting questions in the study of organogels. It is well-known that a given gelator gelatinizes certain solvents better than others but the precise role of the organic solvent in determining the macroscopic properties of the gel and network is still not clear. Furthermore the way the gel is prepared can affect its physical properties. We have shown that 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose gels, depending on the solvent, are characterized by different hydrogen-bonding patterns, which are reflected in a different microstructure of the networks. The thermal stabilities of the studied gels also differ and are dominated by the polarity of the solvent.8-11 Despite many different studies of gels made by LMOG the question about the interaction of the solvent molecules with gelator aggregates within the gel is still open. In this paper we present the proton spin-lattice relaxation measurements of bulk toluene and toluene within the 1,2-O-(1ethylpropylidene)-R-D-glucofuranose gel in the function of magnetic field B0 (so-called nuclear magnetic resonance dispersion (NMRD) profiles) and temperature. The field-cycling technique12,13 is especially suitable to study the solvent-surface interaction in porous media. The studied gel consists of a 3D fibrillar network which entraps a huge amount of toluene molecules in the spaces (pores) within the network and can be treated like a porous material. The frequency dependence of the spin-lattice relaxation of liquids confined in the porous materials depends strongly on the adsorption properties of the surface and liquid. The polar (9) Bielejewski, M.; Łapinski; Kaszynska, J.; Luboradzki, R.; Tritt-Goc, J. Tetrahedron Lett. 2008, 49, 6685–6689. (10) Bielejewski, M.; Łapinski, M.; Luboradzki, R.; Tritt-Goc, J. Langmuir 2009, 25, 8274–8279. (11) Rachocki, A.; Bielejewski, M.; Tritt-Goc, J. Appl. Magn. Reson. 2009, 36, 61–68. (12) Noack, F. Prog. NMR Spectrosc. 1986, 18, 171–276. (13) Kimmich, R.; Anoardo, E. Prog. NMR Spectrosc. 2004, 44, 257–320.
Published on Web 10/05/2010
DOI: 10.1021/la103324s
17459
Article
Bielejewski and Tritt-Goc
molecules on the polar surface will be more strongly adsorbed than the nonpolar molecules and thus the “weak” and “strong” adsorption cases can be distinguished.14 Weak adsorption is usually revealed by flat and strong adsorption by steep, lowfrequency dispersions of spin-lattice relaxation. The studied glucofuranose gelator, like other carbohydrates, creates a polar surface. Toluene is only slightly polar when compared to a strongly polar solvent like water and alcohol. Toluene is characterized by the dielectric constant ε = 2.4, the dipole moment μ=0.36 D and the polarity parameter ET(30)=33.9 (kcal mol-1). The dielectric constant is related to the macroscopic and microscopic properties of the solvent describing the “solvent polarity”.15 The polarity parameter describes the solvent-gelator interactions. Therefore the studied gel is composed from a slightly polar solvent-toluene confined in the pore with polar surface of the glucofuranose aggregates. The interactions between them lead, as will be shown in this paper, to weak spin-lattice relaxation dispersion of toluene observed for the low-frequency range. Moreover it will be also shown that the cooling rate used in the gel preparation influences the interactions between the gelator and solvent molecules and modifies the microstructure of the system as reflected in the images taken by Optical Polarization Microscopy
Experimental Section Preparation of 1,2-O-(1-ethylpropylidene)-r-D-glucofuranose Toluene Based Gel. The gealator was synthesized by the
Figure 1. Proton-spin lattice relaxation rates of degassed (filled circles) and non degassed (open circles) bulk toluene measured at 223 K as a function of the proton Larmor frequency (bottom) and electron Larmor frequency (top). The solid line through the open circles presents the best fit of eq 2 to the experimental points. The dashed lines were simulated on the base of data from Powels and Neale17 work as discussed in the text. The diamond symbols are the experimental values of the relaxation times and the star symbols are the theoretical ones computed on the base of Powels and Neale17 work.
method described in refs 6 and 16. Concentrations of 2% [g/mL] of the gelator were chosen to form the gel with toluene used in this study. The gel is prepared by mixing the appropriate amount of the gelator with a toluene in a closed capped tube and next heating the mixture until it becomes a transparent solution. Next, cooling the solution, with appropriate rate so-called cooling rate, below the characteristic gelation temperature Tgel brings out the transition to the gel phase. As a result a thermoreversible, optically clear and transparent gel is obtained. To check the influence of the cooling rate on the properties of the created physical gel two gels were prepared at cooling rate of 5 (gel 1) and 15 K min-1 (gel 2), respectively. To the best of our knowledge no data on this subject concerning LMOG gels were published. NMR Measurements. Proton spin-lattice relaxation measurements in the studied gel, in the function of magnetic field B0, were performed on field-cycling relaxometer (STELAR, Italy) covering proton frequencies from 10 kHz to 40 MHz. The spectrometer operates by switching current in solenoidal magnet from a polarizing field (BPOL) with a 1H Larmor frequency of 24 MHz to a field of interest (BRELAX) for a variable relaxation period (τ) after which the field is switched to a 1H Larmor frequency of 16 MHz (BACQ) where the magnetization is detected by a τ-π/2 pulse sequence. The switching time was 3 ms. Free Induction Decay signal was measured as a function of the delay time τ at the field of interest BRELAX. From the recovery or decay of the magnetization the relaxation time T1 was calculated. Within the experimental errors, all relaxation curves could be described by monoexponential functions over about 3 decades of the signal amplitude. The measurements were carried out in the temperature range from 183 to 293 K. The measured NMR signal comes only from the toluene protons within the gel. The contribution of the gelator aggregates protons which form the rigid, solid gel region are undetectable under our NMR measuring conditions.
where i is either r, which refers to the ring protons, or m, which referes to the methyl protons, τci is the correlation time of molecular motion responsible for spin-lattice relaxation, and Ci is the spin-lattice coupling parameter. The correlation time for
(14) Bychuk, O. V.; O’Shaughnessy, B. J. Chem. Phys. 1994, 101, 772–780. (15) Katritzky, A. R.; Fara, D. C.; Yang, H.; Tamm, K. Chem. Rev. 2004, 104, 175–198. (16) Sakurai, G.; Jeong, Y.; Koumoto, K.; Friggeri, A.; Gronwald, O.; Sakurai, S.; Okamoto, S.; Inoue, K.; Shinkai, S. Langmuir 2003, 19, 8211–8217.
(17) Powels, J. G.; Neale, D. J. Proc. Phys. Soc. 1960, 3, 737–747. (18) naNagara, B.; O’Connor, R. D.; Blum, F. D. J. Phys. Chem. 1992, 96, 6423– 64427. (19) Wilbur, J. D.; Jonas, J. J. Chem. Phys. 1975, 62, 2800–2807. (20) Spiess, H. W.; Schweitzer, D.; Haberlen, U. J. Magn. Reson. 1973, 9, 444– 460.
17460 DOI: 10.1021/la103324s
Optical Polarization Microscopy Measurements. For imaging, a 2% [g/mL] gel samples were carefully placed at the center of a glass microscope slide. A JENAPOL microscope operating in differential interference contrast and polarization modes was used.
Results and Discussion Proton Spin-Lattice Relaxation Dispersion of Bulk Toluene. Proton spin-lattice relaxation rates T1-1 for degassed and non degassed bulk toluene are presented at 223 K in Figure 1 as a function of magnetic field strength plotted as the Larmor frequency. The observed increase of the relaxation rate (decrease of the spin-lattice relaxation time) in oxygenated toluene as compared to degassed toluene is identified with the paramagnetic contribution from molecular oxygen. The previous NMR relaxation studies17-20 have shown that for bulk toluene there are two relaxation pathways due to the reorientation of the methyl group and of toluene ring. The general form of the Blombergen, Purcell and Pound (BPP) equation applies for both. The observe relaxation time is given by X 1 τci 4τci ¼ Ci þ T1 1 þ ω0 2 τci 2 1 þ 4ω0 2 τci 2
ð1Þ
Langmuir 2010, 26(22), 17459–17464
Bielejewski and Tritt-Goc
the methyl group is described by the Arrhenius law: τcm = τ0m exp(Eam/RT) whereas for the toluene ring takes the form τcr=(h/ kT) exp(Ear/RT þ S/R). Eam is the activation energy for methyl reorientations, Ear and S are the activation energy and entropy of the thermally activated motion of toluene ring, respectively. On the base of the temperature dependence of T1-1 relaxation measurements of toluene at 47.5 MHz from the work of Powels et al.17 the frequency dependencies of relaxation rate of toluene were simulated in a low frequency range with the activation energies values Eam = 6.0 kJ mol-1 and Ear = 7.7 kJ mol-1. The obtained data are shown in Figure 1 in the form of dashed lines. The bold dashed line corresponds to the resultant dispersion profile whereas the other two dashed lines reflect the contribution from the protons of the ring and the protons of the methyl group of toluene, respectively. The experimental values of the relaxation rates from Powels et al.17 work are marked on the lines by the diamond symbols whereas the star symbols are the theoretical values of the corresponding relaxation rates computed on the base of this work for particular frequencies. The obtained NMRD profiles are well approximated by the Lorentzian shape. On the basis of these results we analyzed the relaxation data obtained for our oxygenated and degassed toluene samples (the open and filled circles in Figure 1) in the studied frequency range. In the absence of oxygen, the bulk toluene proton spin-lattice relaxation rate is almost independent of the magnetic field over the range of fields studied. For oxygenated toluene the frequency at which the maximum dispersion occurs is shifted to a lower value and is within the studied frequency range at about 10 MHz. The observed shift is equivalent to the increase of the correlation time and together with the increase of relaxation rate as compared to degassed toluene, is identified with the additional interaction introduced by the oxygen in the non degassed sample. Therefore, the paramagnetic contribution has to be included in order to fit the experimental relaxation dispersion profile for the oxygenated sample. This contribution is expected to be dominated by the relative translational motion of the oxygen molecules and the toluene protons. The appropriate theories have been developed21,22 and the simplified equation for the relaxation can take the following form:23 1 τs þB ð2Þ ¼ A T1 1 þ ωs 2 τ s 2 where A and B are constants, ωs is the electron Larmor frequency, and τs is the correlation time for the electron-nuclear coupling. The first term in eq 2 includes the dispersion from the electron Larmor frequency whereas constant B is the contribution connected with the nuclear Larmor frequency. In our case B corresponds, in the studied frequency range, to the constant vale of the spin-lattice relaxation rate for degassed toluene (T1b-1) and is equal 2.9 s. The constant A includes the magnetogyric ratios of the proton and the electron, the oxygen concentration, the intermoment distance, and the usual physical constants. Equation 2 was used to fit the data for oxygenated sample. The solid lines through the open circles shown in Figure 1 present the best fit with the fitting parameters τs and A, correspondingly equal 5.4 10-12 s and 6.1 1010. The obtained electron spin relaxation parameters for oxygenated toluene compare well with those obtained for other oxygenated solution of organic solvents.23 The bulk toluene does not show any relaxation rate dispersion in the frequency range 104 to 4 106 Hz. The observed frequency (21) Hwang, L. P.; Freed, J. H. J. Chem. Phys. 1975, 63, 4017–4045. (22) Freed, J. H. J. Chem. Phys. 1978, 68, 4034–4037. (23) Teng, Ch.L.; Hong, H.; Kihne, S.; Bryant, R. G. J. Magn. Reson. 2001, 148, 31–34.
Langmuir 2010, 26(22), 17459–17464
Article
Figure 2. Proton-spin lattice relaxation rates of toluene in 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose gel (1, gel created with cooling rate 5 K min-1; 2, gel created with cooling rate 15 K min-1) measured as a function of the magnetic field strength at 223 K. The solid line through the open squares and triangles presents the best fit of eq 4 to the experimental points for gels 1 and 2, respectively. Three dashed lines reflect the contribution from the bulk toluene, and the motionally altered fraction of toluene (ring and CH3 protons) in gel 1. Two dotted lines reflect the contribution from the ring and CH3 protons of motionally altered fraction of toluene in gel 2.
dependence of toluene in the gel system is therefore attributed to toluene-gelator interactions. Proton Spin-Lattice Relaxation Dispersion of Toluene in Gel. Figure 2 presents NMRD profiles of the relaxation rate for toluene in 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose gels. The open squares and triangles represent the relaxation rates of the 2% [g/mL] gel prepared at cooling rates of 5 and 15 K min-1 for gels 1 and 2, respectively. The gels were prepared with the oxygenate toluene whose dispersion curve (open circles) is also shown in Figure 2 for comparison. It is clearly seen that in contrast to bulk toluene the T1-1 of toluene in gels smoothly decreases with increasing frequency. This fact indicates the presence of a molecular dynamic process with characteristic frequency in the kHz range not found in the bulk toluene. Therefore, the observed dispersion for the relaxation rate of toluene in the gels system, in the frequency range 104-106 Hz, is solely due to the interaction between the toluene and the pore surface of gelator aggregates which formed the solid fibrillar networks in the gel. Inspection of Figure 2 also shows that gel 2 prepared with a cooling rate of 15 K min-1 exhibits more pronounced dispersion as compared to gel 1 prepared with 5 K min-1. This means that the toluene-gelator interactions somehow depend also on the way in which the gel was created. The solvent-solid gel matrix interaction is fundamental to understanding the observed magnetic field dependence of the spin-lattice relaxation rate. To explain this phenomenon we may consider two possibilities. The dispersion profile of toluene in the gel may reflect the magnetic field dependence of the gelator aggregates which is transferred to the toluene thanks to the magnetic cross-relaxation24,25 between the toluene and the gelator (24) Edzes, H. T.; Samulski, E. T. J. Magn. Reson. 1978, 31, 207–229.
DOI: 10.1021/la103324s
17461
Article
Bielejewski and Tritt-Goc
aggregates. On the other hand, the dispersion data can be interpreted in terms of the two-fraction fast-exchange model26 which assumes that there are two states or fraction of solvent molecules, one which is next to the gelator aggregates (motionally altered fraction) and the other that behaves like bulk solvent. The more complicated models like the reorientation mediated by translational diffusion (RMTD) and bulk mediated surface diffusion (BMSD)13,14 were not considered because there are experimental evidence that the combination of these models leads to the efficient spin-lattice relaxation mechanism only in the strong absorption case where the polar solvent interacts with polar surface.13 This situation does not apply to the studied gel. The magnetic cross-relaxation mechanism was used in the interpretation of the NMRD dispersion profiles of water in gel systems or proteins.24,25 There are two major contributions to cross-relaxation: the direct dipole-dipole coupling between diffusing solvent protons and the protons of solid gel matrix and the chemical exchange of unstable protons of the gel matrix (i.e., hydroxyl groups) with solvent protons. In the case of toluene used as the solvent, we can reject the contribution from the chemical exchange because the toluene has nonexchangeable protons. From the literature it is well-known that the effective diffusion coefficient of the solvent molecules at the solid matrix in porous media is not very different from that in the bulk.13 Therefore, the contribution to the toluene proton-gelator proton coupling modulated by translation diffusion is very small. As a result the correlation time for coupling is short and the relaxation efficiency low. Consequently, the cross-relaxation mechanism may be excluded in the interpretation of observed low frequency dispersion of the spin-lattice relaxation time of toluene confined in the 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose gel. The experimental data of the spin-lattice relaxation rate of toluene in the gel were therefore interpreted in terms of the twofraction fast-exchange model.26 We consider two phases: a surface-affected molecules of toluene with the T1s-1 and a bulk toluene with T1b-1, and we assume that the exchange rate between them is much faster than the relaxation rate in each of the phases. The two-fraction fast-exchange model is justified by a single value of the T1-1 calculated on the base of a single exponential magnetization curve measured over a studied frequency range. The exchange process of the toluene molecules in the vicinity of the surface of the gelator with that “inside” the pores-bulk toluene takes place by translational diffusion. The evidence that the mechanism for interphase (surface-bulk) transition is via diffusion is provided by the temperature dependence of the relaxation rate of toluene confined in the gel as shown in Figure 3. The relaxation rate increases with decreasing temperature. This is qualitatively consistent with a diffusion-induced relaxation process. In the two-fraction fast-exchange model, the observed relaxation rate is a linear combination of the bulk and surface contributions and we may write27 1 pb ps ¼ þ T1obs T1b T1s
ð3Þ
where pb and ps are the fraction of toluene in the pores behaves as in bulk and toluene at the gelator surface, respectively and pb þ ps =1. T1b is the spin-lattice relaxation time for the bulk toluene and T1s the relaxation time of toluene protons at the gelator aggregates. T1b is constant over the whole experimentally acces(25) Whaley, M.; Lawence, A. J.; Korb, J. P.; Bryant, R. G. Solid State NMR 1996, 7, 247–252. (26) Brownstein, K. R.; Tarr, C. E. J. Magn. Reson. 1977, 26, 17–24. (27) Stapf, S.; Kimmich, R. Chem. Phys. Lett. 1997, 275, 261–268.
17462 DOI: 10.1021/la103324s
Figure 3. Magnetic-field dependences of proton-spin lattice relaxation rates of toluene confined in the pores of studied gel at different temperatures. The dispersion curve at 223 K was the subject of our analysis as discussed in the text.
sible frequency range. Therefore, the frequency dependence of relaxation rate for toluene in gel is entirely determined by the fraction of the toluene molecules at the gelator surface ps. The observed dispersion of the relaxation rate in the frequency range 104 to 4 106 Hz shown in Figure 2 was analyzed with eq 3 but modified for our purpose in the following form: 1 1 τs 1 1 þ ps ð4Þ ¼ pb þA þ T1obs T1b T1r T1m 1 þ ωs 2 τ s 2 In eq 4 the term connected with the fraction of bulk toluene contains the value of the relaxation time for bulk toluene and the contribution from the molecular oxygen. The second term describes the contribution from the toluene at the gelator aggregates due to the reorientation of the methyl group T1m-1 and toluene ring T1r-1. These contributions are in the form given by eq 1. Equation 4 was used to analyze NMRD profiles obtained for gels 1 and 2. The solid line through the open squares in Figure 2 is the best fit of eq 4 to the experimental data. Three dashed lines reflected the contribution from the bulk toluene, and the protons from ring and methyl group of motionally altered fraction of toluene in gel 1, respectively. The fitted values for the activation energies for the ring and the methyl group of the fraction of toluene at the pore surface are Ear = 11.3 kJ mol-1 and Eam = 7.4 kJ mol-1, respectively. These values differ from that for bulk toluene (Ear = 7.7 kJ mol-1, Eam = 6.0 kJ mol-1). However, mainly the motion of the toluene ring is restricted at the surface, and as can be seen from Figure 2 this slow motion is responsible for the observed dispersion which extends down to the kHz region and results in a correlation time for the ring protons up to 10-5 s. It is well documented in the literature that the water or other solvent molecules interacting with internal surfaces of the porous system change their magnetic relaxation behavior.13 Interactions with the surface results in the slow decay of the NMR correlation function compared to the bulk water or bulk solvent. For example water confined in Bioran B30 is characterized by a correlation time up to 10-4 s.28 For comparison, the rotational correlation time in bulk water is in order of 10-12 s at room temperature. Therefore the long correlation times for the toluene ring protons is (28) Stapf, S.; Kimmich, R. J. Chem. Phys. 1995, 103, 2247–2250.
Langmuir 2010, 26(22), 17459–17464
Bielejewski and Tritt-Goc
not surprising. The toluene in gel is confined in the threedimensional network forms by the gelators molecules. The long correlation time of toluene ring results from the toluene-gelator interaction. As a consequence of this interaction the toluene molecules tend to adopt a preferential orientation with respect to the surface. We can assume that in gel 1 the toluene molecules are oriented toward the gelator aggregates with C2 axis of toluene ring perpendicular to the surface and with toluene rings close to the pore surface. Thus, the methyl groups are oriented toward the inside of the gel pore. Such orientation only slightly restricts the methyl reorientation but can significantly influence the motion of toluene rings. When the cooling rate during the gelation process of the sol was increased from 5 K min-1 (gel 1, open squares on Figure 2) to 15 K min-1 (gel 2, open triangles on Figure 2) the T1s-1 for motionally altered fraction of toluene molecules also increased. The solid line through the open triangles in Figure 2 represents the best fit of eq 4 to the experimental data obtained in gel 2. Two dotted lines stand for the contribution from the methyl group and ring protons of motionally altered fraction of toluene molecules, respectively. The fitted values for the activation energies for the ring and the methyl group of the fraction of molecules of toluene at the pore surface are Ear=11.8 kJ mol-1 and Eam=7.4 kJ mol-1, respectively. The value for the activation energy for the ring in gel 2 case also differs from that for bulk toluene (Ear=7.7 kJ mol-1) and only slightly differs for motionally altered fraction of toluene in the gel 1 (Ear = 11.3 kJ mol-1). The correlation times τ0 for methyl groups in gels 1 and 2 were correspondingly equal 6.9 10-12 and 1.11 10-11 s. Analogous to the situation as it was in gel 1, also in gel 2 mainly the motion of the toluene ring is restricted, and as can be seen from Figure 2 this slow motion is responsible for the observed dispersion. However, the correlation time for the methyl group of the motionally altered fraction of toluene in gel 2 has increased with respect to that obtained for gel 1. Such a behavior suggests that in this case also methyl groups are more slightly affected by interaction with the gelator aggregates’ surface. We assume that the orientation of the solvent molecules next to the surface of gelator aggregates is less ordered in comparison with gel 1. Due to the fast cooling rate, the orientation of the toluene molecules is “frozen” near the solid skeleton of the gel and reflects more the random order of toluene molecules in the vacancy of the pores. As a consequence the motion of toluene rings as well as methyl groups is influenced. The question arises about the origin of the orientation of toluene molecules at the pore surface. We assume that it comes from the interaction between the electric dipolar moment of the toluene molecules and polar surface composed of 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose aggregates. The fraction of motionally altered toluene responsible for the observed dispersion of the spin-lattice relaxation obtained from the fitting procedure is 2%. This small number of toluene molecules at the pore surface explains why the observed dispersion is so weak. It also proves the sensitivity of the field-cycling technique in the study of the solvent-surface interactions in gel systems. Figure 4 shows the microstructure of the studied gels created with different cooling rates, gels 1 and 2, respectively. The images are different from each other. The morphology of fibers of gel 1 consists of fibers composed in a starlike form whereas straight and rodlike fibers composed the microstructure of gel 2. Smaller cooling rate leads to gels whose microstructure is more ordered whereas a faster cooling rate leaves less time for ordering of the Langmuir 2010, 26(22), 17459–17464
Article
Figure 4. Optical polarizing micrographs of the gel [2%(g/mL)] of 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose formed with toluene: (a) at cooling rate 5 K min-1 (b) at cooling rate 15 K min-1.
gelator molecules and thus the aggregates are more irregular in form which leads to a less ordered microstructure of the gel as seen in Figure 4, panels a and b, respectively. The higher cooling rate leads also to more random orientation of the solvent molecules at the aggregates surface. However, the main difference of both gels affects the packing of the structure reflected by different density of the network. In gel 1 the gelator molecules formed a denser network as compared to gel 2. We consider that the cooling rate affects the kinetics of the gelation process which influences the microstructure of the gel by different packaging of the aggregates whereas the microstructure of the aggregates itself is dependent mostly on the used solvent and on the physical properties of the gelator such as polarity, dipole moment, solubility. The influence of different solvents on the 1,2-O-(1-ethylpropylidene)-R-D-glucofuranose gel microstructure was shown in our previous work.10 The difference in the microstructure guide to a different shape of pores and thus to different orientations of the toluene molecules at pore surface which influences the strengths of the gelator-toluene interactions as reflected in the NMRD profiles.
Conclusion Thanks to the 1H field-cycling nuclear magnetic resonance relaxometry we were able to measure the spin-lattice relaxtion time of toluene confined in the gel as a function of the magnetic field strenght and temperature. 1,2-O-(1-ethylpropylidene)-R-Dglucofuranose gel with toluene is considered as microporous material in which an exchange process between the entrapped DOI: 10.1021/la103324s
17463
Article
at the pore surface solvent components of this system and bulk solvent phase is responsible for the low magnetic field dispersion of the solvent proton relaxation rate. The occurrence of the dispersion in the frequency range 104 -106 Hz gives strong evidence of the interaction between the toluene and gelator aggregates. The results were interpreted in terms of a two-fraction fast-exchange model in the fast-exchange limit. We belive that our results finally answer some long-standing question as to the existence of the interaction between the solvent and gelator in the LMOG gel system. In addition we also have shown the
17464 DOI: 10.1021/la103324s
Bielejewski and Tritt-Goc
influence of a cooling rate during the gel preperation process on the gel microstructure. A different microstructure leads to the different gelator-solvent interactions and different behavior of the proton spin-lattice relaxation rate of solvent within the gel observed at the low frequency range. Acknowledgment. One of the authors (M.B.) thanks Prof. Ernst Roessler from the University of Bayreuth, Germany for an opportunity to lern in his labolatory the FFC technique and performed the preliminary measurements.
Langmuir 2010, 26(22), 17459–17464
