Diffusive Diffraction Phenomenon Observed by PGSE NMR Technique

Sep 6, 2012 - ... https://cdn.mathjax.org/mathjax/contrib/a11y/accessibility-menu.js .... of solid-state NMR spectroscopy and DFT optimization in the ...
0 downloads 0 Views 485KB Size
Article pubs.acs.org/Langmuir

Diffusive Diffraction Phenomenon Observed by PGSE NMR Technique in a Sugar-Based Low-Molecular-Mass Gel Jadwiga Tritt-Goc* and Joanna Kowalczuk Institute of Molecular Physics, Polish Academy of Sciences, M. Smoluchowskiego 17, 60-172 Poznań, Poland ABSTRACT: The paper presents the diffusive diffraction phenomenon observed by the single-pulse-gradient spin−echo (s-PGSE) NMR technique in a real porous material: a gel composed of low-molecular-mass gelator methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside and toluene. Thanks to this phenomenon, we can probe the true microstructure (not xerogel) in which the toluene diffuses. To analyze the measured diffusion−diffraction pattern, we employed a composite bicompartmental model that superimposes restricted diffusion in small cavities of the gel matrix within the bundles of crossing fibers, with free diffusion in large and unconfined compartments between the bundles of crossing fibers. For restricted diffusion a pore-hopping formalism was applied. The observation of the diffraction pattern and its analysis leads to the conclusion that the pores, in the slow diffusing compartment of studied gel are ordered, at least locally, and relatively monodisperse with a size of 64 μm. Moreover, the restricting walls formed by the crossing fibers are perpendicular to the direction of the diffusion gradient.



INTRODUCTION Different classes of low-molecular-mass organic molecules (LMOGs) possess the ability to gelatinize a wide spectrum of organic solvents and form organogels at a very low gelator concentration.1−4 The characteristic feature of these gels is the coexistence of two phases: the solid, three-dimensional matrix of networks of nanofibers formed by the gelator aggregates and the liquid one confined within the pores of the solid matrix. Therefore, organogels can be considered a porous microstructured material. Gelator molecules in the fibers are selfassembled through noncovalent interactions, such as electrostatic, dipole−dipole, hydrogen-bonding, the π−π stacking, and van der Waals interactions, and thus the gels derived from LMOGs are classified as physical gels.1 The extensive studies of the organogels, conducted in recent years due to their numerous industrial applications and interesting supramolecular structures, have improved the understanding of the structural requirements for a molecule to be a good gelator, of the self-organization of gelator molecules into the network structure, and of the correlation between the molecular packing of gelator molecules in their aggregates. Such studies also help to understand the mechanical, thermodynamic, optical, and other properties of gels and the dependence of the thermal stability and of the gel morphology upon the gelated solvent.5−19 Despite the progress in understanding the properties of the organogels, the accurate design of a new gelator is still a hard task, and they are found mostly serendipitously. In our studies we focus on the gelation phenomenon of a small and effective LMOGs based on glucofuranose derivatives.15,16 In particular, we are interested in the determination of the driving forces responsible for gelator aggregation, the thermal properties and microstructure of gels, solvent−gelator © 2012 American Chemical Society

interactions, and molecular dynamics of solvents within the gel matrix.9−11,17−19 The subject of this work, methyl-4,6-O-(pnitrobenzylidene)-α-D-glucopyranoside, is also a representative of sugar-based gelators. This is a unique gelator because its small and weakly interacting molecules can form large supramolecular structures in both nonpolar and polar solvents and cause their gelation.17 Such “bifunctional” gelators are not often found in the literature.20 Previously, we showed that the hydrogen-bonding interaction is the main driving force for the aggregation of this gelator’s molecules in both polar and nonpolar solvents and discussed the influence of solvent on the thermal stability and organization of self-assembling fibrillar networks in gels.17 Generally, the morphology of the xerogels varied depending on the solvent, cooling rate during gelation, gel concentration, capacity of the gel, xerogel preparation method, and temperature but in all cases may be described as “fibrillar”.21 The molecules’ arrangement within the nanofibers, regardless of their random entanglement, displays significant order. A unidirectional intermolecular structure is observed. The nanofibers in the gel composed of the methyl-4,6-O-(pnitrobenzylidene)-α-D-glucopyranoside with toluene have a rather thick, needle-like shape. The gel matrix consists of bundles of crossing fibers with different lengths, thickness, and orientations, dispersed in the gel.17 Our last paper reports the experimental evidence of solventgelator interaction in a gel composed of methyl-4,6-O-(pnitrobenzylidene)-α-D-glucopyranoside with chlorobenzene18 obtained with the 1H fast field cycling (FFC) relaxometry Received: June 11, 2012 Revised: September 1, 2012 Published: September 6, 2012 14039

dx.doi.org/10.1021/la302364d | Langmuir 2012, 28, 14039−14044

Langmuir

Article

restricting geometry in the system. It is this reciprocal-space dependence of the signal which bears the similarity to diffraction. The diffraction effect arising as a consequence of the confined geometry experienced by the fluid inside the cavities of a porous structure. The so-called q-space formalism of PGSE NMR technique was first developed by Callaghan et al.27,28 In recent years the theoretical curves of diffraction-like pattern have been predicted for different geometries of the pores29−36 and have been observed experimentally in different porous materials, including real systems.37−47 The purpose of this paper is to present the diffusive diffraction phenomenon observed in the gel composed by methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside with toluene and to show that thanks to this effect we can probe the microstructure of the gel in which the toluene diffuses. To our knowledge this is the first evidence of the diffraction pattern observed in gel composed by LMOGs. To analyze the experimental data, we employed a composite bicompartmental model which superimposes restricted diffusion in small cavities of the gel matrix with free diffusion in large and unconfined compartments. As a result, a slow and fast diffusing component was extracted from observed NMR signal decay. Previously, the superposition of two diffusion modes in the analysis of the diffusion curve had been investigated theoretically.44,45 However, the bicompartmental model in which the free diffusion is added to the system, which already exhibits diffusion−diffraction pattern, has been introduced only recently by Shemesh46 and applied to a phantom consisting of waterfilled microcapillaries enclosed in an NMR tube filled with free water. In this study, for the first time, we used the bicompartmental model to investigate the signal decay and to extract the structural information in a real porous system such as the gel of methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside with toluene. For restricted diffusion a pore-hopping formalism worked out by Callaghan et al. was applied.25,28,33

method. Previously, we have studied the dynamics of water in the hydrogel of this gelator19 by s-PGSE methodology.22 Later studies revealed that when the compartments in this gel matrix become comparable with the diffusion distance of water, the confinement effect starts to play a role and so-called restricted diffusion22,23 is observed, which is manifested in the linear decreases of the diffusion coefficient with diffusing time. Generally, the echo attenuation caused by Brownian motions is given by E(g , Δ) = exp[−γ 2g 2δ 2D freeΔ]

(1)

where E(g,Δ) is the normalized echo signal, γ is the magnetogyric ratio of the studied nucleus, g is the field gradient strength, δ is the duration of the gradient pulses, Δ is the gradient pulse interval, which is called the diffusion time, and Dfree is the diffusion coefficient. For free and restricted diffusion, the signal decay is characterized by its distinctive diffusion profile.23−25 In freely diffusing systems, where the molecules are not hindered by barriers, the logarithm of the signal decay given by eq 1 is linearly dependent on the b-value (b = γ2g2δ2(Δ − δ/3)), from which the Dfree can be extracted. Furthermore, Dfree is independent of the diffusion time, and consequently, the root-mean-square displacement (rmsd) of the diffusing molecule can be obtained by Einstein’s relation: (rmsd)2 = 2nDfreetd where n = 1, 2, or 3 depending on the dimension of the diffusion process; td = Δ − δ/3. The plot of rmsd versus td1/2 yields a linear correlation. Free diffusion is also characterized by the same value of rmsd regardless of the direction along which the diffusion measurements are performed. Through the linear decay of the logarithm of the NMR signal as a function of the b-value and the linear dependence of the mean-squared displacements on td the Gaussian nature of free diffusion is manifested. The diffusion profile for restricted diffusion is different: the logarithm of the signal attenuation is linear only for low bvalues (short diffusion time when other parameters are kept constant), and eq 1 can be used to evaluate only an apparent diffusion coefficient Dapp. For sufficiently long diffusion time, when the pores are increasingly sampled by more and more molecules, a deviation from linearity of the signal decay with respect to the b-value is observed, which characterizes restricted diffusion. Therefore, in contrast to free diffusion, the signal attenuation characterizing the restricted diffusion depends on the diffusion period and also on the direction in which the diffusion occurs.25 If all molecules have sufficient time to probe the pore boundaries and fully experience the restriction effects, then they can be used as unique fingerprints for the pore microstructure, but completely different treatment of the signal decay is necessary. Therefore, the PGSE NMR technique can not only be used to characterize diffusion but also to probe the microstructure in which the fluid diffuses from “diffraction” pattern that signal attenuation curves exhibit. The latter is however possible only for a porous system when pores are well ordered, at least locally, and relatively monodisperse in size. The diffraction-like effect in PGSE experiment has been theoretically discussed first by Callaghan et al.26 and by Corry et al.27 and experimentally shown by Callaghan for the system composed of randomly packed polystyrene spheres, measuring water diffusion.26 The diffraction-like pattern is best interpreted using the mathematics of diffraction theory and examining the dependence of E(q,Δ) versus q, where q = (2π)−1γδg is reciprocal lattice wave vector, which describes the reciprocal space of the



EXPERIMENTAL SECTION

Gel Preparation. Methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside was synthesized according to a method described elsewhere.47 To obtain a 2% (w/v) gel with toluene, the gelator was mixed with the appropriate amount of the solvent directly in a closed 6 mm NMR tube. The mixture was heated until the solid was dissolved. The solvent boiling point was reached in this procedure. Cooling the solution below the characteristic gelation temperature (in this case Tgel = 320 K) brings out the transition to the gel phase. As a result, a thermoreversible, stable, optically clear and transparent gel is obtained. NMR Diffusion Experiment. NMR measurements were performed with a Bruker Avance spectrometer operating at 300 MHz for protons and coupled with a 7.1 T, 89 mm bore Bruker magnet. The system is equipped with a Micro2.5 probehead and gradient system capable of producing pulse gradients up to 100 Gs cm−1 at 40 A in each of the three directions. The 6 mm NMR tube with a gel was placed inside a 10 mm radio-frequency (RF) coil and aligned along the z-direction in the magnet. The conventional s-PGSE methodology22 was used for measuring diffusion of toluene confined in the cavities of methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside gel. This method uses the Hahn spin-echo pulse sequence (90°−τ−180°−τ− echo) with a pair of square-shaped gradient field pulses of magnitude g and duration δ, separated by so-called diffusion time Δ. The 90° and 180° RF pulses serve as excitation and refocusing pulses, respectively. The first gradient pulse, applied between the two RF pulses, winds a magnetization helix which imparts a spatially dependent phase shift, and then the spins are allowed to diffuse during time Δ. The second gradient pulse, after the 180° pulse, acts to unwind the magnetization 14040

dx.doi.org/10.1021/la302364d | Langmuir 2012, 28, 14039−14044

Langmuir

Article

E(q , Δ) = |S0(q)|2 F(q , Δ)

helix and, consequently, to refocus the previously acquired phase shift. All stationary spins will acquire zero net phase whereas all mobile spins will attain a finite net phase because they experience a mismatch between the phases acquired during each gradient pulse. Therefore, in s-PGSE the signal is phase encoded according to the molecular displacement over a well-defined time interval. The molecular displacement leads to an attenuation of the echo signal. In our experiment the pulse gradient was applied in the z-direction and varied from 0 to 50 Gs cm−1 in 32 steps. The duration of the pulse gradient, δ, in all experiments was 2 ms. The echo decay curves were measured at 300 K in the function of diffusion time Δ = 30, 50, 100, 150, 200, and 250 ms. The measured 1H NMR signal derives from toluene molecules diffusing within the cavities of the studied gel. Theoretical Analysis: Bicompartment Model. A composite bicompartmental model46 assumes the coexistence of a fast diffusion compartment (FDC) and a slow diffusion compartment (SDC). These two compartments reflect the two diffusion modes: free or almost free diffusion of solvent in large cavities of gel matrix and restricted diffusion of solvent in small pores. Assuming that there is no exchange of solvent between two compartments under consideration, the NMR echo decay signal can be written as the sum of two following expressions:

E(q) = ffree Efree(q) + frest Erest(q)

with |S0(q)| being the average pore structure factor. The modulations of the echo signal due to this factor are observed as a local minima in the plot of E(q,Δ) versus q. F(q,Δ) is sensitive to the details of motions between pores. A plot of E(q,Δ) may display diffraction-like features only at such values of the diffusion time Δ that the molecules have displacements on the same length scale as some characteristic distance of the structure probed by diffusing molecules and only where Δ is sufficiently long to allow a significant number of molecules to perform a hop from a starting pore to the neighbor ones. The characteristic features of the diffraction pattern are the position of nodes or the coherence maxima. The latter occurs under these conditions when the “gradient wavelength” matches the periodicity of the matrix. Inserting into eq 6 the form of the function F(q,Δ) developed for the pore−glass system using the pore-hopping model formalism33 yields the equation ⎡ 6D Δ ⎛ E(q , Δ) = |S0(q)|2 exp⎢− 2 rest 2 ⎜1 − exp(− 2π 2q2ξ 2) ⎢⎣ b + 3ξ ⎝ sin(2πqb) ⎞⎤ ⎟⎥ 2πqb ⎠⎥⎦

(2)

where Efree(q) and Erest(q) are the contributions of the NMR signal attenuations from the solvent in two compartments that are equal to unity at |q| = 0 and f free and f rest are the fractions of the molecules from the solvent in free and restricted compartments, respectively, which satisfy the relationship f free + f rest = 1. The echo attenuation curve for free diffusion is given by eq 1, which in q-space formalism takes the form Efree(q , Δ) = exp[− 4π 2q2DfreeΔ]

(7)

To allow for some variation in the pore spacing b, the standard deviation parameter ξ was introduced in eq 7. The exact form of factor |S0(q)|2 has been calculated for spherical, cylindrical, and rectangular pores29−36 and contains the a parameter, which corresponds to the mean pore size. In this theory the short gradient pulse (SGP) limit approximation is assumed, which means that the gradient pulse length, δ, is so short that the spins do not diffuse during the gradient pulse. For finite pulses Δ in eq 7 is replaced by Δ − δ/3. In the simulation a resulting signal attenuation profile was fitted to experimental data using a Levenberg−Marquardt fitting routine. The following parameters were estimated via this procedure: the pore diameter a, the pore distance b, the diffusion coefficients Dfree and Drest, and the fractions f free and f rest.

(3)

In the computation of the signal from restricted diffusion, we employed a so-called pore-hopping model described below. Pore-Hopping Model. The pore-hopping theory was developed by Callaghan25 for fluid saturated porous media and first applied for systems composed of interconnected boxes33 and for the interpretation of experimental data for beads of monodisperse spherical polystyrene particles.28 Canet et al.37 used a pore-hopping model for the interpretations of a real porous system made up of cross-linked polystyrene synthesized by an inverse emulsion process. The aim of this paragraph is not to give a detailed description of the theory, which can be found elsewhere.25,28,33 Instead, we want to show how the porehopping model can be used for the interpretation of the diffraction pattern of studied organogel and for estimation of gel microstructure parameters. Therefore, only the basic idea and resulting equation from theory are presented. Generally, the pore-hopping theory applies to the diffusion of fluid in a porous system consisting of well-defined, interconnected pores with the assumption that the diffusion between pores is much slower than within a single pore and successive hops are uncorrelated. If the condition

a2 b2 ≪ D0 Dα

(6)

2



RESULTS AND DISCUSSION Dependence of the Signal Decay on the Diffusion Time. The signal decay for s-PGSE experiments performed on toluene protons in the gel of methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside is shown in Figure 1 for different values of Δ at 300 K. Only at very short diffusion time Δ = 30 ms, the signal decay exhibits a nearly free diffusion profile. When Δ increases, the deviation from the Gaussian shape of the echo decay appears and small oscillations are observed,

(4)

for a system consisting of pores with the dimension a and pore distance b, bulk diffusion D0, and long-range diffusion coefficient Dα within the porous network is fulfilled, then the propagator for the molecules to displace over the diffusion time Δ is a convolution of the autocorrelation function of the shape of a single pore ρ(z) (z is the position of spin after a time Δ) and the microstructural correlation function L(Z) describing the relative positions of the pores in the space, weighted by the diffusion envelope C(Z,Δ). The echo attenuation is the Fourier transform of the propagator

E(q , Δ) = |FT[ρ(z)]|2 [FT[L(Z)] ⊗ FT[C(Z , Δ)]]

(5)

Figure 1. Echo intensities E(q) as a function of the gradient wavevector q at 300 K for a series of diffusion times Δ.

Equation 5 can be rewritten as 14041

dx.doi.org/10.1021/la302364d | Langmuir 2012, 28, 14039−14044

Langmuir

Article

of crossing fibers experiences restricted diffusion. In the computation of the signal from restricted diffusion, we employed a pore-hopping model described above (eq 7). For simplicity, the pore structure factor (|S0(q)|2 = |sin(πqa)/πqa|2) corresponds to a spherical pore of diameter a and could be seen as rather unrealistic. However, the assumption about the spherical pore shape can be justified because as shown by Ozarslan34 there are no significant qualitative differences between the signals curves obtained from different geometries in s-PGSE experiment. The signal decay curves obtained from cylindrical, spherical pores or from parallel planes differ only in multipulse-field gradient experiment.34 Figure 2 shows the experimental data (red circles), for Δ = 150 ms, together with theoretical curve (solid red line) obtained using eq 2 (after inserting eqs 3 and 7) in a nonlinear least-squares fit to the parameters a, b, ξ, Dfree, Drest, and f free, f rest. The theoretical curve is in good agreement with the experimental data. Equation 2 describes the diffraction trough at 0.03 μm−1 surprisingly well. The fit yields structural parameters a = 64 μm and b = 120 μm with ξ = 13 μm. The diffusion coefficients for free and restricted diffusion are Dfree = 1.20 × 10−9 m2 s−1 and Drest = 0.98 × 10−9 m2 s−1, respectively. The volume fractions of free and restricted compartments are f free = 0.51 and f rest = 0.49. With the fitted parameters, the two components of the echo decay curve were calculated: the free diffusion profile from the FDC and diffraction patterns expected from the SDC. They are presented in Figure 2 as black dashed and solid blue curves, respectively. Now, we can see that the additive nature of the signal for FDC and SDC smoothes the echo decay because the freely diffusing component of toluene almost completely masks the first diffraction trough occurring in the diffraction pattern at q = 0.015 μm−1 (1/q = a). That is why the structural parameter a cannot be extracted directly from the diffraction pattern but only with the proper model used. Using the bicompartmental model, we could investigate the structural information on the gel matrix and also test how the properties revealed by the s-PGSE experiment are affected by the presence of the two diffusion modes at different q-values. However, we sample a very narrow range of q-values. This explains why the attenuation of the measured signal decay shown in Figure 2 is equally dominated by the contribution from free and restricted diffusion up to q-value equal 0.03 μm−1. Previously, it was shown that the contribution from FDC is shifted to the higher q-value with increasing amount of free water.46 In the studied gel the fractions of toluene in both compartments are almost equal. For q > 0.03 μm−1 the restricted diffusion starts to dominate. The value of the parameter a = 64 μm can be treated as an average distance between the crossing fibers, perpendicular to the direction of the gradient, which formed barriers for diffusion. The value of the parameter ξ equal to 13 (standard deviation in b = 120 μm) is the measure of the sample heterogeneity. From Figure 2 it can be concluded that we are able to generate a quantitative agreement between experiment and eq 2, which proves our hypothesis of the behavior of the echo decay signal being governed by two main types of toluene molecules: free toluene in the FDC and restricted toluene in the SDC. The good agreement between the theory and experiment may be due to the absence of surface relaxation effects. If surface relaxation were significant, then it plays a role in the final appearance of the echo decay smearing out the

which could be remainders of the coherence pattern predicted by q-space formalism of the s-PGSE technique.25,28 The diffraction effect arises as a consequence of the confined geometry experienced by the toluene inside the gel cavities. It is clearly seen in Figure 2 where the diffraction-like pattern

Figure 2. Echo intensities E(q) at Δ = 150 ms. The red circles represent experimental data while the solid red line represents the theoretical curve obtained using the bicompartmental model which superimposes restricted diffusion in small cavities of the gel matrix with free diffusion in large and unconfined compartments.

(experimental data are presented as the red circles) is presented in log scale for Δ = 150 ms. The echo decay shows a bend in the signal attenuation profile at q approximately equal to 0.015 μm−1 and the diffraction trough occurring at q = 0.031 μm−1. The observed minimum is sharp which shows that the SGP condition is met in our experiment. The value of q, at which the first diffusion−diffraction trough occurs, corresponds to the reciprocal of the size of compartment.30 Therefore, directly from the diffraction pattern its size can be estimated. However, it is not the case for the studied gel where a more sophisticated model for the analysis of the attenuation curve is necessary to obtain the structural information on the gel matrix. Structural Information Extracted from the Diffusion Pattern. The studied gel network, based on the micrograph taken by optical polarization microscopy,17 can be considered to contain randomly arranged pores of different sizes and shapes. To analyze the experimental data, we assumed that the gel matrix contained on the average two types of compartments. In one type the average pore size is large, so in the measured diffusion time, the diffusion of the toluene molecules is unrestricted. The large pores are situated in the gel matrix between the bundles of crossing fibers and “form” the FDC. The second compartment is composed of pores much smaller within the bundles of crossing fibers, and the toluene molecules experience the restriction in diffusion forming the SDC. Consequently, the measured echo attenuation signal has two components: from free toluene in the FDC and restricted toluene in the SDC. The restricted diffusion occurs due to the pore boundariesfibers formed due to the self-assembling of gelator molecules. The surfaces of the fibers act as the scattering centers, and we can assume that these fibers are oriented, at least on average, almost perpendicular to the diffusion gradients because in such anisotropic samples as the studied gel only such orientation allows for the observation of the diffusion− diffraction pattern. Freely diffusing toluene in the FDC undergoes Gaussian diffusion (eq 3) while toluene in the pores within the bundles 14042

dx.doi.org/10.1021/la302364d | Langmuir 2012, 28, 14039−14044

Langmuir

Article

maxima and the minima in the echo decay.40 In eq 7 the influence of the spin−lattice or spin−spin relaxation time is not taken into account. We assumed that the fibers have no effect on the relaxation behavior of the toluene protons. The assumption is based on the temperature measurements of proton spin−lattice relaxation time of toluene in the gel performed as a function of the magnetic field by the FFC relaxometry method (data not shown). In the case of the interactions between the solvent and the gel matrix a significant slowing down of the motion of solvent molecules at the gelator surfaces is observed when compared to bulk solvent. The motion manifests itself through the low-frequency dispersion of the proton spin−lattice relaxation time of the solvent in the gel observed below 105 Hz. Such behavior was observed in the gel of methyl-4,6-O-(p-nitrobenzylidene)-α- D-glucopyranoside with chlorobenzene18 but not with the studied gel. The gel with chlorobenzene does not show the diffraction patterns.

Notes

The authors declare no competing financial interest.

ACKNOWLEDGMENTS



REFERENCES

Financial support for this work was provided by the National Centre for Science (Grant N N202 1961 40). Authors are very grateful to Dr Roman Luboradzki from the Institute of Physical Chemistry, Polish Academy of Sciences, Warsaw, for supply us with the organogelator.

(1) Weiss, R. G.; Terech, P. Molecular Gels, Materials with Self− Assembled Fibrillar Network; Springer: Dordrecht, The Netherlands, 2006. (2) George, M.; Weiss, R. G. Molecular organogels. soft matter comprised of low-molecular-mass organic gelators and organic liquids. Acc. Chem. Res. 2006, 39, 489−497. (3) Estroff, L. A.; Hamilton, D. Water gelation by small organic molecules. Chem. Rev. 2004, 104, 1201−1218. (4) Terech, P.; Weiss, R. G. Low molecular mass gelators of organic liquids and the properties of their gels. Chem. Rev. 1997, 97, 3133− 3159. (5) Piepenbrock, M. O. M.; Lloyd, G. O.; Clarke, N.; Steed, J. W. Metal- and anion-binding supramolecular gels. Chem. Rev. 2010, 110, 1960−2004. (6) Steed, J. W. Supramolecular gel chemistry: developments over the last decade. Chem. Commun. 2011, 47, 1379−1383. (7) Samai, S.; Dey, J.; Biradha, K. Amino acid based low-molecularweight tris(bis-amido) organogelators. Soft Matter 2011, 7, 2121− 2126. (8) Allix, F.; Curcio, P.; Pham, Q. N.; Pickaert, G.; Jamart-Gregoire, B. Evidence of intercolumnar π-π stacking interactions in amino-acidbased low-molecular-weight organogels. Langmuir 2010, 26, 6818− 16827. (9) Tritt-Goc, J.; Bielejewski, M.; Luboradzki, R.; Łapiński, A. Thermal properties of the gel made by low molecular weight gelator 1,2-O-(1-ethylpropylidene)-α-D-glucofuranose with toluene and molecular dynamics of solvent. Langmuir 2008, 24, 534−540. (10) Bielejewski, M.; Łapiński, A.; Luboradzki, R.; Tritt-Goc, J. Solvent effect on 1,2-O-(1-ethylpropylidene)-α-D-glucofuranose organogel properties. Langmuir 2009, 25, 8274−8279. (11) Bielejewski, M.; Tritt-Goc, J. Evidence of solvent-gelator interaction in sugar-based organogel studied by field-cycling NMR relaxometry. Langmuir 2010, 26, 17459−17464. (12) Vintiloiu, A.; Leroux, J. C. Organogels and their use in drug delivery  A review. J. Controlled Release 2008, 125, 179−192. (13) Llusar, M.; Sanchez, C. Inorganic and hybrid nanofibrous materials templated with organogelators. Chem. Mater. 2008, 20, 782− 820. (14) Kaszyńska, J.; Łapiński, A.; Bielejewski, M.; Luboradzki, R.; Tritt-Goc, J. On the relation between the solvent parameters and the physical properties of methyl-4,6-O-benzylidene-α-D-glucopyranoside organogels. Tetrahedron 2012, 68, 3803−3810. (15) Luboradzki, R.; Pakulski, Z. Novel class of saccharide-based organogelators: glucofuranose derivatives as one of the smallest and highly efficient gelator. Tetrahedron 2004, 60, 4613−4616. (16) Luboradzki, R.; Pakulski, Z.; Sartowska, B. Glucofuranose derivatives as a library for designing and investigating low molecular mass organogelators. Tetrahedron 2005, 61, 10122−10128. (17) Tritt-Goc, J.; Bielejewski, M.; Luboradzki, R. Influence of solvent on the thermal stability and organization of self-assembling fibrillar networks in methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside gels. Tetrahedron 2011, 67, 7222−7230. (18) Tritt-Goc, J.; Bielejewski, M.; Luboradzki, R. Interaction of chlorobenzene with gelator in methyl-4,6-O-(p-nitrobenzylidene)- α -D-glucopyranoside gel probed by proton fast field cycling NMR relaxometry. Tetrahedron 2011, 67, 8170−8176.



CONCLUSIONS The paper reports the experimental evidence of the diffractionlike effect from proton s-PGSE NMR studies of toluene confined in the cavities of sugar-based physical gels. The organogels formed an interesting class of systems of great importance in a number of technical applications. As a consequence, it is essential to understand the basic properties of such systems, and here the noninvasive s-PGSE NMR technique can provide a valuable tool to obtain the structural and dynamical properties of the systems. The observation of the diffusive diffraction phenomenon in the realistic porous material such as the studied gel is somehow unexpected. Usually, the experiment performed on a fluid confined in real porous media, using the s-PGSE sequence, yields a featureless decay, as an effect of large pores anisotropy and pore size distributions, and due to the large susceptibility artifact arising from susceptibility differences within the porous medium. Contrary to the usual expectation of the monotonically decreasing E(q) with increasing q, we observe the coherence peaks and diffraction troughs, in a s-PGSE experiment performed on toluene protons in the gel of methyl-4,6-O-(p-nitrobenzylidene)-α- D -glucopyranoside. Therefore, we can conclude that the pores in the SDC of the gel are ordered, at least locally, and relatively monodisperse with a mean size of 64 μm. Moreover, the restricting walls formed by the crossing fibers are perpendicular to the diffusion gradient applied along the z-direction. The information obtained by s-PGSE NMR technique concerns the morphology of the gel matrix in the gel phase. Therefore, we can name it a “true” gel matrix. Contrarily, the SEM method gives the morphology of the xerogels, which depends on many parameters21 and cannot always correspond to a “true” gel matrix. The estimation of the structural parameters from the diffraction pattern was possible by employing a composite bicompartmental model which superimposes restricted diffusion in small cavities of the gel matrix with free diffusion in large and unconfined compartments. To our best knowledge, this model was for the first time applied to a real porous system such as the gel of methyl-4,6-O-(p-nitrobenzylidene)-α-Dglucopyranoside with toluene.





AUTHOR INFORMATION

Corresponding Author

*Phone +48-061 8695-226; e-mail [email protected]. 14043

dx.doi.org/10.1021/la302364d | Langmuir 2012, 28, 14039−14044

Langmuir

Article

(40) Wassenius, H.; Nyden, M.; Vincent, B. NMR diffusion studies of translational properties of oil inside core−shell latex particles. J. Colloid Interface Sci. 2003, 264, 538−547. (41) Seland, J. G.; Ottaviani, M.; Hafskjold, B. A PFG-NMR study of restricted diffusion in heterogeneous polymer particles. J. Colloid Interface Sci. 2001, 239, 168−177. (42) Shapiro, Y. E. 1H NMR self-diffusion study of morphology and structure of polyvinyl alcohol cryogels. J. Colloid Interface Sci. 1999, 212, 453−465. (43) Kuchel, P. W.; Coy, A.; Stilbs, P. NMR “diffusion−diffraction” of water revealing alignment of erythrocytes in a magnetic field and their dimensions and membrane transport characteristics. Magn. Reson. Med. 1997, 37, 637−643. (44) van der Weerd, L.; Melnikov, S. M.; Vergeldt, F. J.; Novikov, E. G.; Van As, H. Modelling of self-diffusion and relaxation time NMR in multicompartment systems with cylindrical geometry. J. Magn. Reson. 2002, 156, 213−221. (45) Johnson, C. S. Diffusion ordered nuclear magnetic resonance spectroscopy: Principles and applications. Prog. Nucl. Magn. Reson. Spectrosc. 1999, 34, 203−256. (46) Shemesh, N.; Ozarslan, E.; Bar-Shir, A.; Basser, P. J.; Cohen, Y. Observation of restricted diffusion in the presence of a free diffusion compartment: single- and double-PFG experiments. J. Magn. Reson. 2009, 200, 214−225. (47) Gronwald, O.; Shinkai, S. “Bifunctional” sugar-integrated gelators for organic solvents and water - On the role of nitrosubstituents in 1-O-methyl-4,6-O-(nitrobenzylidene)-monosaccharides for the improvement of gelation ability. J. Chem. Soc., Perkin Trans. 2001, 2, 1933−1937.

(19) Kowalczuk, J.; Jarosz, S.; Tritt-Goc, J. Characterization of low molecular-weight gelator methyl-4,6-O-(p-nitrobenzylidene)-α-D-glucopyranoside hydrogels and water diffusion in their networks. Tetrahedron 2009, 65, 9801−9806. (20) Jung, J. H.; John, G.; Masuda, M.; Yoshida, K.; Shinkai, S.; Shimizu, T. Self-assembly of a sugar-based gelator in water: its remarkable diversity in gelation ability and aggregate structure. Langmuir 2001, 17, 7229−7232. (21) Pyzalska, M.; Smal, M.; Luboradzki, R. Different morphologies of the dry physical gels as a function of the sample preparation method. J. Non-Cryst. Solids 2011, 357, 3184−3189. (22) Stejskal, E. O.; Tanner, J. E. Spin diffusion measurements: spin echoes in the presence of a time-dependent field gradient. J. Chem. Phys. 1965, 42, 288−292. (23) Price, S. W. Gradient NMR. Annu. Rep. NMR Spectrosc. 1996, 32, 51−142. (24) Sen, P. N. Time-dependent diffusion coefficient as a probe of geometry. Concepts Magn. Reson. 2004, 23A, 1−21. (25) Callaghan, P. T. Principle of Nuclear Magnetic Resonance Microscopy; Oxford Science Publications: Oxford, 1991. (26) Callaghan, P. T.; MacGowan, D.; Packer, K. J.; Zelaya, F. O. High-resolution q-space imaging in porous structures. J. Magn. Reson. 1990, 90, 177−182. (27) Corry, D.; Garroway, A. Measurement of translational displacement probabilities by NMR - An indicator of compartmentation. Magn. Reson. Med. 1990, 14, 435−444. (28) Callaghan, P. T.; Coy, A.; MacGowan, D.; Packer, K. J.; Zelaya, F. O. Diffraction-like effects in NMR diffusion studies of fluids in porous solids. Nature 1991, 351, 467−469. (29) Coy, A.; Callaghan, P. T. Pulsed gradient spin echo nuclear magnetic resonance for molecules diffusing between partially reflecting rectangular barriers. J. Chem. Phys. 1994, 101, 4599−4609. (30) Callaghan, P. T. Pulsed gradient spin echo NMR for planar, cylindrical and spherical pores under conditions of wall relaxation. J. Magn. Reson. 1995, 113, 53−59. (31) Gibbs, S. J. Observations of diffusive diffraction in a cylindrical pore by PFG NMR. J. Magn. Reson. 1997, 124, 223−226. (32) Codd, S. L.; Callaghan, P. T. Spin echo analysis of restricted diffusion under generalized gradient waveforms: planar, cylindrical, and spherical pores with wall relaxivity. J. Magn. Reson. 1999, 137, 358−372. (33) Callaghan, P. T.; Coy, A.; Halpin, T. P. J.; MacGowan, D.; Packer, K. J.; Zelaya, F. O. Diffusion in porous systems and the influence of pore morphology in pulsed gradient spin-echo nuclear magnetic resonance studies. J. Chem. Phys. 1992, 97, 651−662. (34) Ozarslan, E.; Basser, P. J. MR diffusion − “diffraction” phenomenon in multi-pulse-field-gradient experiments. J. Magn. Reson. 2007, 188, 285−294. (35) Shemesh, N.; Ö zarslan, E.; Adiri, T.; Basser, P. J.; Cohen, Y. Noninvasive bipolar double-pulsed-field-gradient NMR reveals signatures for pore size and shape in randomly oriented, polydisperse, inhomogeneous porous media. J. Chem. Phys. 2010, 133, 044705. (36) Shemesh, N.; Westin, C.-F.; Cohen, Y. Magnetic resonance imaging by synergistic diffusion-diffraction patterns. Phys. Rev. Lett. 2012, 108, 058103. (37) Kuntz, J. F.; Palmas, P.; Canet, D. Diffusive diffraction measurements in porous media: effect of structural disorder and internal magnetic field gradients. J. Magn. Reson. 2007, 188, 322−329. (38) Kuntz, J. F.; Trausch, G.; Palmas, P.; Mutzenhardt, P.; Canet, D. Diffusive diffraction phenomenon in a porous polymer material observed by NMR using radio freqency field gradients. J. Chem. Phys. 2007, 126, 134904. (39) Avram, L.; Assaf, Y.; Cohen, Y. The effect of rotational angle and experimental parameters on the diffraction patterns and microstructural information obtained from q-space diffusion NMR: implication for diffusion in white matter fibers. J. Magn. Reson. 2004, 169, 30−38. 14044

dx.doi.org/10.1021/la302364d | Langmuir 2012, 28, 14039−14044