bk-2011-1077.ch009

Chapter 9

Downloaded by PURDUE UNIVERSITY on August 5, 2013 | http://pubs.acs.org Publication Date (Web): October 14, 2011 | doi: 10.1021/bk-2011-1077.ch009

NMR Characterization of Canopy Dynamics in Nanoscale Ionic Materials Michael L. Jespersen,1,2 Peter A. Mirau,*,1 Ernst von Meerwall,3 Richard A. Vaia,1 Robert Rodriguez,4,5 Nikhil J. Fernandes,4 and Emmanuel P. Giannelis4 1Materials

and Manufacturing Directorate, Air Force Research Laboratory, Wright-Patterson AFB, OH 45433 2UES, Inc., Dayton, OH 45432 3Department of Physics, University of Akron, Akron, OH 44325 4Department of Materials Science and Engineering and School of Applied and Engineering Physics, Cornell University, Ithaca, NY 14853 5Current address: Intel Corp., 2501 NW 229th Ave., Hillsoboro, OR 97124 *E-mail: [email protected]

Nanoscale ionic materials (NIMs) are organic-inorganic hybrids in which a core nanoparticle is functionalized with a covalently attached corona and an ionically tethered polymer canopy. NIMs exhibit liquid-like character under ambient conditions in the absence of solvent and are of interest for a variety of applications. We have used nuclear magnetic resonance (NMR) relaxation and pulsed-field gradient (PFG) diffusion experiments to measure the canopy dynamics of NIMs prepared from 18-nm silica nanoparticles. NMR studies show that the fast (ns) local dynamics of the canopy are insensitive to the presence of the silica nanoparticles. Canopy diffusion in the NIMs is slowed relative to the neat copolymer, but not all canopy molecules are slowed equally due to crowding at the nanoparticle surface, resulting in a strongly bound fraction at the surface and a weakly bound outer sphere. Electrostatic interactions with other ionic (Na+) species alter the dynamics by screening interactions with the nanoparticle.

© 2011 American Chemical Society In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


Introduction Nanoscale ionic materials (NIMs) are an emerging class of functionalized nanoparticles that have recently generated a great deal of interest due to their unique characteristics (1–3). NIMs are organic-inorganic hybrids consisting of a core nanoparticle functionalized with a covalently bound ionic corona and a bulky counter-ion as the canopy (Figure 1). NIMs are nanoparticles stabilized by an organic ionic liquid coating and either behave as liquids at room temperature or undergo reversible, macroscopic solid-to-liquid transitions near room temperature in the absence of solvent. The library of NIMs reported to date has grown to include NIMs based on metal oxides (SiO2 (1, 4, 5), Fe2O3 (5), TiO2 (2), and ZnO (6)), metals (Au (7–9), Pt (7, 9), Pd (7), and Rh (7)), quantum dots (10), carbon nanotubes (11, 12), nanorods (13) and fullerols (14). The room temperature liquid character of NIMs allows for the design of liquids that retain the unique size-dependent properties of the core nanoparticles (e.g., magnetic fluids with high viscosity, etc.). Exploiting the versatility of the NIMs platform in widespread applications (9, 15, 16) will depend on the degree to which NIMs properties can be tuned by modifying their structure and composition (core shape and size, canopy composition and structure, and ionic content). Flexibility in NIMs design also presents an opportunity to investigate the influence of structure and composition on the dynamics of the canopy, which will likely have a strong impact on the macroscopic properties of NIMs. In this study we report the NMR relaxation and diffusion studies of silicabased NIMs with a polymer canopy in order to determine the relationship between chemical structure and dynamics. The results show how the properties of the canopy relate to the macroscopic properties.

Experimental Section Materials Silica nanoparticles (LUDOX® HS-30 colloidal silica, 30 wt. % suspension in H2O, 18-nm diameter, Aldrich Chemical Co.), 3-(Trihydroxysilyl)-1propanesulfonic acid (30-35 wt. % solution in H2O, Gelest, Inc.), and Jeffamine® M-2070 Polyetheramine (Mn=2263, Mw=2334, PDI=1.03, Huntsman Corporation ) were used as received. Deionized water (18.2 MΩ-cm) was purified using a Barnstead Nanopure system. NIMs Preparation Silica nanoparticles were functionalized according to a previously reported procedure using (trihydroxysilyl)-1-propanesulfonic acid (SIT, 30-35 wt. % in deionized water) (1). The functionalized nanoparticles were purified by dialysis to remove excess SIT and then stirred in the presence of an ion exchange resin (HCR-W2) for at least 48 hours in order to exchange sodium atoms for protons. NIMs were prepared by titration of the bound sulfonic acid groups with M-2070 in deionized water. The equivalence point of the titration represents 150 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.

a 1:1 ratio of canopy to bound corona. The canopy:corona ratio can be tuned by adding the appropriate amount of M-2070 relative to that required to reach stoichiometric equivalence. Samples were prepared at 100% (NIMs100) and 60% (NIMs60) neutralization of the anionic surface of the silica nanoparticles. Once the appropriate amount of M-2070 had been added to generate the targeted corona:canopy ratio, solvent was removed under vacuum at 35-45 °C for several days, resulting in a pale yellow product that flows as a viscous liquid at room temperature. NMR Characterization Downloaded by PURDUE UNIVERSITY on August 5, 2013 | http://pubs.acs.org Publication Date (Web): October 14, 2011 | doi: 10.1021/bk-2011-1077.ch009

13C

relaxation experiments were conducted using a Tecmag Apollo 500 MHz NMR spectrometer equipped with a DOTY Scientific magic-angle spinning probe. Diffusion coefficients were measured at 50.5 °C using proton pulsed-gradient spinecho NMR as previously described (17) using the stimulated echo sequence with pulsed magnetic field gradients of magnitude G = 652 Gauss/cm. The magnetization decay as a function of gradient field strength was used to determine the diffusion coefficient. In the case of a single diffusing species, the signal decay is given by (18, 19)

where γ is the gyromagnetic ratio of the nucleus of interest, δ is the time of the gradient pulse, Δ is the diffusion delay time in the pulse sequence, G is the gradient strength, and the diffusion coefficient D is related to the hydrodynamic radius rH by the Stoke-Einstein equation (19, 20).

In those cases where single-exponential decay is not observed (vide infra), we have chosen to fit the data to a stretched exponential function, given by

where β is a fitting parameter relating to the deviation from single-exponential behavior. In cases where the stretched exponential did not give a good fit, the data was fit to a double exponential function, given by

where the subscripts f and s refer to fast and slow diffusion processes.

Results and Discussion 13C

The dynamics of the NIMs corona-canopy interface were evaluated from the spin-lattice relaxation times and the proton diffusion coefficients measured 151

In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


by pulsed-field gradient (PFG) NMR. The NIMs of interest consist of an 18-nm silica nanoparticle core functionalized with a monolayer of trihydroxysilylpropyl sulfonic acid (SIT). The anionic SIT corona is paired with an amine-terminated (cationic) ethylene oxide/propylene oxide diblock copolymer canopy (Figure 1). NIMs have been described as monolithic hybrid materials with liquid properties, where each nanostructure “carries” its share of the solvent (1, 21). For this reason, a canopy: corona ratio of 1:1 was chosen as the most reasonable starting point for examining the molecular-level dynamics of NIMs using NMR. The first step in characterizing the dynamics by NMR relaxation is measurement of the T1 relaxation across a range of temperatures and identification of the 13C T1 minimum. The T1 minimum occurs when the molecular motions are near the spectrometer frequency (125 MHz) (22). The dynamics of the free canopy (M-2070) and the NIMs canopy in neat, dry samples were measured from the carbon T1 of the methylene carbon peak at 74 ppm, which is attributed primarily to the ethylene oxide segments. Figure 2 compares the temperature dependence of the carbon T1‘s for the bulk M-2070 and NIMs samples prepared with 100% and 60% neutralization of the sulfonic acid groups. The most remarkable feature of this plot is that the relaxation times versus temperature are identical and that the T1 minimum occurs at the same temperature for the NIMs100 and NIMs60 samples as for the bulk M-2070.

Figure 1. NIMs diagram showing the core, corona and canopy. A more quantitative understanding of canopy dynamics in NIMs can be obtained by fitting the temperature dependence of the relaxation to a specific model. A number of models for the C-H autocorrelation function were explored to determine the rotational correlation times from the T1 relaxation data. We observed that the T1 minimum (0.1 s) calculated from the isotropic reorientation model for the C-H autocorrelation function did not adequately predict the 152 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


experimentally observed T1 minimum (0.26 s) for either the neat canopy or the NIMs. Given that the canopy molecules are smaller than an entangled polymer in the melt, we have chosen to use the Lipari-Szabo (LS) model (23, 24) for the autocorrelation function rather than a more sophisticated model like the modified KWW model (25) commonly used for high molecular weight polymers. The key assumption of the LS model is that the overall molecular reorientation is decoupled from faster internal motions, and the autocorrelation function C(t) is given by the product of the correlation functions for overall reorientation C0(t) and the internal motion Ci(t) as

The spectral densities in the LS model are given by

where S2 is the generalized order parameter and where τi is the correlation time for rapid librational motions and τr is the rotational correlation time. The LS model with a generalized order parameter of 0.4 gave a good fit to the T1 minimum (268 K) and was used to calculate correlation times as a function of temperature. The key conclusions from the T1 fits are that the carbon relaxation is due to a combination of rapid (10 ps) librational motions and slower (0.1-2 ns) reorientation. The local dynamics do not appear to depend on the presence of the 18-nm silica nanoparticle for NIMs100 and NIMs60. The silica nanoparticles in the NIMs are functionalized at nearly every surface hydroxyl group on the silica nanoparticle, resulting in an estimated corona density of 4.5/nm2. At this density, the average distance between between acid groups is less than the radius of gyration of the canopy (1.25 nm) calculated for a freely jointed chain (26). The resultant crowding could result in stretched chain configurations extending away from the surface, analogous to a brush. At lower canopy densities than those studied here, configurations where the chain folds back onto the nanoparticle surface, allowing interactions between the oxygen atoms in the ethylene oxide/propylene oxide monomers and the sulfonic acid groups or water molecules strongly bound to the sulfonic acids, are more likely. We are currently investigating systems with the possibility for these interactions, which we would expect to result in slower local dynamics. One of the key assumptions in NIMs design is that each functionalized nanoparticle carries its share of the solvent (i.e., the canopy) because the charge on the ionic terminal functionality of the corona is balanced by a strongly associated, oppositely-charged canopy molecule (1, 21). In the case of strong association of the canopy with the nanoparticle surface, significant differences between the diffusion coefficients measured for the bulk M-2070 and NIMs canopy would be expected. To evaluate this hypothesis, we have measured the canopy diffusion 153 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


coefficients using the stimulated-echo PFG pulse sequence with corrections for gradient artifacts (27). The first measure of the interaction between the anionic terminal functionality on the nanoparticle and the cationic corona is to compare the diffusion coefficients in solutions of the canopy and the NIMs. Figure 3 compares the diffusion curves for 8% (w/w) solutions of the canopy and the NIMs in D2O at 50°C. The most notable feature is that the diffusion coefficients are very similar for the two samples. In both cases, the data was fit to a stretched exponential function and the results are listed in Table I (28).

Figure 2. The carbon T1 vs. temperature for M-2070 (●), NIMs100 (▵), and NIMs60 (▾). The magnetization decay for M-2070 in D2O appears as a single exponential (β=0.98) within experimental error, as expected for a nearly monodisperse polymer (PDI=1.03) (29). The measured diffusion coefficient for M-2070 closely agrees with the value of 2.1 x 10-6 cm2/s calculated from the Stokes-Einstein relation using the viscosity of D2O at 50 °C and a hydrodynamic radius of 1.25 nm, calculated assuming a freely jointed chain (26). We note that rH =1.25 nm compares favorably to the value determined from dynamic light scattering experiments (1.15 nm). By comparison, the decay curve for the NIMs100 canopy shows only a 15% decrease in the diffusion coefficient compared to dissolved M-2070. We can estimate the diffusion coefficient expected if the canopy molecules were bound to the nanoparticles by calculating the hydrodynamic radius for the silica nanoparticle (9 nm), the corona (0.5 nm) and the canopy (2 x 1.25 nm). This gives a hard-sphere radius of 12 nm, from which we would expect the 154 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


diffusion coefficient to decrease by a factor of 9.6 relative to a dissolved canopy molecule in D2O. The fact that we observe only a 15% decrease in the diffusion coefficient shows that the NIMs canopy does not undergo hard-sphere diffusion in solution. Rather, the canopy is exchanging between free molecules and other nanoparticles in D2O on the time scale (ms) of the diffusion measurements.

Figure 3. PFG NMR diffusion for M-2070 (●) and NIMs100 (▵) in D2O. (Reproduced with permission from reference (28). Copyright 2010 American Chemical Society.)

Table I. Self-diffusion coefficients and fitting parameters for the canopy and NIMs in D2O at 50°C D (cm2/s)

Sample

β

M-2070

1.84 ×

10-6

0.98

NIMs100

1.56× 10-6

0.73

Figure 4 compares the self-diffusion decay curves for neat, dry samples of the free canopy, NIMs100, and NIMs100 spiked with NaCl (SIT:Na=1:1). We observe curvature on the semilog plots and good fits to the stretched exponential function for free canopy and Na+-spiked NIMs100. The data for NIMs 100 is fit by a double exponential function, and the diffusion coefficients and fit parameters are listed in Table II. 155 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


Figure 4. PFG diffusion for bulk M-2070 (●), NIMs100 (▵) and NIMs100 spiked with NaCl (▴).

Table II. Self-diffusion coefficients for neat M-2070, NIMs100 and NIMs100 spiked with NaCl Sample Bulk M-2070 NIMs100 NIMs100/NaCl

D (cm2/s)

β

5.3×10-8

0.83

3.2×10-8 (fast) 1.4×10-9 (slow)

--

1.3×10-8

0.90

Even though we observed single exponential decay for M-2070 in D2O solution, Figure 4 and Table II show that a stretched exponential function (β=0.83) is required to fit the data for the neat canopy. Since the solution experiments eliminate polydispersity as an explanation for curvature in the signal decay, we believe the curvature results from self-association of M-2070 in the neat state, which does not occur when the molecule is dissolved in D2O at 50 °C (Figure 3). We note that M-2070 is similar in structure to Pluronic block copolymers, which are known to self-associate in solution, giving rise to distributions in diffusion coefficients (30). We have also observed self-association of M-2070 in dilute solutions of by GPC and in both dilute solutions of M-2070 and neat M-2070 by small angle x-ray scattering (not shown). 156 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


Figure 5. Diagram showing the crowding of polymer chains at the surface of the nanoparticle. Canopy diffusion in NIMs100 is significantly slower than in bulk M-2070 and is best fit with a double exponential function. The double-exponential nature of the curve demonstrates that there are two populations of canopy polymers, one of which is not exchanging on the time scale (ms) of the diffusion experiment. We believe this can be explained by polymer crowding at the particle surface because the radius of gyration of the M-2070 is larger than the spacing between anionic groups on the nanoparticle surface, as shown schematically in Figure 5. A balance of electrostatic attractions between sulfonic acid and amine groups and entropic repulsion arising from stretched chain conformations would result in a diffuse electrostatic coupling zone extending away from the anionic corona. This hypothesis is further supported by the observation that the slow diffusion coefficient (1.4 x 10-9 cm2/s) gives a hard-sphere radius of 14 nm, which closely agrees with the estimated hard-sphere radius (12 nm), defined by the silica nanoparticle diameter plus the SIT layer and two times the hydrodynamic radius of the canopy. The outer sphere canopy molecules are attracted to the surface by Coulombic interactions, but they are unable to reach the surface due to the molecular crowding. We note that the fast diffusion in the NIMs100 sample is similar to that of bulk M-2070. We previously reported that the diffusion 157 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


coefficients are not sensitive to the diffusion delay time, which demonstrates the diffusion is not restricted to the surface of the nanoparticle (28). Canopy exchange is slowed in neat NIMs100, but it is greatly increased in the presence of NaCl. This is illustrated in Figure 4 (▴) for the sample spiked with NaCl at a level of one sodium atom per acid group. Under these conditions, the diffusion coefficient is an average of that observed for hard-sphere diffusion and for the bulk canopy. The diffusion coefficient measured for the NaCl-spiked sample is similar to that reported for another NIMs sample which we have subsequently determined also contained approximately one sodium ion per acid group (28). This behavior suggests that the presence of small amounts of contaminant ions can have a dramatic effect on the canopy dynamics in NIMs and that the diffusion properties can potentially be tuned by controlling the concentration and valence of additional ions in the NIMs.

Conclusion In summary, we have studied the dynamics and diffusion of the organic canopy in SiO2 NIMs using a combination of NMR relaxation and pulsed-field gradient diffusion. The local molecular relaxation times are similar for the bulk canopy and the completely neutralized NIMs, suggesting that the dynamics are liquid-like at the surface of the nanoparticle. The diffusion experiments reveal two different populations of canopy polymers with different translational diffusion characteristics. The canopy polymers strongly tethered to the surface exhibit hard-sphere-like diffusion, whereas the outer sphere layer of the canopy exhibits bulk-like M-2070 diffusion. These data are explained with a model in which canopy crowding at the nanoparticle surface prevents all of the amine-terminated canopy from interacting directly with the acid groups on the surface. We are currently investigating the effects of canopy molecular weight, canopy:corona ratios, nanoparticle size, and contaminant ion concentrations on the dynamics and rheology of NIMs.

Acknowledgments Funding provided by the Air Force Office of Scientific Research is gratefully acknowledged. The diffusion portion of this work was supported by the National Science Foundation under Grant No. DMR 04 55117. This publication is based on work supported by Award No. KUS-C1-018-02, made by King Abdullah University of Science and Technology (KAUST). A portion of this research was carried out while M. Jespersen was a National Research Council Associate at the Air Force Research Laboratory. Jeffamine M-2070 was generously donated by Huntsman Corporation (Houston, TX). The authors would like to thank George Fultz and Timothy Reid (University of Dayton Research Institute) for viscosity and ICP-MS data supporting this research. M. Tchoul and H. Koerner (AFRL/RX) contributed GPC and SAXS data, respectively, in support of this study. The authors also thankfully acknowledge Dr. Rajiv Berry and Phuong Ngo (AFRL/RX) for helpful discussions regarding this work. 158 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.

References 1. 2. 3. 4.


5. 6. 7. 8. 9.

10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25.

Rodriguez, R.; Herrera, R.; Archer, L. A.; Giannelis, E. P. Adv. Mater. 2008, 20, 4353–4358. Bourlinos, A. B.; Chowdhury, S. R.; Herrera, R.; Jiang, D. D.; Zhang, Q.; Archer, L. A.; Giannelis, E. P. Adv. Funct. Mater. 2005, 15, 1285–1290. Rodriguez, R.; Herrera, R.; Bourlinos, A. B.; Li, R.; Amassian, A.; Archer, L. A.; Giannelis, E. P. Appl. Organomet. Chem. 2010, 24, 581–589. Bourlinos, A. B.; Chowdhury, S. R.; Jiang, D. D.; An, Y.-U.; Zhang, Q.; Archer, L. A.; Giannelis, E. P. Small 2005, 1, 80–82. Bourlinos, A. B.; Herrera, R.; Chalkias, N.; Jiang, D. D.; Zhang, Q.; Archer, L. A.; Giannelis, E. P. Adv. Mater. 2005, 17, 234–237. Bourlinos, A. B.; Stassinopoulos, A.; Anglos, D.; Herrera, R.; Anastasiadis, S. H.; Petridis, D.; Giannelis, E. P. Small 2006, 2, 513–516. Warren, S. C.; Banholzer, M. J.; Slaughter, L. S.; Giannelis, E. P.; DiSalvo, F. J.; Wiesner, U. B. J. Am. Chem. Soc. 2006, 128, 12074–12075. Zheng, Y.; Zhang, J.; Lan, L.; Yu, P.; Rodriguez, R.; Herrera, R.; Wang, D.; Giannelis, E. P. ChemPhysChem 2010, 11, 61–64. Voevodin, A. A.; Vaia, R. A.; Patton, S. T.; Diamanti, S.; Pender, M.; Yoonessi, M.; Brubaker, J.; Hu, J.-J.; Sanders, J. H.; Phillips, B. S.; MacCuspie, R. I. Small 2007, 3, 1957–1963. Sun, L.; Fang, J.; Reed, J. C.; Estevez, L.; Bartnik, A. C.; Hyun, B.-R.; Wise, F. W.; Malliaras, G. G.; Giannelis, E. P. Small 2010, 6, 638–641. Bourlinos, A. B.; Georgakilas, V.; Boukos, N.; Dallas, P.; Trapalis, C.; Giannelis, E. P. Carbon 2007, 45, 1583–1595. Bourlinos, A. B.; Georgakilas, V.; Tzitzios, V.; Boukos, N.; Herrera, R.; Giannelis, E. P. Small 2006, 2, 1118–1191. Bhattacharjee, R. R.; Li, R.; Estevez, L.; Smilgies, D.-M.; Amassian, A.; Giannelis, E. P. J. Mater. Chem. 2009, 19, 8728–8731. Fernandes, N.; Dallas, P.; Rodriguez, R.; Bourlinos, A. B.; Georgakilas, V.; Eiannelis, E. P. Nanoscale 2010, 2, 1653–1656. Eastman, J. A.; Choi, S. U. S.; Li, S.; Yu, W.; Thompson, L. J. Appl. Phys. Lett. 2001, 78, 718–720. Park, S. I.; Lim, J. H.; Kim, J. H.; Yun, H. I.; Roh, J. S.; Kim, C. G.; Kim, C. O. Phys. Status Solidi B 2004, 241, 1662–1664. Iannacchione, G.; Von Meerwall, E. J. Polym. Sci., Part B: Polym. Phys. 1991, 29, 659–668. Price, W. S. Concepts Magn. Reson. 1998, 104, 299–336. Johnson, C. S. Prog. Nucl. Magn. Reson. Spectrosc. 1999, 34, 203–256. Price, W. S. Concepts Magn. Reson. 1998, 104, 197–237. Agarwal, P.; Qi, H.; Archer, L. A. Nano Lett. 2010, 10, 111–115. Mirau, P. A Practical Guide to the NMR of Polymers; John Wiley & Sons: Hoboken, 2004; p 418. Lipari, G.; Szabo, A. J. Am. Chem. Soc. 1982, 104, 4546–4559. Lipari, G.; Szabo, A. J. Am. Chem. Soc. 1982, 104, 4559–4570. Qiu, X.; Moe, N. E.; Ediger, M. D.; Fetters, L. J. J. Chem. Phys. 2000, 113, 2918–2926. 159

In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.


26. Hiemenz, P. C. Polymer Chemistry, The Basic Concepts; Marcel Dekker, Inc.: New York, 1984. 27. Von Meerwall, E.; Kmat, M. J. Magn. Reson. 1989, 83, 309–323. 28. Jespersen, M. L.; Mirau, P. A.; von Meerwall, E.; Vaia, R. A.; Rodriguez, R.; Giannelis, E. P. ACS Nano 2010, 4, 3735–3742. 29. Hakansson, B.; Nyden, M.; Soderman, O. Colloid Polym. Sci. 2000, 278, 399–405. 30. Nilsson, M.; Hakansson, B.; Soderman, O.; Topgaard, D. Macromolecules 2007, 40, 8250–8258.

160 In NMR Spectroscopy of Polymers: Innovative Strategies for Complex Macromolecules; Cheng, H., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 2011.

bk-2011-1077.ch009

Recommend Documents