Electrostatic Control of Spin Exchange Between Mobile Spin

Apr 8, 2011 - Time-Resolved Electron Paramagnetic Resonance Spectroscopy. Malcolm D.E. Forbes , Lauren E. Jarocha , SooYeon Sim , Valery F. Tarasov...
0 downloads 0 Views 4MB Size
ARTICLE pubs.acs.org/Langmuir

Electrostatic Control of Spin Exchange Between Mobile Spin-Correlated Radical Pairs Created in Micellar Solutions Paula Caregnato,† Lauren E. Jarocha,‡ Hali S. Esinhart,‡ Natalia V. Lebedeva,‡ Valery F. Tarasov,§ and Malcolm D. E. Forbes*,‡ †

Instituto de Investigaciones Fisicoquímicas Teoricas y Aplicadas (INIFTA), Facultad de Ciencias Exactas, Universidad Nacional de la Plata, C.C. 16, suc. 4, (1900) La Plata, Argentina ‡ Caudill Laboratories, Department of Chemistry, CB #3290, University of North Carolina, Chapel Hill, North Carolina 27599-3290, United States § Semenov Institute of Chemical Physics, Kosygin St 4, Moscow 119991, Russia ABSTRACT: A series of photoinduced H-atom abstraction reactions between anthraquinone-2,6,-disulfonate, disodium salt (AQDS) and differently charged micellar substrates is presented. After a 248 nm excimer laser flash, the first excited triplet state of AQDS is rapidly formed and then quenched by abstraction of a hydrogen atom from the alkyl chain of the micelle surfactant, leading to a spin-correlated radical pair (SCRP). The SCRP is detected 500 ns after the laser flash using time-resolved (direct detection) electron paramagnetic resonance (TREPR) spectroscopy at X-band (9.5 GHz). By changing the charge on the surfactant headgroup from negative (sodium dodecyl sulfate, SDS) to positive (dodecyltrimethylammonium chloride, DTAC), TREPR spectra with different degrees of antiphase structure (APS) in their line shape were observed. The first derivative-like APS line shape is the signature of an SCRP experiencing an electron spin exchange interaction between the radical centers, which was clearly observable in DTAC micelles and absent in SDS micellar solutions. Solutions with surfactant concentrations well below the critical micelle concentration (cmc) or solutions where micellar formation had been disrupted (1:1 v/v CH3CN/H2O) also showed no APS line shapes in their TREPR spectra. These results support the conclusion that electrostatic forces between the sensitizer (AQDS) charge and the substrate (surfactant) headgroup charge are responsible for the observed effects. The results represent a new example of electrostatic control of a spin exchange interaction in mobile radical pairs.

’ INTRODUCTION The control of magnetic and electronic interactions in molecules, and between them, is a topic of current interest in the fields of organic molecular magnetism1 and spintronics.2 There are numerous reports of attempts to control the sign and magnitude of exchange and/or dipolar couplings in the solid state,3 in frozen matrices,4 in host guest complexes,5 and in rigid molecular frameworks in solution.6 However, there are few studies of control of such interactions in systems with high mobility, such as radical pairs that are created in a micellar or reverse micellar environment. These are called mobile spin-correlated radical pairs (SCRPs),7 and they are amenable to study by time-resolved magnetic resonance techniques.8 The relationship between spin and molecular dynamics (diffusion and rotation) in SCRPs has been a focal point of our research program for many years.9 Time-resolved electron paramagnetic resonance (TREPR) spectroscopy gives immediate information regarding interactions between the radicals of a micellar SCRP, in the form of a splitting due to electron spin exchange coupling (J). The average splitting is typically smaller than or comparable to C-centered radical hyperfine interactions (1 20 G), and it removes the double r 2011 American Chemical Society

degeneracy of the EPR transitions of a pair of magnetically nonequivalent electron spins. The observed additional splitting of the EPR spectral lines is similar to that observed in some stable nitroxide biradicals.10 The new transitions that are now resolved due to this interaction have polarizations of opposite phase, leading to a first derivative-type line shape called antiphase structure (APS, Figure 1).11 It is important to note that the APS arises from two EPR transitions, one absorptive (A) and one emissive (E), and is distinct from the “normal” first derivative line shape in a single absorptive transition, typically observed in steady-state EPR spectroscopy using 100 kHz field modulation. The spectral pattern in Figure 1 is analogous to an AX-type NMR spectrum for a spin 1/2 nucleus, with the only difference being that both transitions are polarized equally (but with opposite phases). For TREPR spectra of organic radicals with multiple hyperfine interactions, each individual line in the spectrum will show a similar splitting to those illustrated in Figure 1. The Received: February 15, 2011 Revised: March 24, 2011 Published: April 08, 2011 5304

dx.doi.org/10.1021/la2005997 | Langmuir 2011, 27, 5304–5309

Langmuir

Figure 1. Calculated direct detection TREPR spectrum showing antiphase structure in a mobile spin-correlated radical pair. Each radical experiences a splitting (into a doublet) due to Heisenberg-type spin exchange, with the two components of each doublet exhibiting opposite phase (low field E, high field A). The result is a spectrum with first derivative-like line shapes, but should not be confused with a steady-state EPR spectrum that would arise from two lines (one for 3 R1 and one for 3 R2) experiencing field modulation during detection.

resulting spectrum is a complex pattern of E/A doublets that can often overlap. The strong chemically induced electron spin polarization (CIDEP) in SCRPs, combined with the opposite phases of the transitions, renders the APS observable even when the absolute magnitude of the splitting (defined qualitatively here as the distance between the extrema within each radical signal in Figure 1) is much smaller than the line width. This provides an advantage for studying exchange or dipolar couplings in transient radical pairs compared to nitroxide biradicals,12 particularly those with large end-to-end distances between the radical centers. Turro and co-workers have used optical detection methods to study the location, mobility, and CIDEP patterns of neutral free radicals created in micelles.13 They found that hydrophobic effects dominated the escape rate of small radicals from SDS micelles. Hydrophobic effects also dominate the location of nitroxide spin probes in supramolecular assemblies.14 Several research groups have investigated electron transfer reactions in micelles or across micellar boundaries and found electrostatic effects on reaction rates.15 van Willigen and Levstein reported Fourier transform EPR spectra of charged SCRPs, created by ET reactions in or near micelles, which clearly show the influence of those charges on the manifestations of the exchange interaction in the relevant TREPR spectra.16 In this paper, we report TREPR spectra for charged SCRPs created by H-atom abstraction reactions, which are an important addition to this data set because the creation of the radical pair must take place in or near the micelle interior rather than across an interface as in electron transfer reactions. The photochemistry leading to the SCRPs studied here is outlined in Scheme 1. Solutions of anthraquinone-2,6-disulfonate, disodium salt (AQDS) in aqueous micellar solution are irradiated at 308 or 248 nm, which leads to

ARTICLE

H-atom abstraction by the AQDS excited triplet state from the alkyl chain of the micelle monomer (The AQDS triplet state is inferred but not shown in Scheme 1). We have purposely drawn the AQDS molecule outside the micelle, as it is unlikely to be enclosed in the interior with two negative charges. The possible surfactant radical structures resulting from this photochemistry are shown in Chart 1. While the AQDS ketyl radical is common to all pathways, there are several possible alkyl radical structures. As noted previously by us,7 secondary radicals 1a and 1b are observed with essentially statistical intensity ratios (i.e., radical 1a structures represent abstraction reactions at C2 C10, whereas radical 1b is strictly from C11). Secondary radical 1c is not observed, presumably because it is destabilized by the neighboring anion. Primary radical 1d is also not observed, as it is less stable based on a hyperconjugative argument. Below, we will demonstrate that the spin exchange interaction between the radicals making up this SCRP can be controlled on the sub-microsecond time scale, and even turned off completely, via electrostatic interactions between the two charged radicals. The present study is restricted in scope to the effect of the charge of the surfactant headgroup: sulfate (anionic) versus dodecyltrimethylammonium (cationic), and the one doubly charged chromophore.

’ EXPERIMENTAL SECTION Materials. Anthraquinone-2,6-disufonate, disodium salt (AQDS), sodium dodecyl sulfate (SDS), dodecyltrimethylammonium chloride (DTAC), and acetonitrile (Aldrich) were used as received. Ultrapure water from a Millipore system was used for all sample preparations and to flush the TREPR flow system between experiments. The concentration of AQDS in each sample was 1.1 mM, while the concentration of SDS or DTAC was 0.05 M, well above their respective critical micellar concentrations. Samples were prepared by dissolving AQDS and surfactant in the Millipore water and then sonicating for several minutes followed by sitirring/degassing. TREPR Spectroscopy. Our TREPR apparatus has been described previously in several recent publications.17 Briefly, the 308 or 248 nm excimer laser is fired at a repetition rate of 60 Hz while sampling the direct detection EPR signal from the microwave bridge (CW mode) using a gated boxcar signal averager. The external magnetic field is swept over 2 or 4 min with 100 or 300 ns gates sampling the EPR signal 5 10 times at each magnetic field point. Samples were flowed through the microwave resonator using a micropump from a reservoir that was constantly purged with nitrogen gas bubbles (for 10 min prior to and during EPR experiments). All spectra have a center field of approximately 3375 G, microwave frequency of 9.47 GHz, and microwave power of 10 mW. Transitions appearing below the baseline in each spectrum represent emissive spin polarization, while those above the baseline exhibit enhanced absorption.

’ RESULTS AND DISCUSSION Figure 2A shows the TREPR spectrum acquired after 248 nm excimer laser excitation of an aqueous solution of AQDS and SDS micelles, which have anionic headgroups. The central line that is cut off in the spectrum is due to the AQDS ketyl radical, which deprotonates in solution to give the AQDS radical anion. This radical has very small hyperfine interactions and thus exhibits an unresolved intense transition in the center (emissively polarized by the triplet mechanism). We have observed this radical in many previous experiments where water-soluble SCRPs are created in reverse micelles.9a The other lines in the spectrum, which also carry a small amount of net emissive polarization, are assigned to two 5305

dx.doi.org/10.1021/la2005997 |Langmuir 2011, 27, 5304–5309

Langmuir

ARTICLE

Scheme 1

Chart 1

Figure 2. X-band TREPR spectra acquired 500 ns after a 248 flash of AQDS (1.1 mM) dissolved in aqueous solutions of (A) SDS and (B) DTAC micelles (both 50 mM). The central line in the center of the spectra is due to radical 2 and has been purposely chopped so that vertically expanded signals of radicals 1b (marked with asterisks) and radical 1a (all other transitions) can be clearly observed. See Scheme 1 and Chart 1 for photochemical mechanism and radical structures.

different types of secondary alkyl radical whose structures are shown in Chart 1 (radicals 1a and 1b). It is immediately apparent that the spectrum in Figure 2A shows only normal near-Lorentzian line shapes typically observed for absorptive (A, at high field) or emissive (E, at low field) transitions in TREPR of monoradicals, with no evidence for any APS line shape. The low field E/high field A multiplet polarization pattern observed is from the radical pair mechanism (RPM)18 of CIDEP due to S T0 mixing in the radical pair state originating from a triplet precursor with a negative exchange interaction. There is also a small net polarization (E) superimposed from the triplet mechanism (TM), which is known to be strong for many radical reactions involving AQDS. Figure 2A represents the TREPR spectrum expected of radical pairs that are not interacting at the time of observation; that is, they show no indication that both radicals are experiencing any kind of restricted diffusion.

The same photochemical reaction was run with micelles made from DTAC molecules, which have cationic headgroups but the same chain length as SDS. The formation of DTAC micelles takes place at a slightly higher critical micellar concentration (cmc), but the aggregation number (Na) of DTAC at 50 mM is very similar to that for SDS (Na = ∼ 60)19 that is, the only major structural difference between the DTAC and SDS micelles at this concentration and temperature is the headgroup charge. The TREPR spectrum obtained with DTAC micelles, shown in Figure 2B, exhibits significant differences from that in Figure 2A. Although the same radicals are produced, there is an intense APS line shape (E/A doublets), and the overall line widths are somewhat broader for each transition than those observed in Figure 2A for SDS. The effect of the positively charged surfactant head groups on the spectrum in Figure 2 5306

dx.doi.org/10.1021/la2005997 |Langmuir 2011, 27, 5304–5309

Langmuir

Figure 3. X-band TREPR spectra acquired 500 ns after a 248 flash of AQDS (1.1 mM) dissolved in aqueous solutions of SDS at (A) 4 mM and (B) 50 mM. Intensities are normalized to the highest alkyl radical peak hight in (B), and all other experimental conditions (data collection settings and temperature) were identical. The AQDS anion spectrum in the center has been purposely cut off in (B).

provides strong evidence for control of the magnitude of the spin exchange interaction between two mobile radical pairs, based on electrostatic attraction versus repulsion between the AQDS chromophore and the micelle headgroup. It could be argued that, in the case of extensive repulsion between AQDS and the micelle exterior, the photochemistry may be taking place only in free solution between the “free” AQDS and individual surfactant molecules that are in the process of exchanging between micelles. The maximum concentration of nonmicellar surfactant molecules in free solution must be well below the cmc, which is approximately 8 mM for SDS20 and 22 mM for DTAC.21 Indeed, it is likely to be below or near the critical aggregate concentration of SDS, which is reported in the literature to be 4 mM or less.22 Figure 3 shows a comparison of the signal intensities for two solutions that are 1.1 mM in AQDS, with 4 mM SDS (Figure 3A) and 50 mM SDS (Figure 3B, same as Figure 1A). It should also be noted that when the SDS concentration is lowered to 1 mM, no alkyl radicals are observed from this photochemistry, only a weak central line. We therefore conclude that the majority of the alkyl radical signal intensity in Figure 1A arises from micellar radicals. We also attempted to run similar experiments below the cmc of DTAC. However, a rather interesting phenomenon prevented TREPR experiments from being carried out on such samples. As they were being prepared at high dilution, a precipitate formed that was determined by NMR spectroscopy to be the salt of DTAC and AQDS. This salt would easily redissolve in higher concentrations of DTAC, that is, above the cmc. The phenomenon of salt formation between AQDS and surfactant cations is under further exploration in our laboratory, as it may provide a mechanism for drastically altering the distribution of charge species around the exterior of DTAC micelles. At the present time, we are unable to collect TREPR spectra for DTAC at low concentrations for this reason.

ARTICLE

Figure 4. X-band TREPR spectra acquired 500 ns after a 248 flash of AQDS (1.1 mM) dissolved in 1:1 acetonitrile/water (v/v) solutions of (A) SDS (50 mM) and (B) DTAC (50 mM). (C) Computer simulation of the alkyl radicals, with hyperfine coupling parameters for 1a, 20.6 G (1 HR) and 24.9 G (4 Hβ), and for 1b, 20.4 G (1 HR) and 24.4 G (5 Hβ). Note the much lower intensity of the central transition in (B) for DTAC (see text for discussion). The AQDS radical has been omitted from the simulation for clarity.

Figure 4 shows TREPR spectra obtained after photolysis of AQDS in solutions of SDS and DTAC that have 50% acetonitrile added to the aqueous solution. In this solvent system, micelles do not form,23 and it is clear that only radical pair mechanism (RPM) polarization from noninteracting radicals is observed. This spectrum is the same at all delay times, and the concentration of surfactant (50 mM) and AQDS (1.1 mM) is the same in these experiments as in the micellar systems in Figure 2. Note the similarity in spectral features of these two spectra with the AQDS/SDS spectrum shown in Figure 2A. The results in Figures 3 and 4 support our hypothesis that the magnitude of the spin exchange interaction between the radicals of the SCRP in Scheme 1 can be controlled, but only under conditions where micelles exist. There is one other interesting feature in Figure 4 regarding relative signal intensities of the alkyl radicals and the AQDS anion radical. When DTAC is the substrate (Figre 4B), the central line is much less intense than for SDS (Figure 4A). We suggest that this discrepancy arises due to an absorptive component in the central line in Figure 4B, originating from a separate photophysical process involving exciplex formation between AQDS and the chloride counterion. This process has been demonstrated to produce absorptive CIDEP for the monosulfonated analog of AQDS (anthraquinone-2-sulfonate, AQS) and chloride ion by Moribe et al.15b The observed salt formation between DTAC and AQDS discussed above shows that significant ground state interaction can exist between these two molecules, and this is likely to be present in the excited state as well. Given that we know the H-atom abstraction reaction must still be taking place between the AQDS triplet state and the SDS micelles in Figure 2A, the question arises as to how the reaction takes place at all with electrostatic repulsions between the sensitizer and the substrate. The answer is provided by a previous computational chemistry result from our laboratory illustrated in Figure 5.24 This is a snapshot of a molecular dynamics simulation 5307

dx.doi.org/10.1021/la2005997 |Langmuir 2011, 27, 5304–5309

Langmuir

ARTICLE

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT We thank the National Science Foundation for continued support of our program through Grant No. CHE-0809530. P.C. is a research member of Consejo Nacional de Investigaciones Científicas y Tecnicas (CONICET) of Argentina and thanks them for research support. V.F.T. thanks the Russian Foundation for Basic Research for support through Grant No. 10-03-00791. ’ REFERENCES

Figure 5. Graphic showing one geometry of a SDS micelle in water from the molecular dynamics simulation in ref 24. This particular “snapshot” is taken after 160 ps of simulation. For clarity, free SDS monomers were excluded.

of an SDS micelle that has equilibrated in aqueous solution. The surrounding water molecules (27 000 of them, simulated explicitly with 60 SDS units) have been removed for clarity. The blue spheres represent the sodum counterions, which are about 30% completely solvated, in agreement with experimental determinations.19 A key feature of this snapshot is the vast number of accessible H-atoms on the alkyl chains of the surfactant (white spheres in Figure 5). Clearly, there are enough active sites without very many neighboring negative charges, so that the AQDS triplet state can access the reactive sites near the surface of the micelle. On the time scale of observation of TREPR spectra, however, electrostatic repulsion is strong enough that the AQDS radical diffuses away rapidly after its formation. The TREPR signals for AQDS/SDS are much weaker than for AQDS/DTAC, for two reasons. First, there is less chance of a reaction in the first place due to electrostatic repulsion. Second, the RPM polarization observed is generally weaker than that from the SCRP polarization. This has been observed and commented upon previously,7 and modern theoretical methods of calculating polarization intensities support this.10

’ CONCLUSIONS We have clearly demonstrated that electrostatic interactions can be used to control the magnitude of the J coupling in a mobile SCRP. Using the same photochemistry and supramolecular structures of similar size, the attraction or repulsion of radical pairs determines the level of observed interaction on the submicrosecond time scale. The APS might therefore be used as a signaling event in light-induced drug delivery processes, and the magnitude of this effect might be controllable with mixed or zwitterionic micelles or with added salts. Such studies are currently underway in our laboratory, in addition to further study on the DTAC AQDS salt precipitate and its possible role in this chemistry.

(1) (a) Kahn, O. Molecular Magnetism; VCH Publishers, Inc.: New York, 1993; p 380. (b) Roques, N.; Mugnaini, V.; Veciana, J. Magnetic and Porous Molecule-Based Materials. In Functional Metal-Organic Frameworks: Gas Storage, Separation and Catalysis; Springer-Verlag: Berlin, 2010; Vol. 293, p 207. (2) Bader, S. D.; Parkin, S. S. P. Spintronics. In Annual Review of Condensed Matter Physics, Vol 1; Annual Reviews: Palo Alto, 2010; Vol. 1, p 71. (3) Lebedeva, N. V.; Tarasov, V. F.; Resendiz, M. J. E.; Garcia Garibay, M. A.; White, R. C.; Forbes, M. D. E. J. Am. Chem. Soc. 2010, 132, 82. (4) (a) Yamauchi, S.; Takahashi, K.; Islam, S. S. M.; Ohba, Y.; Tarasov, V. J. Phys. Chem. B 2010, 114, 14559. (b) Yamauchi, S. Bull. Chem. Soc. Jpn. 2004, 77, 1255. (c) Teki, Y.; Tamekuni, H.; Haruta, K.; Takeuchi, J.; Miura, Y. J. Mater. Chem. 2008, 18, 381. (5) Krumkacheva, O. A.; Gorelik, V. R.; Bagryanskaya, E. V.; Lebedeva, N. V.; Forbes, M. D. E. Langmuir 2010, 26, 8971. (6) (a) Scott, A. M.; Miura, T.; Ricks, A. B.; Dance, Z. E. X.; Giacobbe, E. M.; Colvin, M. T.; Wasielewski, M. R. J. Am. Chem. Soc. 2009, 131, 17655. (b) Butler, A.; Solomon, G. C.; Colvin, M. T.; Scott, A. M.; Chen, K.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2010, 132, 15427. (c) Li, Y.; Lei, X.; Jockusch, S.; Chen, J. Y.-C.; Frunzi, M.; Johnson, J. A.; Lawler, R. G.; Murata, Y.; Murata, M.; Komatsu, K.; Turro, N. J. J. Chem. Soc. 2010, 132, 4042. (7) (a) Closs, G. L.; Forbes, M. D. E.; Norris, J. R. J. Phys. Chem. 1987, 91, 3592. (b) Buckley, C. D.; Hunter, D. A.; Hore, P. J.; McLauchlan, K. A. Chem. Phys. Lett. 1987, 135, 307. (8) McLauchlan, K. A. In Advanced EPR: Applications in Biology and Biochemistry; Hoff, A.J., Ed.; Elsevier: New York, 1989; p 350 ff. (9) (a) White, R. C.; Gorelik, V. R.; Bagryanskaya, E. G.; Forbes, M. D. E. Langmuir 2007, 23, 4183. (b) Tarasov, V. F.; White, R. C.; Forbes, M. D. E. Spectrochim. Acta, Part A 2006, 63, 776. (c) White, R. C.; Tarasov, V. F.; Forbes, M. D. E. Langmuir 2005, 21, 2721. (10) (a) Parmon, V. N.; Zhidomirov, G. M. Mol. Phys. 1976, 32, 613. (b) Parmon, V. N.; Zhidomirov, G. M. Mol. Phys. 1974, 27, 367. (c) Luckhurst, G. R. Mol. Phys. 1966, 10, 543. (d) Glarum, S. H.; Marshall, J. H. J. Chem. Phys. 1967, 47, 1374. (e) Hudson, A; Luckhurst, G. R. Chem. Rev. 1969, 69, 191. (f) Shultz, D. A.; Boal, A. K.; Lee, H.; Farmer, G. T. J. Org. Chem. 1999, 64, 4386. (g) Komaguchi, K.; Iida, T.; Goh, Y.; Ohshita, J.; Kunai, A.; Shiotani, M. Chem. Phys. Lett. 2004, 387, 327. (h) Sartori, E.; Khudyakov, I. V.; Lei, X.; Turro, N. J. J. Am. Chem. Soc. 2007, 129, 7785. (11) Tarasov, V. F.; Forbes, M. D. E. Spectrochim. Acta, Part A 2000, 56, 245. (12) We note here that the analysis of the steady-state EPR spectra of nitroxides is not exactly comparable to that carried out for SCRPs detected in TREPR experiments. The SCRPs studied here have both longitudinal polarization (population differences) and coherence (phase shifts). In stable nitroxide biradicals, there is no polarization and so the origin of the observed splittings has slightly different physical origin. 5308

dx.doi.org/10.1021/la2005997 |Langmuir 2011, 27, 5304–5309

Langmuir

ARTICLE

This does not affect the qualitative “J on” vs “J off” phenomenon reported here. (13) (a) Wu, C. H.; Jenks, W. S.; Koptyug, I. V.; Ghatlia, N. D.; Lipson, M.; Tarasov, V. F.; Turro, N. J. J. Am. Chem. Soc. 1993, 115, 9583. (b) Khudyakov, I. V.; McGarry, P. F.; Turro, N. J. J. Phys. Chem. 1993, 97, 13234. (14) Lebedeva, N. V.; Ranganathan, R.; Bales, B. L. J. Phys. Chem. B 2007, 111, 5781. (15) (a) Miura, T.; Kageyama, A.; Toru, S.; Murai, H. J. Phys. Chem. B 2010, 114, 14550. (b) Moribe, S.; Ikoma, T.; Akiyama, K.; Tero-Kubota, S. Chem. Phys. Lett. 2008, 457, 66. (c) Jockusch, S.; Zeika, O.; Jayaraj, N.; Ramamurthy, V.; Turro, N. J. J. Phys. Chem. Lett. 2010, 1, 2628. (d) PradoGotor, R.; Jimenez, R.; Perez-Tejeda, P.; Lopez-Cornejo, P.; Lopez-Lopez, M.; Sanchez, A.; Muriel-Delgado, F.; Sanchez, F. Prog. React. Kinet. Mech. 2000, 25, 371. (e) Gratzel, M.; Kozak, J. J.; Thomas, J. K. J. Chem. Phys. 1975, 62, 1632. (16) Levstein, P. R.; van Willigen., H. Chem. Phys. Lett. 1991, 187, 415. (17) (a) Lebedeva, N. V.; Forbes, M. D. E. In Carbon Centered Radicals and Radical Cations; Forbes, M. D. E., Ed.; Wiley Series on Reactive Intermediates in Chemistry and Biology; Wiley: New York, 2010; Vol. 3, p 323. (b) Forbes, M. D. E. Photochem. Photobiol. 1997, 65, 73. (18) (a) Harbron, E. J.; Forbes, M. D. E. In Encyclopedia of Chemical Physics and Physical Chemistry; Moore, J. H., Spencer, N. D., Eds.; Institute of Physics Publishing: Philadelphia, 2001; p 1389. (b) Spin Polarization and Magnetic Effects in Radical Reactions; Salikhov, K. M., Molin, Y. N., Sagdeev, R. Z., Buchachenko, A. L., Eds.; Elsevier: Amsterdam, 1984. (19) Bales, B. L.; Zana, R. J. Phys. Chem. B 2002, 106, 1926. (20) Mandal, A. B.; Nair, B. U.; Ramaswamy, D. Langmuir 1988, 4, 736. (21) Perger, T. M.; Bester-Rogac, M. J. Colloid Interface Sci. 2007, 313, 288. (22) Cui, X. H.; Mao, S. Z.; Liu, M. L.; Yuan, H. Z.; Du, Y. R. Langmuir 2008, 24, 10771. (23) (a) Lin, C. E.; Chiou, W. C.; Lin, W. C. J. Chromatogr., A 1996, 723, 189. (b) Misra, P. K.; Mishra, B. K.; Behera, G. B. Colloids Surf. 1991, 57, 1. (24) Bruce, C. D.; Berkowitz, M. L.; Perera, L.; Forbes, M. D. E. J. Phys. Chem. B 2002, 106, 3788.

5309

dx.doi.org/10.1021/la2005997 |Langmuir 2011, 27, 5304–5309