Macroscopic Dynamics as Reporter of Mesoscopic Organization: The

ARTICLE pubs.acs.org/JPCA

Macroscopic Dynamics as Reporter of Mesoscopic Organization: The Belousov Zhabotinsky Reaction in Aqueous Layers of DPPC Lamellar Phases Grazia Biosa,† Sandra Ristori,‡ Olivier Spalla,§ Mauro Rustici,† and Marcus J. B. Hauser||,* †

Departement of Chemistry, University of Sassari, Via Vienna 2, 07100 Sassari (SS), Italy Department of Chemistry and CSGI, University of Florence, Via della Lastruccia 3, 50019 Sesto Fiorentino (FI), Italy § CEA Saclay, DSM/IRAMIS/SIS2M/LIONS, UMR CEA-CNRS 3299, 91191 Gif-sur-Yvette, France Abteilung Biophysik, Institut f€ur Experimentelle Physik, Otto-von-Guericke Universit€at Magdeburg, Universit€atsplatz 2, 39106 Magdeburg, Germany

)

‡

ABSTRACT:

The propagation of traveling chemical waves in the excitable Belousov Zhabotinsky (BZ) system when performed in the presence of 1,2-dipalmitoyl-sn-glycero-3-phosphatidyl choline (DPPC) bilayers responds sensitively to the phospholipid content. The characteristic features of wave propagation, such as spiral pitch, rotation period, and size of the spiral core region, show two regions of different behavior, one below and the other above a DPPC content of 12.5% (w/w) thus suggesting a transition in the organization of the lipid domains at a DPPC content of ∼12.5% (w/w). This transition is supported by small-angle X-ray scattering data, which show pronounced changes in the coherence lengths of the lyotropic smectic domains. Thus, the dynamics of the chemical system occurring at a macroscopic length scale reflects the organization of the water/lipid domains which extend over mesoscopic lengths. These findings indicate that in the BZ/DPPC system, there is an interaction between processes that occurs at length scales differing by as much as 3 orders of magnitude.

’ INTRODUCTION Living systems are generally complex and show the ability to self-organize, for instance, forming excitable patterns or leading to rhythmic oscillations. Some chemical systems share such properties; however, they usually involve a considerably smaller number of components. It has been remarked very early that these features make oscillatory reactions suitable models to understand the organization of biological rhythms.1 When self-organizing chemical reactions are performed in extended and unstirred containers, their dynamics are determined by the interplay between the reactions and the involved transport processes. To suppress hydrodynamic convection, the spatiotemporal dynamics of these systems are often studied in gels, which are chemically inert with respect to the reaction. Hence, the observed dynamics stems solely from the interplay between reaction and diffusion. Pattern formation in reaction diffusion systems is widely studied,2 4 and among these patterns the behavior of spiral-shaped waves plays a prominent role. Recently, r 2011 American Chemical Society

experimental investigations have been extended to the study of the dynamics of three-dimensional patterns occurring in reaction diffusion systems.5 The Belousov Zhabotinsky (BZ) reaction has become a prototypical example for pattern formation. In its classical version, it is studied either in aqueous media or in aqueous gels.2,4 Recently, both the compartmentalization of the reaction medium and the response of excitable reactions to it have attracted scientific interest, since such systems mimic more accurately the conditions found in the natural environment of many enzymes and cells. This is also reflected in a study of a biomimetic model for the cytochrome P450 system that shows oscillations and involves the transfer of electrons and reactants between an aqueous and a membrane phase.6 Received: January 15, 2011 Revised: March 3, 2011 Published: March 24, 2011 3227

dx.doi.org/10.1021/jp2004375 | J. Phys. Chem. A 2011, 115, 3227–3232

The Journal of Physical Chemistry A Some aspects of the BZ reaction in the presence of membranes and vesicles have also been studied. Typical features of oscillatory chemical reactions, e.g., the periods and amplitudes of oscillations and the lengths of induction period, are affected by vesicles in the reaction medium. Surfactants and phospholipid bilayers exert an influence on the patterns formed in the BZ system.7 11 For instance, in the presence of sodium dodecyl sulfate (SDS) the wave velocity decreases as the SDS content is augmented and, concomitantly, the viscosity of the reaction medium increases.12 A variety of novel patterns were found when the BZ reaction was performed in a highly compartmented medium, namely, oilin-water microemulsions.13,14 These patterns include dashed waves,13 inwardly rotating spiral waves,14 Turing patterns,15 and stationary solitary structures.16 The key feature for the occurrence of these patterns is the partition of the autocatalytic species between the organic and the aqueous phase and the differences of its effective diffusion coefficients in these phases. When conducted in phospholipid lamellar phases, the BZ reaction has been found to induce structural variations of the host matrix.17 As shown by small-angle X-ray scattering (SAXS), the phospholipid bilayers retain their lamellar structure,18 but the persistence length of lyotropic smectic domains is increased with respect to comparable, nonreacting media.17 It is conjectured that this effect is due to a deeper penetration of some BZ reaction intermediates into the polar region of the phospholipid bilayer. In the present article, we investigate the dynamics of spiralshaped patterns in a BZ reaction medium containing DPPC bilayers. We focus on the dependence of the dynamical features of this excitable chemical system on the phospholipid content in the medium. Remarkably, we observe two different regimes, one occurring at a DPPC content below ∼12.5% (w/w) and the other at higher phospholipid contents. In addition, we show that this bimodal behavior is also reflected at the level of the supramolecular arrangement of the lipid domains, as evidenced by SAXS. Our findings indicate that the mutual coupling between the excitable chemical reaction and the underlying phospholipid matrix has pronounced effects on both components under investigation.

’ EXPERIMENTAL SECTION Compartmented phospholipid systems of 1,2-dipalmitoyl-snglycero-3-phosphatidyl choline (DPPC, Avanti Polar Lipids) in 0.34 mol L 1 aqueous KBrO3 (Merck) were prepared by sonication. Experiments were performed at a constant temperature of 22 ( 1 °C using the BZ reaction containing 0.11 mol L 1 malonic acid (Merck), 0.35 mol L 1 H2SO4 (Riedel de Ha€en), 0.34 mol L 1 KBrO3 (Merck), 57 mmol L 1 KBr (SigmaAldrich), and 2.9 mmol L 1 ferroin (prepared from FeSO4 (Merck) and 1,10-phenanthroline (Riedel de Ha€en)), and a suitable amount of DPPC, such that the additions of DPPC contributed from 0 to 25% to the total weight of the sample (i.e., 0 25% w/w). The concentrations are the initial concentrations of the different reaction components in the reactor. To study the dynamic behavior, 1.0 mL of the BZ/DPPC suspension was transferred into a circular Plexiglas dish of 30 mm inner diameter, yielding a ∼0.8 mm high layer of the BZ/DPPC mixture. This dish was placed in a 2D spectrophotometric setup, between a LCD projector (Hitachi CP5860) as light source and a monochrome CCD camera (Hamamatsu C3077). A circular milky glass plate and an optical filter (493 ( 8 nm) were used to ensure homogeneous illumination and image contrast enhancement,

ARTICLE

Figure 1. Spiral wave of the BZ reaction in a DPPC matrix (0.25% w/ w). The trajectory of the spiral tip is superimposed, and this tip moves with a tangential velocity ctan. The area inside the trajectory is never visited by the wave and represents the spiral core of radius r. The distance between two subsequent wave trains is the pitch λ, and the normal velocity of these wave trains is the propagation velocity c of the wave. Size of the image: 7.5  7.0 mm2.

respectively. The video signal from the CCD camera was digitized using an image-acquisition card. To trace the spiral tip precisely, the resolution of the images was adjusted to 0.025 mm pixel 1. Pictures were sampled at 1 Hz yielding 20 75 images per rotation of the spiral (depending on the DPPC content). For the analysis of the trajectory of the spiral tip, the contrast of the images was further enhanced by background subtraction followed by a histogram stretching. The background was calculated as the temporal average of all images. The tip of the spiral wave is defined as the intersection of contour lines (at 0.6  amplitude) of two subsequent images. This method was described in detail by Grill and colleagues19 and has found wide application in studies of spiral tip motion.20 22 Time space plots were constructed such that they passed through the origin of the spiral core and were used to determine the characteristic kinematic features, such as wave velocity c, the pitch λ (or wavelength) of the spiral, and the period T needed for a rotation around its core (Figure 1). The wave velocity was measured during the first hour of each experiment. The wave fronts were allowed to propagate at least 8 mm away of the tip before the velocity was measured so that the effect of curvature of the wavefront is negligible. SAXS diagrams were recorded either on a home-built apparatus (pinhole camera with Cu KR source radiation, collimating optics, and rotating anode) or at the ID02 beamline of the European Synchrotron Radiation Facility (Grenoble, France). The latter SAXS apparatus was chosen for samples having low DPPC content, due to its enhanced sensitivity. Measuring cells were 1 mm thick and in both cases SAXS measurements were started 5 min after sample preparation. The accumulation time was 40 min and 1 s for the standard and synchrotron cameras, respectively. The full width at half-maximum (fwhm) of Bragg peaks was calculated by fitting with a Lorentzian line shape. Direct comparison between the two data sets recorded by different apparatuses was not carried out, since instrumental factors also affect the width of Bragg peaks. 3228

dx.doi.org/10.1021/jp2004375 |J. Phys. Chem. A 2011, 115, 3227–3232

The Journal of Physical Chemistry A

Figure 2. Velocity of the propagating BZ waves as function of the DPPC content.

’ RESULTS In aqueous solution, i.e., in the absence of DPPC, the BZ system forms spiral waves, whose tips move on meandering trajectories. Meandering spirals were also found at low DPPC contents, while for lipid contents g2.5% (w/w) only spiral tips describing rigid rotations around circular core areas are observed (Figure 1). The propagation velocity c of the rigidly rotating spiral waves decreases monotonically with increasing content of DPPC in the reaction medium (Figure 2). This reflects the general increase in viscosity as the concentration of DPPC is increased, which in turn, leads to lower effective diffusion coefficients of the reacting species. In contrast to the wave propagation velocity, other parameters which describe the characteristic properties of propagating spiral waves were found to present a nonmonotonous behavior in BZ/ DPPC medium (Figure 3). More precisely, we can distinguish between one form of behavior for DPPC content of up to ∼12.5% (w/w) and another that occurs at higher lipid content. The spiral performs rigid rotations around a core of ∼0.15 mm radius when the DPPC content in the medium lies between 2.5 and 12.5% (w/w), while the core radius r is dramatically increased for BZ/DPPC media of higher lipid content (i.e., DPPC >12.5% (w/w), Figure 3b). The spiral rotates with a period of rotation T, which also depends on the DPPC content in a nonmonotonous, bimodal manner. We observe two domains of increasing T with increasing lipid content, which are separated by a pronounced minimum at 10 12.5% (w/w) DPPC (Figure 3a). The core radius r and the rotation period T define the tangential velocity ctan = 2πr/T of the spiral tip along its circular trajectory. This velocity has a minimum at 5 7.5% (w/w) DPPC content and a maximum at about 15% (w/w) DPPC. Again, the transition between minimum and maximum occurs around 12.5% (w/w) DPPC (Figure 3d). Finally, the spiral pitch λ (i.e., the characteristic wavelength which separates consecutive wave passages along a line) also changes with the DPPC content: two domains of increasing λ are found at increasing lipid content, which are separated by a pronounced minimum at 10 12.5% (w/w) DPPC (Figure 3), following a trend that is similar to that observed for the period T. The evolution of the spiral core radius r, spiral pitch λ, rotation period T, and tangential velocity ctan as a function of the phospholipid content show two distinct modes, the first of which ranging from 2.5 to 12.5% (w/w) DPPC, while the second mode

ARTICLE

is observed for 12.5 25% (w/w) DPPC. The correspondence of these features as well as the common DPPC value for the transition between the two modes suggests the occurrence of a pronounced transition in the nature of the BZ/DPPC medium at a content of ∼12.5% (w/w) DPPC. SAXS experiments were performed to obtain insights in the supramolecular structuring of DPPC in the BZ reaction medium as a function of the lipid content. All SAXS diagrams recorded in the BZ/DPPC system reflected the classic lyotropic smectic stacking of phospholipids, with alternating DPPC bilayers and aqueous domains. Remarkably, in the presence of a reacting BZ system, the overall structure consists of a unique, well-defined lamellar phase in the entire range of DPPC content investigated. On the other hand, in control systems where the catalyst is absent, microscopic demixing leads to the coexistence of two lamellar phases with slightly different spacing at DPPC >20% (w/w) (Figure 4). No evidence of vesicle formation was found in the SAXS diagrams of the (BZ/DPPC) system, either from data obtained by a laboratory apparatus (DPPC >2.5% (w/w)) or by using a SAXS camera at a synchrotron radiation facility (DPPC e5% (w/w)). As previously observed for some specific lipid content,17 the comparison between SAXS diagrams obtained in the BZ/DPPC system and those of the corresponding nonreactive media showed that Bragg reflections are narrowed in the reacting BZ/ DPPC system (Figure 5). Since the width of Bragg peaks is related to the extension of coherence domains in the lyotropic smectic phase, the observed trend reflects a bimodal variation of the persistence length in the arrangement of DPPC bilayers.23 In fact, while for high phospholipid content (i.e., DPPC g 12.5% (w/w)) the peak width is relatively constant, there is a more marked dependence on DPPC content in the range 0.05% e DPPC e 12.5% (w/w) (Figure 5 and Table 1) with a maximum at ∼5% (w/w). This suggests a different structural organization of the DPPC lamellae in the low DPPC (DPPC e 12.5% (w/w)) and the concentrated DPPC regimes (DPPC g 12.5% (w/w)).

’ DISCUSSION We have investigated the behavior of spatiotemporal patterns of the Belousov Zhabotinsky reaction in aqueous layers of DPPC lamellar phases, focusing on the dependence of the dynamic features of the generated spiral patterns on the content of the phospholipid DPPC in the medium. We also studied the supramolecular organization of the DPPC lamellae as a function of the DPPC content. Thus, we could show for the first time that there is a pronounced and peculiar bimodal dependence of both the characteristics of BZ patterns and the persistence lengths of DPPC lamellae on the DPPC content of the medium. Spatiotemporal patterns in the BZ reaction are affected by the presence of DPPC bilayers in the reaction medium. As expected, the propagation velocity c of the reaction fronts (Figure 2) decreases monotonically with the DPPC content, that is with increasing viscosity of the medium. This leads to a slower effective diffusion of the autocatalytic compound of the BZ reaction and ultimately to a decrease in wave velocity. A similar dependence of the wave propagation velocity of BZ waves on the content of the surfactant SDS has been reported recently.12 However, our present study shows that the behavior of BZ waves in dependence of the DPPC content differs significantly from previous studies using micellar surfactants.12 Indeed, the 3229

dx.doi.org/10.1021/jp2004375 |J. Phys. Chem. A 2011, 115, 3227–3232

The Journal of Physical Chemistry A

ARTICLE

Figure 3. Dependence of the characteristic parameters of a spiral-shaped wave on the content of DPPC in the medium. These parameters show a bimodal behavior in the regions of low (12.5% w/w) contents. While (a) the period of rotation T shows two maxima, one in each region, the (b) radius of the spiral core remains unaffected for low DPPC contents and increases for DPPC >12.5% (w/w). (c) The pitch λ behaves analogously to the rotation period T. (d) The tangential velocity ctan of the spiral tip shows a drastic increase at about DPPC = 12.5% (w/w).

Figure 4. SAXS diagrams of a BZ solution (black line) and of a BZ analogue lacking the catalyst ferroin (gray line). Both systems contain 25% (w/w) DPPC.

effect of phospholipid layers on the period of rotation, pitch, radius of the spiral core, and tangential velocity ctan of the rotating spiral waves cannot be explained solely by an increased viscosity

Figure 5. Full width at half-maximum of first Bragg peak of SAXS diagrams in dependence of the DPPC content.

of the medium. Surprisingly, the changes in these characteristic features of spiral waves are not monotonic as the DPPC content increases. Instead, two distinct regions with different behavior can be distinguished, one below a 12.5% (w/w) DPPC and the 3230

dx.doi.org/10.1021/jp2004375 |J. Phys. Chem. A 2011, 115, 3227–3232

The Journal of Physical Chemistry A

ARTICLE

Table 1. Position and Peak Widths (full widths at half maxima, FWMH) of the first Bragg Peak in SAXS Diagrams Recorded at the ERSF Synchrotron Facility for the BZ/DPPC System at Low DPPC Content

a

DPPC content (% w/w)

d (nm 1)a

FWHM (nm 1)b

0.05

0.965

0.058

0.25 1

0.975 0.983

0.060 0.061

3.75

0.943

0.067

5

0.915

0.078

Error = (0.001. b Error = (0.002.

other for DPPC exceeding 12.5% (w/w). Therefore, we conclude that an important feature of the BZ/DPPC system is fundamentally changed at ∼12.5% (w/w) DPPC content, giving rise to a phase transition or a bifurcation in the dynamic behavior. The radius r of the spiral core is determined by the interplay of the autocatalytic and refractory steps of the BZ reaction.24 At low DPPC content, the radius r is virtually unaffected by the phospholipid concentration, while for DPPC g12.5% (w/w) it increases and shows a maximum (Figure 3b). Since the reaction remains the same, the medium must either slow down the autocatalytic reaction steps or enhance the refractory ones when DPPC g12.5% (w/w). This may happen due to a preferential solubilization of BrO2, which is the intermediate involved in the catalytic reaction step, into the lipid bilayers. This leads to a decreased availability of the autocatalyst in the aqueous domain where the reaction takes place, so that the refractory reaction steps are slightly enhanced. The accumulation of lipophilic intermediates into surfactant phases of compartmented BZ media has already been proposed as a mechanism leading to deceleration of the BZ reaction.7 12 Interestingly, the bimodal behavior of the characteristic features of spiral waves (i.e., its period of rotation, pitch, radius of the spiral core, and tangential velocity ctan) as a function of the phospholipid content correlates well with SAXS measurements (Figure 5). The line widths of the Bragg reflections, which report on the structuring of the DPPC bilayer, also behave in a binomial fashion, showing changing values for DPPC e12.5% (w/w) and remaining almost constant for higher DPPC content. Here, again, the transition between the two types of behavior occurs at a DPPC content of ∼12.5% (w/w). The SAXS data indicate that at DPPC content above ∼12.5%, the lamellar structure is stiffer than at low phospholipid content. This may come either from a solubilization of BrO2 into the bilayer cores, making them thicker and less fluctuating, or from a change of the osmotic pressure in the confined aqueous layer.25 It is worth noting that the present SAXS study goes beyond previous investigations reported in the literature,17,18 in that it describes the behavior of the BZ/DPPC system in a wide range of phospholipid content. At any rate, the original contribution of the on-going BZ reaction is not simply reflected by a thickness variation in the bilayer, as it may be induced by insertion of guest molecules. Rather, an ordering effect is produced at the mesoscopic scale that is in between the molecular size and the characteristic length of the chemical patterns. The changes in the order parameter obtained from SAXS diagrams of the BZ/DPPC system at varying phospholipid content indicate that there is a substantial change in the persistence lengths of the lamellar domain

arrangement, which also takes place at ∼12.5% (w/w) DPPC. This novel piece of evidence is in excellent agreement with the trends observed in the BZ pattern formation. The bimodal changes in the parameters of rotating BZ spiral waves, as well as that of the structural order only occur in the BZ/ DPPC system, but neither in a nonreactive analogue nor in BZ media of higher viscosity (in absence of phospholipids). These findings underline that the BZ reaction is indispensable for generating the observed effects. The correlation between the DPPC dependences of the characteristic features of the BZ spiral dynamics (Figure 3) and the structural organization of DPPC bilayers (Figures 4 and 5) strongly points to a mutual interaction between the reaction and the phospholipid domain arrangement, possibly via solubilization of BrO2 in the apolar region of the lamellae. This is remarkable, since we observe a coupling between the bilayer arrangement that is characterized by persistence lengths of 0.5 1.0 μm and patterns of the excitable BZ reaction, which have a typical pitch λ of 1.0 2.5 mm. Thus, two phenomena that occur at length scales differing by as much as 3 orders of magnitude are coupled to each other. The molecular mechanism leading to this fascinating coupling is still unknown and awaits experimental elucidation. Research along these lines is in progress.

’ CONCLUSIONS Characteristic features of the propagation of chemical waves in the BZ system show a pronounced dependence on the content of the phospholipid DPPC in the reaction medium. Surprisingly, two behavioral domains can be distinguished, one for DPPC content up to ∼12.5% (w/w) DPPC and the other for higher phospholipid concentration. Although the phsopholipid phase remains lamellar at all times, the coherence length in the lamellar arrangement is found to increase considerably for DPPC content g12.5% (w/w). Therefore, we conclude that there is a mutual interaction between the features of the BZ reaction, which takes place on a scale of millimeters with the ordering of the lamellar phase occurring at lengths of 500 nm to 1 μm. ’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by the Deutsche Forschungsgemeinschaft (DFG). G.B. thanks the Deutscher Akademischer Austauschdienst (DAAD) for a stipend for a research stay in Magdeburg. We thank the IDO2 beamline of the ESRF in Grenoble for beam time allocation and Dr. Federico Rossi (University of Siena) for fruitful discussions on the SAXS measurements of BZ reaction media. ’ REFERENCES (1) Bredig, G.; Weinmayer, J. Z. Phys. Chem. 1903, 42, 601–611. (2) Field, R. J.; Burger, M., Oscillations and traveling waves in chemical systems; Wiley: New York, 1985. (3) Kapral, R.; Showalter, K., Chemical waves and patterns; Kluwer: Dordrecht, 1995. (4) Mikhailov, A. S.; Showalter, K. Phys. Rep. 2006, 425, 79–194. 3231

dx.doi.org/10.1021/jp2004375 |J. Phys. Chem. A 2011, 115, 3227–3232

The Journal of Physical Chemistry A

ARTICLE

(5) Luengviriya, C.; Storb, U.; Lindner, G.; M€uller, S. C.; B€ar, M.; Hauser, M. J. B. Phys. Rev. Lett. 2008, 100, 148302. (6) Schenning, A. P. H. J.; Spelberg, J. H. L.; Driessen, M. C. P. F.; Hauser, M. J. B.; Feiters, M. C.; Nolte, R. J. M. J. Am. Chem. Soc. 1995, 117, 12655–12656. (7) Mariato, M.; Nikles, J.; Romsted, L. S.; Tramontin, M. J. Phys. Chem. 1985, 89, 1341–1344. (8) Cavassino, F. P.; Cervelatti, R.; Lombardo, R.; Turco Liveri, M. L. J. Phys. Chem. B 1999, 103, 4285–4291. (9) Paul, A. J. Phys. Chem. B 2005, 109, 9639–9644. (10) Sciascia, L.; Lombardo, R.; Turco Liveri, M. L. J. Phys. Chem. B 2007, 111, 1354–1360. (11) Grancicova, O.; Olexova, A. Z. Phys. Chem. 2009, 223, 1451–1463. (12) Rossi, F.; Lombardo, R.; Sciascia, L.; Sbrizolo, C.; Turco Liveri, M. L. J. Phys Chem. B 2008, 112, 7244–7250. (13) Vanag, V. K.; Epstein, I. R. Phys. Rev. Lett. 2001, 87, 228301. (14) Vanag, V. K.; Epstein, I. R. Science 2001, 294, 835–837. (15) Vanag, V. K.; Epstein, I. R. Chaos 2005, 15, 047510. (16) Kaminaga, A.; Vanag, V. K.; Epstein, I. R. Angew. Chem., Int. Ed. 2006, 45, 3087–3089. (17) Ristori, S.; Rossi, F.; Biosa, G.; Marchettini, N.; Rustici, M.; Tiezzi, E. Chem. Phys. Lett. 2007, 436, 175–187. (18) Magnani, A.; Marchettini, N.; Ristori, S.; Rossi, C.; Rossi, F.; Rustici, M.; Spalla, O.; Tiezzi, E. J. Am. Chem. Soc. 2004, 126, 11406–11407. (19) Grill, S.; Zykov, V. S.; M€uller, S. C. J. Phys. Chem. 1996, 100, 19082–19088. (20) Goldschmidt, D. M.; Zykov, V. S.; M€uller, S. C. Phys. Rev. Lett. 1998, 80, 5220–5223. (21) Kheowan, O.-U.; Zykov, V. S.; M€uller, S. C. Phys. Chem. Chem. Phys. 2001, 3, 4747–4752. (22) Luengviriya, C.; Storb, U.; Hauser, M. J. B.; M€uller, S. C. Phys. Chem. Chem. Phys. 2006, 8, 1425–1429. (23) Vainshtein, B. K. Diffraction of X-rays by chain molecules; Elsevier: New York, 1966; Chapter 5. (24) Zykov, V. S. Biophysics 1980, 25, 329–333. (25) Brotons, G.; Dubois, M.; Belloni, L.; Grillo, I.; Narayanan, T.; Zemb, T. J. Chem. Phys. 2005, 123, 024704.

3232

dx.doi.org/10.1021/jp2004375 |J. Phys. Chem. A 2011, 115, 3227–3232