A Mechanistic Approach on the Self-Organization of the Two

Nov 15, 2007 - The rate constant data from the CD spectra show a small positive temperature coefficient, but the rate constant values from the PL data...
1 downloads 6 Views 725KB Size
Download PDF
13126

Langmuir 2007, 23, 13126-13135

A Mechanistic Approach on the Self-Organization of the Two-Component Thermoreversible Hydrogel of Riboflavin and Melamine Abhijit Saha, Swarup Manna, and Arun K. Nandi* Polymer Science Unit, Indian Association for the CultiVation of Science, JadaVpur, Kolkata-700 032, India ReceiVed August 24, 2007. In Final Form: September 28, 2007 The riboflavin (R) and melamine (M) supramolecular complex in the mole ratio of 3:1 (RM31) produces a thermoreversible gel in aqueous medium. The gelation mechanism has been elucidated from morphological investigations using optical, electron, and atomic force microscopy together with time-dependent circular dichroism (CD) and photoluminescence (PL) spectroscopy. Optical microscopy indicates spherulitic morphology at lower gelation temperature (e25 °C), but at higher temperature fibrillar network morphology develops. Electron and atomic force microscopy indicate the presence of left handed helical structures in the fibrils. Kinetic study of gel formation using circular dichroism (CD) and photoluminescence (PL) spectra indicates that there are three steps: (1) RM complex formation, (2) conformational ordering, and (3) π-π-stacking of ordered conformers. The first step of RM complex formation is already established from Fourier transform infrared (FTIR) spectroscopy (Manna, S.; Saha, A.; Nandi, A. K. Chem. Commun. 2006, 4285), and the second step is detected from the CD spectra. Here, the ellipticity value of the n-π* transition increases by 600 times during gel formation. The dramatic increase of ellipticity is attributed to conformational ordering of the ribityl chain followed by helical fibril formation. The third step is concluded from fluorescence spectroscopy, which also shows a 30 times increase in intensity. The substantial increase in PL intensity is caused by hydrophobic core formation during π-stacking of the complex. Both the ellipticity and PL intensity show a sigmoidal increase with time, and analysis of data using the Avrami equation shows n values close to 1.5 for the former and close to 2 for the latter. The rate constant values obtained from the intercepts of Avrami plots are different in the two methods. The rate constant data from the CD spectra show a small positive temperature coefficient, but the rate constant values from the PL data show a negative temperature coefficient except the data at 30 and 35 °C. Arrhenius treatment of the rate constant values of the CD data indicates an activation energy of ∼13 kcal/mol, signifying that the conformational transition is the cause of ellipticity increase. The negative temperature coefficient of the rate constant obtained from the fluorescence data has been attributed to the spherulitic crystal formation, and the increase of the rate constant at 30 and 35 °C has been attributed to fibril formation. The fluorescence intensity and peak position change with temperature and with the concentration of the RM complex in the gel. A probable explanation from fibrillar thickness is offered.

Introduction Small molecular hydrogelators and organogelators have received great attention during the past few years.1-8 The formation of a nanostructured self-assembly traps a significantly large amount of solvent due to its large solid-liquid interfacial area. Until now, a major portion of the work has been devoted to organogels,1,3-5,8 and small molecular hydrogelators are really very uncommon.2,6 For the potential use of hydrogels in biomedical applications, the search of new small molecular hydrogelators has now been extended to two-component systems producing hydrogels through supramolecular complexation of the components.5 The two-component hydrogels may have some benefit over the one-component small molecular hydrogels to exploit the labile supramolecular bonding for the end use of the gel. * To whom correspondence should be addressed. E-mail: psuakn@ mahendra.iacs.res.in. (1) Terech, P.; Weiss, R. G. Chem. ReV. 1997, 97, 3133. (2) de Loos, M.; Feringe, B. L.; van Esch, J. H. Eur. J. Org. Chem. 2005, 3615. (3) Ishi-I, T.; Shinkai, S. Top. Curr. Chem. 2005, 258, 119. (4) Sangeetha, N. M.; Maitra, U. Chem. Soc. ReV. 2005, 34, 821. (5) Hirst, A. R.; Smith, D. K. Chem.sEur. J. 2005, 11, 5496. (6) Estroff, L. A.; Hamilton, A. D. Chem. ReV. 2004, 104, 1201. (7) Molecular Gels: Materials with Self Assembled Fibrillar Network; Weiss, R. G., Terech, P., Eds.; Spinger: Dordrecht, 2006. (8) (a) Ajayaghosh, A.; Praveen, V. K. Acc. Chem. Res. 2007, 40, 644. (b) Schmidt, R.; Schmutz, M.; Mathis, A.; Decher, G.; Rawiso, M.; Me´sini, P. J. Langmuir 2002, 18, 7167.

Recently, we reported the thermoreversible hydrogel formation of riboflavin (R) and melamine (M), which produce a supramolecular complex in the mole ratio of R/M ) 3:1 (RM31, Scheme 1).9 This complex produces a fibrillar network structure in aqueous medium via π-stacking of the complex. In this article, we shed light on the mechanism of the molecular stacking of the RM31 complex producing the gel by exploiting the asymmetric structure of the ribityl chain and fluorescence property of riboflavin. During gelation, a conformational change of the ribityl chain of the RM31 complex is possibly needed to produce an ordered structure required for the π-stacking process.8-10 So, to understand the mechanism of the stacking of the RM31 complex in the gelation process, circular dichroism (CD) may be a useful tool to follow conformational transition, if any, in the gelation process. A solvent has a marked effect on the CD spectra of riboflavin, indicating strong interactions of the isoalloxazine ring and ribityl side chain.11 Such an interaction would also affect the RM complex during gelation. Apart from the molecular asymmetry, molecular dissymmetry may also arise during the π-stacking of the RM complex due to the development of nonplanarity or helicity during gel formation. This new supramolecular dissymmetry may have an additional effect on the CD spectra. Thus, the CD spectra may (9) Manna, S.; Saha, A.; Nandi, A. K. Chem. Commun. 2006, 4285. (10) Takeuchi, M.; Tanaka, S.; Shinkai, S. Chem. Commun. 2005, 5539. (11) (a) Tollin, G. Biochemistry 1968, 7, 1720. (b) Buckingham, A. D.; Stephens, P. J. Annu. ReV. Phys. Chem. 1966, 17, 399.

10.1021/la7026219 CCC: $37.00 © 2007 American Chemical Society Published on Web 11/15/2007

Gelation Mechanism of a RM31 Hydrogel

Langmuir, Vol. 23, No. 26, 2007 13127

Scheme 1. Supramolecular Complex Formation of Riboflavin and Melamine through H-Bonding9

be an important tool to investigate the supramolecular organization of the RM31 complex to produce the thermoreversible gel. Riboflavin has a good photoluminescence (PL) property, and its quantum yield also depends on the polarity of the solvent.12-14 It decreases in hydrogen bond forming solvents and also in the presence of electrolytes, proteins, metal ions, and organic compounds such as phenols, purines, pyrimidine, and thiols.12 In the RM31 supramolecular complex, the H-bonding sites of the isoalloxazine moieties are completely used by the melamine molecules. As a result, the fluorescence intensity of the RM complex shows a sharp increase.9 During aggregation of this complex through π-π interactions, hydrophobic cores are produced, giving the complex very little opportunity to interact with water for H-bond formation. As aggregation of the RM complex increases, fluorescence is expected to increase with time. So, to get a detailed mechanism of aggregation of the RM complex for producing a gel, a kinetic investigation using fluorescence spectroscopy would be very helpful. Here, we report time-dependent CD and time-dependent fluorescence spectral studies in water at different temperatures during gelation. Finally, the micro and macro morphologies of the gels are obtained via atomic force microscopy (AFM), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and polarized microscopy, as a wealth of morphology is reported in the small molecular organogels.1-8 It is interesting to study the morphology in the two-component hydrogelator to understand if there is any difference in morphology from that of the singlecomponent gelator. A correlation of the time-dependent CD and time-dependent fluorescence spectral data with the morphological study would be fruitful to understand the gelation process more effectively. This is because the morphology is a reflection of internal molecular arrangement, which also influences the CD and PL spectra in the system. So, in this article, the gelation mechanism of the two-component RM complex would be elucidated with the help of CD, fluorescence spectra, and morphological investigation. Though there are two recent reports on the elucidation of the gelation mechanism of small molecular organogels,15,16 to our knowledge there is no such mechanistic study on the small molecular hydrogels. Also, with regard to the (12) Penzer, G. R.; Radda, G. K. Q. ReV. Chem. Soc. 1967, 21, 43. (13) Thimann, K. V.; Curry, G. M. In ComparatiVe Biochemistry; Florkin, M., Mason, H. S., Eds.; Academic Press: New York, 1960; Vol. 1, pp 281. (14) Massey, V. Biochem. Soc. Trans. 2000, 28, 283. (15) Huang, X.; Terech, P.; Raghavan, S. R.; Weiss, R. G. J. Am. Chem. Soc. 2005, 127, 4336. (16) Huang, X.; Raghavan, S. R.; Terech, P.; Weiss, R. G. J. Am. Chem. Soc. 2006, 128, 15341.

gelation mechanism of the two-component system, such mechanistic study is yet unavailable. So, it is an important report attempting to elucidate the gelation mechanism of a twocomponent hydrogel of melamine and riboflavin forming a supramolecular complex. Experimental Section A. Materials and Methods. (-)-Riboflavin (R) and melamine (M) were purchased from Aldrich Chemical Co. A mixture of riboflavin and melamine in a 3:1 mole ratio was taken in a glass tube, and water was added to make the total complex concentration 0.2% (w/v). The mixture was then sealed and heated to 120 °C to solubilize the components homogeneously and quenched to room temperature to yield a yellow colored gel. The gel was then freezedried to obtain the dry RM31 complex. All the subsequent studies of the RM31 hydrogel were performed using the method mentioned above by preparing the gel in a sealed container and by taking a required amount of the above freeze-dried RM31 complex and water in definite weight ratios. B. Microscopy. (i) Field Emission Scanning Electron Microscopy (FESEM). To understand the network morphology of the gel, small portions of hydrogels of the RM31 complex of different concentrations, produced at 30 °C, were placed on glass cover slips. They were dried in air at room temperature and finally in vacuum, and then they were observed through a FESEM instrument (JEOL, JSM 6700F) operating at 5 kV. (ii) Transmission Electron Microscopy (TEM). TEM study of the 0.055% RM31 gel sample was done by taking a small portion of the gel on a carbon coated copper grid (200 mesh). It was dried in air at room temperature, and then it was observed through a TEM instrument (JEOL, model 2010EX) directly under a voltage of 200 kV. The TEM picture was taken with the help of a charge-coupled device (CCD) camera attached to the instrument. (iii) Optical Polarization and Fluorescence Microscopy. Optical microscopy of different samples prepared by aging at different isothermal temperatures was performed by using a Leitz Biomed polarized optical microscope fitted with a wild MPS-12 semiautomatic camera system. The sample was observed through a 40× eyepiece with a perfectly crossed polarizer, and the picture was taken with a total of 400× magnification. Fluorescence micrographs of the RM31 (0.055% w/v) gel were taken in a fluorescence microscope (Olympus, BX61) by exciting the gel sample with UV radiation using a FITC filter. (iV) Atomic Force Microscopy. The morphology of the dried RM31 gels was studied using atomic force microscopy (Veeco, model AP0100). The AFM study was conducted in noncontact mode at a resonance frequency of the tip end of ∼300 kHz. The sample was cast on a glass slide. Similar objects formed by the primary focusing step were further scanned, and the pictures were taken in amplitude mode.

13128 Langmuir, Vol. 23, No. 26, 2007

Saha et al.

Figure 1. Optical micrographs (polarized) of RM31 0.055% (w/v) hydrogels formed upon gelling the sols at (a) 15, (b) 20, (c) 25, (d) 30, and (e) 35 °C. C. Spectroscopy. (i) UV-Vis Spectroscopy. UV-vis spectra of the hydrogel sample were taken by using a Hewlett-Packard UVvis spectrophotometer (model 8453). (ii) Photoluminescence Spectroscopy. Fluorescence studies of RM31 hydrogel samples prepared in a sealed cuvette were conducted by using a Horiba Jobin Yvon Fluoromax 3 instrument. The gel samples were taken in a quartz cell of 1 cm path length and were excited at 365 nm. The emission scans were taken from 400 to 800 nm using a slit width of 2 nm with an increment of 1 nm wavelength having an integration time of 0.5 s. Fluorescence lifetimes were measured by using a time-correlated single photon counting fluorometer (Fluorecule, Horiba Jobin Yvon). The system was excited with a 375 nm NanoLED from Horiba Jobin Yvon having λmax at 368 nm with a pulse duration of CdO and >NH groups of the isoalloxazine ring and melamine) of the RM31 supramolecular complex. Thus, it reduces the H-bonding ability with water enhancing the photoluminescence intensity.9

Figure 9. (a) Fluorescence spectra with time for RM31 0.055% (w/v) hydrogel formation at 25 °C, (b) plot of the PL peak intensity with time at indicated isothermal temperatures for RM31 0.055% (w/v) hydrogels, and (c) plot of the PL peak intensity with time at 25 °C for RM31 hydrogels at different concentrations (% w/v).

Also, the π-π-stacking of the RM31 complex creates a more hydrophobic core in the supramolecule. Thus, there is a tendency of an increase of the photoluminescence intensity with time during gelation, and Figure 9a corroborates the same. Here, about an ∼30 times increase in the photoluminescence intensity is found after gel formation, and a similar increase is observed for the same sample during gelation at other temperatures (Figure 3 in the Supporting Information). In Figure 9b, the photoluminescence intensity at the peak position (547 nm) is plotted with time at different temperatures. It is interesting that all the plots exhibit a sigmoidal rise indicating that the increase of photoluminescence is cooperative in nature. In other words, the π-stacking process of the RM31 supramolecular complex is an autocatalytic process during gelation. From Figure 9b, it is apparent that the height of the photoluminescence increase is not the same at all temperatures. For 15 and 20 °C, it is almost same, but at 25 °C the increase is the highest. Again, at 30 and 35 °C, it has decreased to some extent. One possible explanation that may be applicable here is that at lower temperatures (e.g., 15 and 20 °C) the

13132 Langmuir, Vol. 23, No. 26, 2007

spherulites are more compact than those at 25 °C, creating more paths for the nonradiative decay of excitons, causing a greater relative decrease in the photoluminescence intensity than that of the coarser spherulite at 25 °C (Figure 1). The cause of the decrease in the PL intensity for the samples at 30 and 35 °C than that at 25 °C is not clear. One possible way to explain the results is that the equilibrium of the π-stacking process decreases with an increase in temperature, causing progressively lower fractions of hydrophobic cores, and 25 °C may be the optimum temperature of a lesser amount of nonradiative decay and maximum π-πstacking. From Figure 9b, it is noteworthy that the onset time of the sigmoidal increase increases with an increase in temperature as in Figure 7. It is also due to the fact that with an increase in temperature molecular motion increases, causing it to take longer time to form the supramolecular complex and hence π-stacking. In Figure 9c, the increase of fluorescence intensity with time is shown for different concentrations of RM31 gels at 25 °C. It is apparent from the figure that all the samples exhibit sigmoidal plots indicating the cooperative nature for all the concentrations studied here. The leveling intensity is highest (i.e., highest increase) for the 0.055% RM31 gel, and it then gradually decreases with an increase in RM31 concentration. With an increase in RM31 concentration, the fibrils/fibers are fatter (cf. Figure 2), causing additional paths of exciton decay, and as a result the leveling value of fluorescence intensity decreases. In Figure 10a, the temperature dependency of the fluorescence spectra of the 0.16% RM31 complex gel is shown. It is apparent from the figure that with an increase in temperature the fluorescent intensity at first increases and then decreases to almost zero at 96 °C. This trend is also true for the 0.13% and 0.20% RM31 gels, but there is some difference in the cases of the 0.055% and 0.066% gels (Figure 10b). In the figure, the fluorescence peak intensity values are plotted with temperature for five indicated concentrations of the RM31 complex gel. It is apparent from the figure that with an increase in temperature the intensity decreases slowly initially and then sharply for the 0.055% gel. For the 0.066% gel, initially it remains almost constant and then begins a sharp fall. At a lower temperatures, these systems have higher photoluminescence intensities because of the lower fibrillar diameters causing a lesser amount of decay paths of excitons. With an increase in temperature, the thinner fibrils start melting, resulting in a decrease of the hydrophobic core, and at a temperature near the melting point all the fibrils start rapidly melting, reducing the fluorescence intensity abruptly. In the cases of higher gel concentrations (e.g., 0.13, 0.16, and 0.2%), the intensity first increases with an increase in temperature to a maximum value at ∼90 °C and then the fluorescence intensity suddenly falls down as melting starts producing hydrophilic pockets. The gradual increase of the fluorescence intensity may be explained by the thinning of the thicker fibrils by the separation of fibrils from the fiber. This thinning process decreases the number of decaying paths for the nonradiative decay of excitons, and the fluorescence intensity reaches a maximum before the melting of the majority of the fibrils starts. This result suggests that the fibers are produced by the agglomeration of fibrils. At higher temperatures, the cohesive force of attraction between the fibrils becomes insufficient to keep them together. In Figure 10c, the peak wavelength λmax is plotted with temperature for different concentrations of the RM31 gel. Here, the λmax at first remains constant and then it shows a gradual blue shift followed by a rapid blue shift with an increase in temperature. Possibly, with increasing temperature, the fibers become progressively thinner, and as a result the delocalization of exciton energy levels

Saha et al.

Figure 10. (a) Temperature-dependent PL spectra of RM31 0.16% (w/v) hydrogels, (b) plot of the fluorescence peak intensity vs temperature of RM31 hydrogels at indicated concentrations (% w/v), (c) plot of the peak position (λmax) vs temperature of RM31 hydrogels at indicated RM31 complex concentrations (% w/v), and (d) plot of the melting point (obtained from (b) and (c) vs concentration (% w/v) of RM31 hydrogels.

is decreased, causing a blue shift of λmax. The onset temperature for the melting process may be calculated from the intercept of the tangents of two portions of the curves, and they are plotted

Gelation Mechanism of a RM31 Hydrogel

Langmuir, Vol. 23, No. 26, 2007 13133 Table 1. Lifetime Data for RM31 Gel Samples at Different Concentrations (% w/v) lifetime from calculation sample

first order (ns)

second order (ns)

pure riboflavin 0.055 0.07 0.1

4.7 4.7 4.7 4.6

4.7 4.6 4.7 4.7

the supramolecular complex formation and its aggregation to produce the fibrils. Therefore, it may be inferred that the fluorescence property of the gel is arising from riboflavin and no dynamic quenching process is involved in the gel state.22 The lifetime measurements conclude that enhancement of the photoluminescence efficiency is only for hydrophobic core formation during supramolecular organization producing the fibrillar structure.

Discussion

Figure 11. Plots of the PL (a) peak position vs concentration (% w/v) and (b) intensity at the peak position vs concentration (% w/v) of RM31 hydrogels at 25 °C.

in Figure 10d. It indicates that the gel melting temperature increases linearly with an increase in RM31 complex concentration. The onset temperature obtained from the intersection of the tangents of the intensity data in Figure 10b are also presented in Figure 10d. It is apparent from Figure 10d that the gel melting data obtained from both of the methods are almost same, signifying that both the intensity and λmax originate from the same process, namely, hydrophobic core formation during gelation. In Figure 11a, the wavelength of maximum emission (λmax) is plotted with different concentrations of the RM31 complex. It is evident from the figure that λmax increases linearly with an increase in RM concentration in the gel. The red shift of λmax with increasing concentration is interesting, and it may be attributed to the formation of more fatter fibers in concentrated solution than in dilute solution. The fatter fiber helps to delocalize the exciton, reducing the energy gap during nonradiative decay and showing a red shift with increasing concentration. Figure 11b presents the plot of the emission intensity at the peak position of the photoluminescence spectra versus RM31 complex concentration, and it exhibits a maximum at 0.055% (w/v) of the complex. The initial sharp increase of the intensity may be due to the increase of the complex concentration and hence the increase of the number of fibrils. After 0.055%, the fibrils start bundling into fatter fibers and this bundling creates more decay paths for the excitons. As a result, the photoluminescence intensity gradually decreases with an increase in the concentration of the RM complex. In Table 1, the fluorescence lifetime data are presented for both first-order and second-order calculations. On analyzing the decay curve (Figure 4 in the Supporting Information) with the single-exponential fit (χ2 ) 1.1, DW ) 1.81), a lifetime of 4.6 ns is observed. It is apparent from the table that the lifetime value of pure riboflavin at the same concentration as in the 0.055% gel is almost same. With an increase in RM31 complex concentration, the lifetime of the excitons of riboflavin remains same even after

Avrami Treatment. In the sol to gel transformation, spherulites/fibrils (a heterogeneous phase) grow from a homogeneous liquid (sol) and the growth of a new phase from its mother phase is usually explained by the Avrami theory of crystallization.23 The theory has been successfully applied to understand the nucleation and growth mechanism of polymer crystals,24-26 and it is now also used to determine the nucleation and growth mechanism of nanoparticles27,28 and also organogels.15 Encouraged by the successful application of the Avrami theory to a single-component organogel,15,16 we have applied the theory to the two-component hydrogel of riboflavin and melamine. The Avrami equation is usually expressed as23

1 - V(t) ) exp(-ktn)

(1)

where V(t) is the fraction transformed, k is the rate constant, t is the time of transformation, and n is a constant whose value depends on the nature of the nucleation and growth process. The fraction transformed can be obtained from the ellipticity data of the n-π* transition in Figure 7 or from the emission intensity data in Figure 9b. If the ellipticity/intensity at t ) 0 is I0 and at t ) ∞ is I∞, then V(t) is equal to (It - I0)/(I∞ - I0). Putting these values into eq 1, we obtain

I∞ - It ) exp(-ktn) I∞ - I o Taking logarithms on both sides, the equation transforms to

(

ln -ln

)

I∞ - It ) ln k + n ln t I∞ - I o

(2)

Thus, by plotting the left-hand side of eq 2 with ln t, straight lines are expected, and from the slope the Avrami exponent n can be obtained. The rate constant k of the process can be obtained from the intercept of the plot. (22) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 2nd ed.; Kluwer Academic/Plenum Publisher: New York, 1999. (23) (a) Avrami, M. J. Chem. Phys. 1939, 7, 1103. (b) Avrami, M. J. Chem. Phys. 1940, 8, 212. (24) (a) Mandelkern, L. Crystallization of Polymers; McGraw-Hill: New York, 1964. (b) Wunderlich, B. Macromolecular Physics; Academic Press: New York, 1976. (25) Cheng, S. Z. D.; Wunderlich, B. Macromolecules 1988, 21, 3327. (26) Pal, S.; Nandi, A. K. Polymer 2005, 46, 8321. (27) Manna, S.; Batabyal, S. K.; Nandi, A. K. J. Phys. Chem. B 2006, 110, 12318. (28) Dawn, A.; Mukherjee, P.; Nandi, A. K. Langmuir 2007, 23, 5231.

13134 Langmuir, Vol. 23, No. 26, 2007

Saha et al.

Figure 13. Avrami plot of the PL intensity data of Figure 9b for RM31 0.055% (w/v) hydrogels at different isothermal temperatures.

Figure 12. (a) Avrami plot of the CD data of Figure 7 for RM31 0.055% (w/v) hydrogels at different isothermal temperatures and (b) Arrhenius plot of ln k vs 1/T obtained from (a).

(i) Circular Dichroism Data. In the application of the theory to both the CD and PL data, we have to consider the incubation time (t ) 0) because it is the time to produce the RM supramolecular complex from the components. As is apparent from Figure 7, the incubation time increases with an increase in temperature for a particular concentration of the complex. In the figure, the onset of a sharp rise of the ellipticity or photoluminescence intensity has been considered as the incubation time, and it is considered as the zero time (t0) for the transformation. Thereafter, all the times subtracted from the incubation times are considered as the real time for the Avrami calculations. In Figure 12a, Avrami plots obtained from the circular dichroism data are presented and certainly the data fit well in straight lines with slope (n) values of 1.51, 1.52, 1.68, and 1.31 for 15, 20, 25, and 30 °C, respectively. The rate constant values obtained from the least-square intercepts are 1.25 × 10-2, 2.28 × 10-2, 3.51 × 10-2, and 3.66 × 10-2 min-1 for 15, 20, 25, and 30 °C, respectively. The n values are close to 1.5, and this is better interpreted as heterogeneous nucleation with one-dimensional growth of the supramolecules.15,24 In Figure 12b, the rate constant values are plotted according to the Arrhenius equation and the data fit very well in a straight line. From the slope, the activation energy is calculated to be 12.8 kcal/mol. This activation energy is probably for the conformational transition of the supramolecular RM31 complex, because this value is close to the activation energy values of the conformational transition of other polymers. For example, in poly(3-alkyl thiophene), the activation energy of the conformational transition is 15 kcal/mol (theoretical)29 and its experimental value from gelation kinetics measurements is 23.7 kcal/ mol.30 For poly(vinylidene fluoride), the theoretical value is 9 (29) Shibaev, P. V.; Schaumburg, K.; Bjornholm, T.; Norgaard, K. Synth. Met. 1998, 97, 97. (30) Malik, S.; Jana, T.; Nandi, A. K. Macromolecules 2001, 34, 275.

kcal/mol31 and the experimental value is 15 kcal/mol.32 In the DNA-poly(o-methoxy aniline) hybrid, the activation energy of the conformational transition of poly(o-methoxy aniline) is 14.5 kcal/mol.33,34 Thus, the present analysis confirms that, in the gelation of the RM31 supramolecular polymer, the conformational transition of the supramolecule occurs before the beginning of π-stacking process to produce the fibrillar network. (ii) Photoluminescence Data. The increase in photoluminescence intensity with time is due to the formation of the hydrophobic core, which decreases the H-bonding interaction of the complex with water.9 So, analysis of the intensity increase would give an idea on how the hydrophobic core surface is produced. In Figure 13, the Avrami plots are presented for the PL data applying the same principle as that applied for the CD spectral data. Here, the data also fit in straight lines very well. The slope values (n) are 1.81, 2.12, 2.09, 1.64, and 1.41 for 15, 20, 25, 30, and 35 °C, respectively. It is interesting that the exponent values are close to 2 for the first three temperatures, and this may be interpreted as two-dimensional thermal (heterogeneous) nucleation with diffusion controlled growth.23,24 At 30 and 35 °C, the Avrami exponent values indicate thermal nucleation with one-dimensional growth.24 These results are in accordance with the morphology changes from a spherulite to fibrillar network. The different values of the Avrami coefficient from those of the CD data are interesting and certainly indicate that two different processes are involved in the gelation of the RM31 complex. This is also supported by the rate constant values obtained from the intercepts of Figure 13. They are 3.01 × 10-2, 2.32 × 10-3, 2.81 × 10-4, 1.75 × 10-3, and 1.8 × 10-3 min-1 for 15, 20, 25, 30, and 35 °C, respectively, and they are quite different from those obtained from the CD spectroscopy data. Further, the k values obtained from the CD spectroscopy data have a small positive temperature coefficient; on the other hand, the k values obtained from the photoluminescence data indicate a strong negative temperature coefficient except at 30° and 35 °C (Figure 14). These results indicate that the two processes are different from each other. Presumably, the conformational change together with the helical fibril formation is associated with the increase in intensity of the n-π* transition of the CD spectra. And the photoluminescence intensity increase is associated with the formation of the hydrophobic core surface, that is, aggregation of supramolecules to fibrils. Thus, the two experimental techniques indicate that there are two different processes occurring during the gelation of the RM31 complex, for example, (31) Farmer, B. L.; Hopfinger, A. J.; Lando, J. B. J. Appl. Phys. 1972, 43, 4293. (32) Dikshit, A. K.; Nandi, A. K. Macromolecules 1998, 31, 8886. (33) Dawn, A.; Nandi, A. K. Macromolecules 2005, 38, 10067. (34) Dawn, A.; Nandi, A. K. Langmuir 2006, 22, 3273.

Gelation Mechanism of a RM31 Hydrogel

Langmuir, Vol. 23, No. 26, 2007 13135

during gelation, and it has been attributed to the conformational transition of the ribityl chain followed by the formation of helical fibrils. Also, an ∼30 times increase in the photoluminescence intensity is observed during gel formation of the above complex, and this has been attributed to the hydrophobic core formation during the π-stacking of the RM31 complex. Thus, the gelation in this system is considered to constitute of three steps:

Figure 14. Plot of ln k (obtained from Figures 12a & 13) vs gelation temperature.

conformational change and aggregation of the RM31 supramolecular complex. Temperature coefficient analysis of the rate constants of the crystals is really difficult, as different theories are involved in the spherulitic and fibrillar crystallization in solution.24a,32,35,36 Nonetheless, it may be concluded from the rate constant temperature plot that with an increase in temperature spherulite crystallization is hindered due to lower undercooling but when the fibrillar crystallization starts there is a sudden hike in the rate constant. This indicates that fibrillar crystallization in this system is easier than spherulite crystallization at temperatures greater than 25 °C. No definite explanation can be afforded for this unique behavior, and it may be postulated that the spherulite formation of the supramolecular process might occur by chain folding, which requires more energy than the fibril formation, as generally the end surface energy has a greater value than the lateral surface energy.37

Conclusion The gelation mechanism of a two-component hydrogel of riboflavin and melamine at a 3:1 composition has been elucidated from circular dichroism, photoluminescence spectroscopy, and morphology investigation of a 0.055% (w/v) composition of the complex. The polarized microscopy experiments indicate spherulite formation in the gel at lower temperatures (e25 °C) and above that temperature fibrils are produced, generating the network structure. SEM, TEM, and AFM studies showed the presence of twisted (helical) fibrils in the system; this helical structure is supposed to originate from the π-stacking of the nonplanar RM31 supramolecular complex. The n-π* transition peak of CD spectra shows a 600 times rise of the ellipticity (35) Mal, S.; Maiti, P.; Nandi, A. K. Macromolecules 1995, 28, 2371. (36) Boon, J.; Azcue, J. M. J. Polymer Sci., Part A-2 1968, 6, 885. (37) Treatise on Solid State Chemistry; Hoffman, J. D., Davis, T., Lauritzen, J. I. J., Eds.; Plenum Publisher: New York, 1976; Vol. 3.

Step 1 occurs via H-bonding as evidenced from the FTIR study reported earlier, step 2 occurs via conformational ordering (activation energy 12.8 kcal/mol), and step 3 occurs via the π-stacking process. Step 2 is clearly manifested from the CD spectra, and step 3 is manifested from the photoluminescence spectra. Both the n-π* ellipticity of the CD spectra and the photoluminescence intensity exhibit a sigmoidal increase with time, indicating the cooperative nature of the above processes. Analysis of the data using the Avrami equation yields Avrami exponent (n) values close to 1.5 for the CD data and close to 2 for the photoluminescence data. The former n value is interpreted as heterogeneous nucleation with one-dimensional growth, while the latter is interpreted as two-dimensional heterogeneous nucleation with diffusion controlled growth. Also, the rate constant values obtained from the CD data indicate a small positive temperature coefficient, but those of the PL data show a negative temperature coefficient except at 30 and 35 °C. Thus, the CD and PL data represent two different processes, namely, conformational ordering and aggregation processes, for the twocomponent supramolecular gel. The fluorescence intensity and peak position change differently with temperature and with concentration. The gel melting points obtained from both the PL intensity and λmax plot with temperature are almost equal and increase with the RM complex concentration in the gel. Acknowledgment. We gratefully acknowledge the Department of Science and Technology, New Delhi (Grant No. SR/ S1/PC-32/2004) for financial support. A.S. and S.M. acknowledge CSIR, New Delhi for providing the fellowship. We acknowledge Prof. T. Ganguly of the Spectroscopy Department for discussion on fluorescence lifetime data. Supporting Information Available: AFM image, timedependent isothermal CD spectra, PL spectra at different temperatures, and time-resolved fluorescence decay of the RM31 0.055% (w/v) hydrogel. This material is available free of charge via the Internet at http://pubs.acs.org. LA7026219