J. Phys. Chem. B 2008, 112, 16485–16489
16485
Effect of Geometrical Confinement on the Nucleation and Crystallization Behavior of n-Alkane Mixtures Kai Jiang,†,‡ Yunlan Su,† Baoquan Xie,†,‡ Shichun Jiang,§ Ying Zhao,† and Dujin Wang*,† Beijing National Laboratory for Molecular Sciences, Key Laboratory of Engineering Plastics, Joint Laboratory for Polymer Science and Materials, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, P. R. China, State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, P. R. China, and Graduate School of Chinese Academy of Sciences, Beijing 100190, P. R. China ReceiVed: August 17, 2008; ReVised Manuscript ReceiVed: October 17, 2008
The condensed structure of normal alkane (n-alkane) mixtures in confined geometry is an interesting topic concerning the difference in crystallization behavior of odd and even alkanes. In the present work, the crystallization of mixtures of normal octadecane (n-C18H38) and normal nonadecane (n-C19H40) in microcapsules with narrow size distribution was investigated using the combination of differential scanning calorimetry (DSC) and X-ray diffraction (XRD). A surface freezing monolayer for microencapsulated n-C18H38, n-C19H40, and their mixture was detected by DSC, which for the mixture is a mixed homogeneous crystalline phase with continuous change in the composition. A more stable rotator phase (RI) was observed for the microencapsulated n-C18H38/n-C19H40 ) 95/5 (molar ratio) mixture, confirmed by an increased supercooling of the transition from RI to stable phase compared to that of the mixture in bulk. Two nucleation mechanisms were speculated as “liquid-to-solid” heterogeneous nucleation and “solid-to-solid” homogeneous nucleation, which occur at different crystallization stages in microcapsules and might be attributed to the surface effect and confinement effect, respectively, in the confined geometry. Introduction Normal alkanes (n-alkanes), having a linear chain molecular structure, are among the most fundamental building blocks of organic, biological, and polymeric systems.1 The pure n-alkanes have some peculiar features in phase equilibriums such as odd-even effect and surface freezing effect.2-4 The mixtures of alkanes are prevalent in practical applications and useful model systems to understand the physical behavior of various polydispersed linear chain aggregates such as linear polymers and biological lipids. Therefore, these mixtures have been extensively investigated using both experimental methodologies and theoretical simulations.5-19 Generally speaking, mixing different chain lengths of alkanes leads to a dramatic influence on the phase behavior, in particular, when even alkanes are mixed with odd alkanes. The chain length mixing reduces the interaction of interlayer coupling, resulting in the stability of the rotator phase, and correspondingly, the odd-even effect disappears in alkane mixtures.13 The alkanes exhibit the wellknown odd-even effect related with their melting points and crystal structures. The melting temperatures of n-alkanes and their derivatives do not show a monotonic increase with increasing chain length. Instead the melting points of odd alkanes are relatively lower than those of even alkanes with an additional carbon number. Another typical odd-even effect is that the supercooling of the rotator-to-crystal transition in even alkanes is larger than that in odd alkanes. However, the structures of the rotator phases and the surface freezing * Corresponding author. Phone: +86-10-82618533. Fax: +86-1082612857. E-mail:
[email protected]. † Institute of Chemistry, Chinese Academy of Sciences. ‡ Graduate School of Chinese Academy of Sciences. § Changchun Institute of Applied Chemistry, Chinese Academy of Sciences.
monolayer in alkane mixtures are found to be very similar to those of single component alkanes.20-24 All of these findings enable researchers to gain insight into the crystallization behavior of n-alkane mixtures, but to date, there are few reports on a comprehensive understanding of the phase change behavior and the condensed states of n-alkane mixtures at the molecular level. It has been known that the surface freezing of n-alkanes in a confined space becomes more prominent than that in the bulk, which can be detected by simple measurements because of the higher surface-to-volume ratio.25,26 More recently, novel rotator phases, absent in the bulk, were observed for medium length n-alkanes confined in emulsified microdroplets,27-29 microcapsules,25,26 and mesoporous silica matrix or mesopores.30-32 These results indicate more intimate connections between the surface freezing and bulk phase behavior in confined geometry. Therefore, one can expect more fascinating information on the nucleation and crystallization behavior of n-alkane mixtures by confining them into finite-size space. In our previous reports, we have studied the crystallization behaviors of n-C18H38 (abbreviated as C18) and n-C19H40 (abbreviated as C19) in microcapsules using melamine-formaldehyde resin as a shell by in situ polymerization.25,26 Direct DSC evidence for surface freezing monolayer and novel rotator phases absent in the bulk were found for confined alkanes in microcapsules. In the present work, we extended our interests from pure n-alkane components to the mixtures of odd and even n-alkanes. We prepared a series of microcapsules containing the n-C18H38/n-C19H40 mixtures (abbreviated as m-C18/C19). The crystallization behavior of the n-alkane mixtures in the microcapsules has been investigated as the function of crystallization temperature and C19 concentration. On the basis of the experi-
10.1021/jp807347d CCC: $40.75 2008 American Chemical Society Published on Web 12/02/2008
16486 J. Phys. Chem. B, Vol. 112, No. 51, 2008
Jiang et al.
mental observations, we proposed the phase transition mechanism of n-alkane mixtures in the confined geometry. Experimental Methods n-C19H40 (purity >99%) and n-C18H38 (purity >99%) were purchased from Aldrich Company and Acros Company, respectively. All compounds were used as received. The solid pure components were weighted in the desired molar fractions and melting-mixed. Using melamine-formaldehyde resin (MFR) as the shell material and n-alkane or n-alkane mixture as the core material, microcapsules were prepared by in situ polymerization according to the literature.33 This method provides us with nearly monodispersed and highly heat-resistant microcapsules, inside which n-alkanes and their mixtures were confined to individually small microdomains surrounded by a noncrystalline wall of MFR. The particle size and surface morphology of the prepared microcapsules were examined by a JEOL-JSM-6700F scanning electron microscope (SEM), fitted with a field emission source and operated at an accelerating voltage of 5 kV. The calorimetry measurements were performed using a Mettler DSC822e calorimeter. The samples sealed in aluminum pans were heated from 25 to 50 °C, followed by a cooling scan and then a second melting scan. Both the cooling and heating rates were 2 °C/ min. The first cooling and second heating thermograms were recorded. Temperature dependent X-ray diffraction (XRD) measurements were performed on a Rigaku D/max-2500 X-ray diffractometer, which works in the reflection mode with Cu KR radiation (λ ) 1.54 Å), power of 200 mA/40 kV. The data were collected from 2 to 50° with the 2θ step of 0.02°. The percentage of surface monolayer molecules inside a single microcapsule can be calculated approximately as F ) (nj × 1.25 Å × 4πr12)/(4/3πr13), where nj is the average carbon number in n-alkane mixtures and r1 corresponds to the radius of the core, assuming that the average length of the C-C single bond is 1.25 Å in the all-trans conformational chain. According to the hypothesis that the wall and core material have the same density:34 r1 ) [Wc/(Ww + Wc)]1/3r2, where r2 corresponds to the radius of the shell and Wc and Ww represent the weight of the core and wall layer, respectively. The quantities of Wc/(Ww + Wc) and r2 can be obtained from DSC and SEM measurements, respectively.
Figure 1. SEM micrographs of the microcapsules containing alkanes prepared by in situ copolymerization of melamine and formaldehyde. Triton X-100 was used as an emulsifier.
Figure 2. DSC curves of m-C18/C19 as the functions of molar fraction (x of C19) and temperature during the cooling process.
Results and Discussion Surface Freezing of n-Alkane Mixtures in Confined Geometry. The microcapsules prepared show the mean diameter of ca. 3 µm and narrow size distribution with a porous surface (Figure 1). The volume percentage of the surface monolayer molecules inside a microcapsule was calculated to be ca. 1%.34 Such a result means that 1% of the total alkane molecules stand at the core-shell interface of the microcapsules, which consequently exert a prominent influence on the bulk crystallization of n-alkanes. As shown in Figure 2, above the bulk freezing temperature of about 3 °C, a small, sharp exothermic peak emerges for all m-C18/C19 samples, corresponding to the surface freezing. The monolayer at the free surface of the binary mixture in bulk has been observed by X-ray reflectivity, grazing incidence X-ray diffraction, and surface tension measurements as well.22-24 Here, we detected it directly by the usual DSC method because of the higher surface-to-volume ratio of alkane molecules in confined geometry. In microcapsules, the surface freezing temperature (Ts) varies continuously and monotonically with x (the molar fraction of C19) between C19 and C18 (Figure 3, upper line), which indicates a continuous change in the
Figure 3. The surface freezing temperature (Ts) and bulk freezing temperature (Tb) in m-C18/C19 as a function of composition (x of C19).
composition of the surface layer. Since both C19 and C18 molecules are aligned parallel and packed randomly into the surface layer of the microcapsules, the competition between entropic mixing and repulsive interaction should be prominent. Apparently, the difference in carbon numbers between C19 and C18 is negligible, so the entropic mixing is dominant and a mixed homogeneous crystalline monolayer is favored. It has been reported that a surface freezing monolayer would serve as an ideal nucleation site for the bulk rotator phase.4 Therefore, it is reasonable to assume that it is the surface freezing monolayer in alkane mixtures that induces the nucleation of the bulk rotator phase in microcapsules, which also exhibits a monotonous
Nucleation and Crystallization Behavior of n-Alkanes
J. Phys. Chem. B, Vol. 112, No. 51, 2008 16487 TABLE 1: Phase Transition Temperatures of n-Alkanes, n-Alkane Mixtures, and Their Confined Samplesa rotator phases sample
liquid
RI
C18 C18/C19 ) 95/5 C19 m-C18 m-C18/C19 ) 95/5 m-C19
26.2 25.6 30.6 24.9 25.3 30.1
18.8 20.3 18.5 13.6 15.0
stable phases T
O
• • • • • •
a
The stable phases at low temperature are indicated with the symbol •.
Figure 5. DSC curves of m-C18, m-C19, C18/C19 ) 95/5, and m-C18/ C19 ) 95/5 during the cooling process.
Figure 4. The supercooling of phase transition in C18/C19 and m-C18/ C19 as a function of composition (x of C19): (a) rotator-to-crystal transition; (b) liquid-to-rotator transition.
increase of bulk freezing temperature (Tb) with the increase of C19 concentration (Figure 3, lower line). From Heterogeneous to Homogeneous Nucleation of n-Alkane Mixtures in Microcapsules. The homogeneous nucleation of alkanes was first measured by Turnbell and Cormia35 and followed by Herhold et al.,36 Montenegro and co-workers,29,37 and Kraack et al.,38 using the droplet method. It has been shown that the large supercooling in emulsions of alkanes is usually regarded as an indication of homogeneous nucleation. Comparing Figure 4a with Figure 4b, it is easy to find that the supercooling difference between m-C18/ C19 and C18/C19 in the whole compositional range for rotatorto-crystal transition (about 3 °C) is larger than that for liquidto-rotator transition (less than 1 °C), implying that the two transition mechanisms are different in nucleation kinetics. We speculate that for n-alkane mixtures in microcapsules the rotator-to-crystal transition is a solid-to-solid homogeneous nucleation, while the liquid-to-rotator transition is a surface-induced heterogeneous nucleation. Note that the nucleation mechanisms in microcapsules are different from those in emulsions, where both the rotator-to-crystal transition and the liquid-to-rotator transition are caused by homogeneous nucleation.27,29,36 Here, m-C18/C19 ) 95/5 was used as an example for the detailed explanation of two different nucleation kinetics. The phase transition temperatures of C18, C19, C18/C19 ) 95/5,
m-C18, m-C18/C19 ) 95/5, and m-C19 are different from each other (Table 1). As shown in Figure 5, a small, sharp exothermic peak emerges above the bulk freezing temperature of about 3 °C for all of the microencapsulated samples, which is absent for all of the unencapsulated ones. This small peak corresponds to the surface freezing of n-alkanes. For m-C18/ C19 ) 95/5, a big exothermic peak appears at 25.3 °C corresponding to the liquid-to-RI transition, while a small exothermic peak is observed at 13.6 °C corresponding to the RI-to-triclinic transition. The temperatures of phase transition observed in DSC curves are consistent with those observed in X-ray diffraction (Figure 6). For the m-C18/C19 ) 95/5 system, with sample cooling to 26 °C, there are two characteristic peaks of (110) and (200) for the orthorhombic rotator phase RI.39 When the temperature was lowered to 14 °C, another four stronger characteristic diffraction peaks for the triclinic phase were observed. There is an overlap between the (200) reflection of the RI phase and the (100) reflection of the triclinic phase. Obviously, a state of triclinicorthorhombic coexistence was detected in the m-C18/C19 ) 95/5 system between 16 and 12 °C. Below 12 °C, the state of triclinic-orthorhombic coexistence changes to a single triclinic phase state. Both XRD and DSC results show that the temperature of liquid-to-solid transition (26 °C) in m-C18/C19 ) 95/5 is quite similar to that in C18/C19 ) 95/5 (27 °C). However, the temperature of solid-to-solid transition (15 °C) in m-C18/C19 ) 95/5 is much lower than that in C18/C19 ) 95/5 (21 °C). This can be explained in terms of different nucleation mechanisms. The bulk C18/C19 ) 95/5 follows the classic heterogeneous nucleation mechanism; i.e., some critical
16488 J. Phys. Chem. B, Vol. 112, No. 51, 2008
Jiang et al. Conclusions Confining different alkane chain lengths into microcapsules causes enhancement of the surface-to-volume ratio of molecules with respect to the bulk, which provides an unusual opportunity for investigating the nucleation and phase behavior of n-alkane mixtures. The present work reported the surface freezing and nucleation mechanisms in a binary alkane mixture of n-C18H38 and n-C19H40 in microcapsules. It was found that the properties of the surface freezing monolayer in m-C18/C19 vary continuously with the temperature and composition. The supercooling difference between m-C18/C19 and C18/C19 for the rotator-to-crystal transition is larger than that for the liquid-to-rotator transition. It is proposed that the rotator-to-crystal transition corresponds to the confinement-induced solid-to-solid homogeneous nucleation, the liquid-to-rotator transition to the surface-induced liquid-to-solid heterogeneous nucleation. This work may help to explain the nucleation and phase behavior of other materials with different chain lengths such as polymers, biological materials, and especially some phase change materials for thermal energy storage. Acknowledgment. The financial support from the National Natural Science Foundation of China (No. 50573086) was acknowledged. References and Notes
Figure 6. X-ray diffraction patterns of C18/C19 ) 95/5 (a) and m-C18/ C19 ) 95/5 (b) at selected temperature on cooling.
nuclei clusters of C19 are first formed in the bulk, which lower the nucleation barrier and induce the bulk heterogeneous nucleation, followed by the crystal growth. For m-C18/C19 ) 95/5, it is possible to form two kinds of nuclei including a surface freezing monolayer and critical nuclei clusters of C19. The surface freezing monolayer is preferential and prevalent at the liquid-to-solid transition, which serves as an ideal nucleation site for the bulk rotator phase. In this case, the nucleation mechanism is a surface-induced heterogeneous nucleation mechanism. As for the solid-to-solid transition, the nucleation cannot be induced by the surface freezing any more and only the critical nuclei clusters of C19 act as the nucleation sites. However, these nuclei clusters formed are now distributed among a large number of isolated microcapsules. In a single microcapsule, the nuclei concentration is so small that the heterogeneous nucleation can be prevented until such a low temperature is reached, at which homogeneous nucleation in the microcapsule begins. Therefore, the crystal grows rapidly only in that one microcapsule but does not affect the others. As a result, the solid-to-solid nucleation mechanism is shifted from heterogeneous nucleation to homogeneous nucleation.
(1) Small, D. M. The Physical Chemistry of Lipids; Plenum: New York, 1986. (2) Sirota, E. B.; Wu, X. Z.; Ocko, B. M.; Deutsch, M. Phys. ReV. Lett. 1997, 79, 531. (3) Ocko, B. M.; Wu, X. Z.; Sirota, E. B.; Sinha, S. K.; Gang, O.; Deutsch, M. Phys. ReV. E 1997, 55, 3164. (4) Wu, X. Z.; Ocko, B. M.; Sirota, E. B.; Sinha, S. K.; Deutsch, M.; Cao, B. H.; Kim, M. W. Science 1993, 261, 1018. (5) Mondieig, D.; Espeau, P.; Robles, L.; Haget, Y.; Oonk, H. A. J.; Cuevas-Diarte, M. A. J. Chem. Soc., Faraday Trans. 1997, 93, 3343. (6) Oonk, H. A. J.; Mondieig, D.; Haget, Y.; Cuevas-Diarte, M. A. J. Chem. Phys. 1998, 108, 715. (7) Clavell-Grunbaum, D.; Strauss, H. L.; Snyder, R. G. J. Phys. Chem. B 1997, 101, 335. (8) Dirand, M.; Bouroukba, M.; Chevallier, V.; Petitjean, D.; Behar, E.; Ruffier-Meray, V. J. Chem. Eng. Data 2002, 47, 115. (9) Dorset, D. L. Macromolecules 1985, 18, 2158. 1986, 19, 2965. (10) Dorset, D. L.; Snyder, R. G. J. Phys. Chem. 1996, 100, 9848. (11) Mondieig, D.; Rajabalee, F.; Metivaud, V.; Oonk, H. A. J.; CuevasDiarte, M. A. Chem. Mater. 2004, 16, 786. (12) Robles, L.; Mondieig, D.; Haget, Y.; CuevasDiarte, M. A.; Alcobe, X. Mol. Cryst. Liq. Cryst. 1996, 281, 279. (13) Sirota, E. B.; King, H. E.; Shao, H. H.; Singer, D. M. J. Phys. Chem. 1995, 99, 798. (14) Snyder, R. G.; Clavell-Grunbaum, D.; Strauss, H. L. J. Phys. Chem. B 2007, 111, 13957. (15) Snyder, R. G.; Goh, M. C.; Srivatsavoy, V. J. P.; Strauss, H. L.; Dorset, D. L. J. Phys. Chem. 1992, 96, 10008. (16) MacDowell, L. G.; Vega, C.; Lopez-Rodriguez, A. J. Chem. Phys. 1999, 111, 3183. (17) Xia, T. K.; Landman, U. Science 1993, 261, 1310. (18) Wickramarachchi, P.; Spells, S. J.; de Silva, D. S. M. J. Phys. Chem. B 2007, 111, 1604. (19) Annis, B. K.; Londono, J. D.; Wignall, G. D.; Snyder, R. G. J. Phys. Chem. 1996, 100, 1725. (20) Sirota, E. B.; King, H. E.; Hughes, G. J.; Wan, W. K. Phys. ReV. Lett. 1992, 68, 492. (21) Ungar, G.; Masic, N. J. Phys. Chem. 1985, 89, 1036. (22) Sloutskin, E.; Sirota, E. B.; Kraack, H.; Ocko, B. M.; Deutsch, M. Phys. ReV. E 2001, 64, No. 031708. (23) Wu, X. Z.; Ocko, B. M.; Tang, H.; Sirota, E. B.; Sinha, S. K.; Deutsch, M. Phys. ReV. Lett. 1995, 75, 1332. (24) Sloutskin, E.; Wu, X. Z.; Peterson, T. B.; Gang, O.; Ocko, B. M.; Sirota, E. B.; Deutsch, M. Phys. ReV. E 2003, 68, No. 031605. (25) Xie, B. Q.; Shi, H. F.; Jiang, S. C.; Zhao, Y.; Han, C. C.; Xu, D. F.; Wang, D. J. J. Phys. Chem. B 2006, 110, 14279.
Nucleation and Crystallization Behavior of n-Alkanes (26) Xie, B. Q.; Shi, H. F.; Liu, G. M.; Zhou, Y.; Wang, Y.; Zhao, Y.; Wang, D. J. J. Phys. Chem. B, in press. (27) Montenegro, R.; Antonietti, M.; Mastai, Y.; Landfester, K. J. Phys. Chem. B 2003, 107, 5088. (28) Herhold, A. B.; Ertas, D.; Levine, A. J.; King, H. E. Phys. ReV. E 1999, 59, 6946. (29) Montenegro, R.; Landfester, K. Langmuir 2003, 19, 5996. (30) Henschel, A.; Hofmann, T.; Huber, P.; Knorr, K. Phys. ReV. E 2007, 75, No. 021607. (31) Huber, P.; Soprunyuk, V. P.; Knorr, K. Phys. ReV. E 2006, 74, No. 031610. (32) Huber, P.; Wallacher, D.; Albers, J.; Knorr, K. Europhys. Lett. 2004, 65, 351. (33) Xie, B. Q.; Shi, H. F.; Liu, G. M.; Zhou, Y.; Wang, Y.; Zhao, Y.; Wang, D. J. Chem. Mater. 2008, 20, 3099.
J. Phys. Chem. B, Vol. 112, No. 51, 2008 16489 (34) Sliwka, W. Angew. Chem., Int. Ed. Engl. 1975, 14, 539. (35) Turnbull, D.; Cormia, R. L. J. Chem. Phys. 1961, 34, 820. (36) Herhold, A. B.; King, H. E.; Sirota, E. B. J. Chem. Phys. 2002, 116, 9036. (37) Montenegro, R.; Antonietti, M.; Mastai, Y.; Landfester, K. J. Phys. Chem. B 2003, 107, 5088. (38) Kraack, H.; Deutsch, M.; Sirota, E. B. Macromolecules 2000, 33, 6174. (39) Sirota, E. B.; King, H. E.; Singer, D. M.; Shao, H. H. J. Chem. Phys. 1993, 98, 5809.
JP807347D
