The Journal of
Physical Chemistry
0 Copyright, 1989, by the American Chemical Society
VOLUME 93, NUMBER 25 DECEMBER 14, 1989
LETTERS Clathrate Hydrate Formation in Reversed Micellar Solutions Huyen Nguyen, John B. Phillips, and Vijay T. John* Department of Chemical Engineering, Tulane University, New Orleans, Louisiana 70118 (Received: August 1 1 , 1989)
We describe the thermodynamic conditions for the nucleation of clathrate hydrates in a new environment, that constituting the microaqueous pools of reversed micellar solutions. Hydrate formation is highly dependent on the water-to-surfactant molar ratio w owhich defines the size of the micelles and influences the state of the microaqueous phase. Hydrate formation behavior approaches that in pure water, as the microaqueous droplets approach pure water properties. The ability of hydrates to nucleate from the water in reversed micelles may have implications to the solubilizing properties of the micelles and to the behavior of macromolecular solutes in the microaqueous phase.
Introduction Gas hydrates are crystalline clathrate or inclusion compounds, formed from water (the host species) and small gas molecules (the guest species). They form in one of two structures; a unit cell of structure I hydrate is shown in Figure l a illustrating the cagelike lattice formed by the hydrogen-bonded water molecules. Gas molecules enclosed within the cavities stabilize the structure through van der Waals interactions with the host water molecules. From a thermodynamic viewpoint, hydrates exhibit interesting univariant properties, the two-component (water gas) system existing in three-phase (gas + liquid waterlice hydrate) equilibria.’ Prior studies of hydrates have focused on hydrate formation in aqueous environments, primarily for relevance to gas recovery from hydrates in suboceanic and arctic environments.2 This study describes the nucleation of the clathrate hydrate of methane in a new environment, that constituting the microaqueous phase within reversed micelles. These water-in-oil microemulsions, shown schematically in Figure lb, are capable of solubilizing a variety of compounds through encapsulation in the microaqueous phase.
+
+
*To whom correspondence should be addressed.
0022-365418912093-8 123$01.50/0
For example, protein-containing reversed micelles have been extensively studied for their biomembrane mimetic properties3 and for their potential in biocatalysis4-’ and protein extraction proc e s s e ~ . ~Micelle ,~ radii in reversed micelles can easily exceed 2.5 nmIbl2 while clathrate hydrates have unit cells of sizes 1.2 nm (1) van der Waals, J. H.; Platteeuw, J. C. Adv. Chem. Phys. 1959, 2, 1. (2) Berecz, E.; Balla-Achs, M. Gas Hydrates; Elsevier: New York, 1983. (3) Wirz, J.; Rosenbusch, J. P. In Reverse Micelles; Luisi, P. L., Straub, B. E., Eds.; Plenum Press: New York, 1984. (4) Leser, M. E.; Wei, G.; Luthi, P.; Haering, G.; Hochkoeppler, A.; Bliichliger, E.; Luisi, P. L. J . Chim. Phys. 1987, 84, 9, 1113. (5) Martinek, K.; Berezin, I. V.; Khmelnitski, Yu. L.; Klyachko, N. L.; Levashov, A. V. Biocatalysis 1987, I , 9. Trotta, E.; (6) Magid, L.; Walde, P.; Zampieri, G.; Battistel, E.; Peng, Q.; Maestro, M.; Luisi, P. L. Colloids Surf. 1988, 30, 193. (7) Luisi, P. L.; Giomini, M.; Pileni, M. P.; Robinson, B. H. Biochim. Biophys. Acta 1988, 947, 204. (8) Leser, M. E.; Wei, G.; Luisi, P. L.; Maestro, M. Biochem. Biophys. Commun. 1986, 135, 629. (9) Rahaman, R. S.; Chee, J. Y.; Cabral, J. M. S.; Hatton, T. A. Biotechnol. Prog. 1988, 4, 218. (IO) Waks, M. Proteins: Struct., Funct., Genet. 1986, I , 4. (1 1) Sheu, E.; Goklen, K. E.; Hatton, T. A.; Chen, S.-H.Biotechnol. Prog. 1986, 2, 175.
0 1989 American Chemical Society
8124 The Journal of Physical Chemistry, Vole 93, No. 25, 1989
Letters /
a
wo = 15. [AOT] = 555 mM
/
/ /
-
.
3.01&5 0
Hydrates Present No Hydrates
- - - P* - T *
LOCUS
...................................... 273.2 275.2 277.2 279.2
'
Temperature (K)
Figure 2. Pressure vs temperature trajectories for hydrate dissociation in a reversed micellar solution.
ul
-
a
No [H20xNl= l 15 [AOTl=SSSmU A w0 = 15. = 55.5 mU W ~ ~ - 1 0[kXI=SS.SmM ,
C
.-
[m
V
formation = -63.5 kJ/g.mol
Figure 1. Illustrations of (a) a unit cell of gas hydrate of structure I and (b) a reversed micelle.
(structure I) and 1.73 nm (structure II).l This intuitively implies the possibility of hydrate nucleation within the microaqueous droplets under appropriate thermodynamic conditions. Our initial study described here characterizes the thermodynamics of hydrate formation in a macroscopically homogeneous reversed micellar phase containing empty reversed micelles (without any solute, or added electrolyte).
Experimental Section Components of the reversed micellar solution included sodium bis(2-ethylhexyl) sulfosuccinate (AOT; Aldrich, 96% and 99% purities), water (double distilled, no added electrolyte), and isooctane (Aldrich, spectrophotometric grade). Methane (Matheson, ultrahigh purity) was used as the hydrate-forming gas species. The experiments were carried out in a high-pressure glasswindowed cell suspended in a temperature-controlled bath. The setup is similar to that used by Holder13 for hydrate formation studies in bulk aqueous systems. Pressures were measured accurate to 6.85 kPa (1 psi), and the temperature was controlled to fO.l K. An optically clear reversed micellar solution was introduced into the cell, and hydrate formation was initiated by pressurizing and cooling the cell. Nucleation of hydrates from the water in reversed micelles leads to the formation of a solid hydrate phase at the cell bottom (hydrates being denser than the bulk organic) which was visually observed. The supematznt phase retains optical clarity . Results and Discussion Figure 2 illustrates our procedure for determining univariant hydrate dissociation pressure data as a function of temperature, for a specific reversed micellar solution with initial composition w0([H20]/[AOT]) = 15 and [AOT] = 555 mM. Hydrates were first formed by cooling the cell at elevated pressures. Then slowly
273.2
274.2
275.2
wo=5,
[ror] =55.5mU
---- wmta fmm fin wotar* 276.2
277.2
Temperature T* (K)
Figure 3. Equilibrium dissociation pressure vs temperature data for hydrate formation in reversed micelles.
heating the cell in increments, the equilibrated pressure was monitored at each temperature increment. Approximately 24 h was required for equilibration at each temperature. Several of these experimental pressure vs temperature trajectories are shown in Figure 2 for the specified reversed micellar solution. The transition from the hydrate-present stage to the no-hydrate stage is the"incipient" hydrate formation condition or the equlibrium dissociation pressure P* at a temperature P.The locus of P * - P data represents the univariant dissociation pressure vs temperature curve for the specified solution. The procedure of Figure 2 was followed for other reversed micellar solutions with different compositions. Characteristics of the univariant equilibrium for hydrate formation/dissociation from the various reversed micellar solutions are thus delineated in Figure 3. The slopes of the curves of Figure 3 are all approximately equal; thus, the heats of hydrate formation in reversed micelles, as calculated from the Clapeyron equation, are the same as that for methane hydrate formation in pure water (-63.5 kJ/gmol gas). The interesting observations relate to the quantity wo, the water/surfactant molar ratio. It is seen that increasing the surfactant concentration at constant wo does not affect hydrate formation conditions, while decreasing wo leads to an increased difficulty in hydrate formation. These observations may be qualitatively attributed to the nature of the water pool in reversed micelles. Increasing the surfactant concentrations at constant wo (and therefore higher water concentrations) simply increases the number of micelles without significantly affecting size and aggregation number.I4 Higher AOT concentrations at constant wo therefore do not affect the nature of water in reversed micelles and thus have no effect on hydrate formation. This is true at least over the surfactant concentration range studied; it ~~~~
(12) Chatenay, D.;Urbach, W.; Nicot, C.; Vacher, M.; Waks, M. J. Phys. Chem. 1987, 91, 2198. (13) Holder, G. D.; Hand, J. H. AIChE J. 1982, 28, 440.
278.2
~~~~~
(14) Fletcher, P. D. I.; Howe, A. M.; Perrins, N. M.; Robinson, B. H.;
Toprakcioglu, C.; Dore, J. C. In Surfactants in Solution; Mittal, K. L., Lindmann, B., Eds.; Plenum: New York, 1984; Vol. 3, p 1745.
The Journal of Physical Chemistry, Vol. 93, No. 25, 1989 8125
Letters may be expected that, at very high surfactant concentration, micelle-micelle interactions would affect aggregrate characteristics. On the other hand, it has been shown that increasing wo increases the size and aggregation number of reversed micelles.'*I2 Deutron spin relaxation studiesIs indicate that the water pool within reversed micelles contains localized or bound water near the oil-water interface; as the center of the micelle is approached, the water molecules approach bulk water properties. At an wo of 4 or less, all intramicellar water is bound and insusceptible to freezing.15 Indeed, our experiments showed extremely little hydrate formation with a reversed micellar solution of wo = 5; taking the pressure to 7.5 MPa visually indicated practically no additional hydrate formation. Pressure limitations of the cell prevented achieving higher pressures. We thus postulate that the ease of hydrate formation is inversely correlated to the degree of binding of water molecules. Additional insight may be gained from the thermodynamic model for hydrate equilibria'
the dissociation pressure through the Langmuir isotherm' (4)
At 273.15 K, the Langmuir constants for the single species (methane) in the small and large cavities of structure I are 1.3 and 12.0 MPa-I, respectively.18 By use of the extrapolated values of P,,(or P*), at T* = 273.15 K from Figure 3, the calculated fractional occupancies of the hydrate cavities formed from reversed micelles of wo = 15 are 0.78 for the small cavities and 0.97 for the large cavities. At wo = 10, the small cavity occupancy increases to 0.8 with the large cavities staying almost completely occupied. Finally, at wo = 5 , the fractional occupancy of the small cavity increases to 0.86. In an independent experiment, we formed hydrates from a reversed micellar solution, originally at w,, = 15, and stabilized the system at 6.5 MPa and 273.85 K. A sample of the supernatant solution was taken out at these conditions and brought to atmospheric pressure and room temperature, and the water content was measured through Karl Fischer analysis (Mettler DL-18). pWH(T,P) = p,@(T,P)- kTCvi In (1 CCijPj) (1) Assuming constant AOT concentration in the supernatant, the i i measured water content indicated a reversed micellar solution of w, = 4.2, thus verifying removal of water and the reduction of where pwHrefers to the chemical potential of water in the solid wo through hydrate formation. hydrate phase, pw@refers to the chemical potential of water in A brief discussion of where exactly the hydrates form is also the hypothetical j3 phase of empty hydrate cavities, vi is the number of use. In our experiments described here, we start with a maof cavities of type i per water molecule, Cij is the Langmuir croscopic single-phase reversed micellar system. Initial nucleation constant of gas speciesj (methane in this case) in a cavity of type of hydrates thus must occur in the microaqueous pools. Since i, and Pi is the partial pressure of gas species j . The Langmuir the micelles are dynamic structures that continually break open constant is the configuration integral describing guest-host inand re-form through collisions, it is felt that the first hydrate teractions, and is fixed for a given species j at a temperature T . crystals are precipitated during this process. We have visually At equilibrium, the chemical potentials of water in the hydrate observed that larger precipitated crystals are formed if the phase (pwH)and in the reversed micellar phase (pwM)are equal; high-pressure cell is kept stationary. Perhaps micelle collisions i.e. with initially precipitated hydrate crystals lead to subsequent M nucleation and growth of the crystals. P w H = Pw In macroscopic two-phase systems (a reversed micellar solution in equilibrium with a bulk aqueous phase) hydrate formation with would utilize both reversed micellar water and bulk water, since the activities of water in both phases must be equal. Continued P , ~ ( T , P= ) rWo(T,P)+ kT In uwM (3) hydrate formation would eventually convert all the bulk water to solid form. Subsequently, further hydrate formation from the where pwois the chemical potential of pure water and awMrefers water in the reversed micelles would tend to decrease micelle size. to the activity of water in reversed micelles (reference activity of In a recent publication, Fulton and co-workers19 showed evidence unity for pure water). Equations 1-3 thus relate the dissociation of hydrate formation when water was admitted to a system pressure P* (=P.for a single gas species) to the activity of water containing AOT and supercritical xenon. The phenomenon in reversed micehes, which is a function of ionic species concenpresented is that of xenon hydrate formation both from liquid tration and distribution. We get an estimate of average counterion water and from the water in reversed micelles existing in the (Na+) concentration within the micelles, by using the size-wo correlation given by Pileni and co-workers16 ( R (nm) = 0 . 1 5 ~ ~ ) supercritical dense gas phase. together with an approximate surfactant heat group diameter of Conclusions 0.8 nmI2 to get the aggregation number which should also be the These initial studies with clathrate hydrate formation in reversed number of cations in the water pool per micelle. Thus, at an wo micelles indirectly reveal very interesting aspects on the behavior of 15 and a micelle radius of 2.25 nm,I6 the average mobile of the microaqueous phase. We have shown that the conditions counterion concentration is 4.4 g.ion/L microaqueous water. At for hydrate formation are dependent on the water-to-surfactant an wo of 10, the micelle radius is about 1.5 nm leading to an molar ratio wo and that, at each wo, the system is thermodyaverage concentration 6.65 gion/L, and at wo = 5 (micelle radius namically univariant. To some extent, the ability of hydrates to 0.75 nm), a concentration of 13.2 gion/L. The increase in micellar form in reversed micelles can be used as a probe to characterize electrolyte concentration with wo is only a crude indication of the the micellar state since the incipient hydrate formation condition increase in the dissociation pressure with decreasing woe A comfor a reversed micellar solution is characteristic of the wo of the prehensive model is under development, to correlate water activity solution, as shown by the wo-P*-T* relationship of Figure 3. to the electrostatic potential in the water pool as described by Further work needs to be done to understand hydrate formation Kubik et al." and thence to the equilibrium dissociation pressure in reversed micelles containing added electrolyte. The effect of through eqs 1-3. hydrate formation on the solubility of macromolecules (e.g. A calculation of the fractional occupancy of the hydrate cavities proteins) within reversed micelles is another important area to formed from micelles at different wo yields information on the be researched. It has been reported that protein conformation amount of occluded gas required to reorient the water to crystalline and enzyme activity in reversed micelles are very dependent on hydrate form and to stabilize the hydrate crystal. The fractional the water-to-surfactant molar ratio, wpZo Several enzymes have occupancy of gas species j , in a cavity of type i, flu, is related to
+
(15) Quist, P.-0.; Halle, B. J . Chem. Soc., Furuduy Trum. 1 1988, 84, 1033. (16) Pileni, M.-P.; Zemb, T.; Petit, C. Chem. Phys. Lett. 1985, 118, 4. (17) Kubik, R.; Eicke, H.-F.; J6nsson, B. Helu. Chim. Acto 1982,65, 170.
(18) John, V. T.; Papadopoulos, K. D.; Holder, G. D. AIChE J . 1985,31, 252. (19) Fulton, J. L.; Blitz, J. P.; Tingey, J. M.; Smith, R. D. J. Phys. Chem. 1989, 93, 4198. (20) Kabanov, A. V.; Levashov, A. V.; Klyachko, N . L.; Namyotkin, S. N.; Pshezhetesky, A. V.; Martinek, K. J . Theor. Eiol. 1988, 133, 327.
8126
J . Phys. Chem. 1983, 93, 8126-8127
maximum activity at optimal wo values ranging from 10 to 15. At the same time, adjusting a reversed micellar solution of high wo to the optimal range by adding surfactant may not improve catalytic efficiency due to adverse proteinsurfactant interactions.% The ability to adjust wo through hydrate formation (keeping
surfactant concentration constant) may thus have implications to the design of protein-containing reversed micellar systems. Acknowledgment. Support from the National Science Foundation (Grant CBT-8802564) is gratefully acknowledged.
Electron Afflnity of Chlorine Dioxide L. M. Babcock,**+T. Pentecost: and W. H. Koppeno1t.t Department of Chemistry and Biodynamics Institute, Louisiana State University, Baton Rouge, Louisiana 70803 (Received: September 5, 1989)
The flowing afterglow technique was used to determine the electron affinity of chlorine dioxide. A value of 2.37 f 0.10 eV was found by bracketing between the electron affinities of HS' and SF4 as a lower limit and that of NOz as an upper limit. This value is in excellent agreement with 2.32 eV predicted from a simple thermodynamic cycle involving the reduction potential of the C102/C102- couple and a Gibbs hydration energy identical with that of SO;-.
Introduction Recently, the reduction potential of the formate radical, COz' was calculated to be -1.8 V from the electron affinity of carbon dioxide and a Gibbs hydration energy estimated to be similar to those of other bent triatomic anions, namely, 03'-, NOT, and S02*-.l The assumption that these anions have similar Gibbs hydration energies can be tested by determining the electron affinity of the chlorine dioxide radical and calculating the Gibbs hydration energy from the difference with the well-known reduction potential of the CIOz,o/CIO;ao couple. In the literature the following values for the electron affinity of chlorine dioxide are found: 2.8 eV,2 3.4 eV? 1.8 eV$ and a range of 1.3-2.2 eV.5 If one assumes, given the similarity in molecular parameters: that the Gibbs solvation energy of ClO; is similar to that of SOz'-, 134 kJ/mol' relative to A,-G(H+) = 0, a value of 2.32 eV is predicted from the simple thermodynamic cycle: EA &Go(CIOz-) = Eo(C102,g/CIO~ao).The parameters used in this calculation are the reduction potential of Eo(CIOz,ao/CIOz~ao), 0.934 V at 25 OCJ', and the Gibbs hydration energy of chlorine dioxide, -0.4 kJ/moL9 As shown below, a value close to this theoretical estimate is found. An accurate value for the electron affinity of chlorine dioxide might possibly be relevant to reactions of this molecule in the stratosphere. At night chlorine dioxide acts as a reservoir of chlorine monoxidelo which acts catalytically in the destruction of ozone."
+
Experimental Section Chlorine dioxide was synthesized according to Bray.12 Briefly, 15 g of oxalic acid and 4 g of potassium chlorate were mixed in a three-neck round-bottom flask that was subsequently kept at 55-60 OC in a water bath. Reaction 1 started after addition of 2C103-
+ HZCz0,
-
2C10z
+ 20H- + 2C02
(1) 2 mL of water. Yellow chlorine dioxide gas, free from chlorine, evolved slowly for about 3 h. The presence of carbon dioxide minimizes the risk of an e x p l ~ s i o nand ' ~ did not interfere with subsequent reactions. As an additional precaution, the reaction vessel was shielded from direct light by aluminum foil. The employed synthesis is considered safe,13 and it is therefore unfortunate that the experimental details are not mentioned in a recent monograph on chlorine dioxide.I4 Excess chlorine dioxide was allowed to bubble through a solution of sodium hydroxide solution where it disproportionated. Department of Chemistry. *Biodynamics Institute.
0022-3654/89/2093-8126$01.50/0
Experimental determination of the electron affinity of chlorine dioxide was carried out on a flowing afterglow apparatus which has been described in detail e1se~here.l~The buffer gas employed in all cases was helium which, prior to introduction into the flow tube, was passed through a molecular sieve a t 77 K to trap any condensable impurities. All reactions were studied a t ambient temperatures at a pressure of 0.3 Torr (40 Pa), corresponding to typical helium flows of on the order of 10 standard liters/min. For charge-exchange reactions involving neutral chlorine dioxide, reactant anions were generated from corresponding neutrals by electron attachment. The NOz-, SF,, SF6-,and SO2- reactant ions were produced from the corresponding parent neutral molecules, while SF5-, HS-, C1-, Br-, and I- were generated by dissociative electron attachment to SF,, HzS, CF2CIz,CH3Br, and CH31, respectively. The NO< anion was produced from NO2. These reactant anions were produced upstream near the ion source, and neutral C102 was introduced approximately 65 cm downstream. C l o y was produced upstream via electron transfer from SF6-1 (3) Reaction 2 occurs at approximately the theoretical collision capture Koppenol, W. H.; Rush, J. D. J . Phys. Chem. 1987, 91, 4429-4430. Weiss, J. Tram. Faraday Soc. 1947, 43, 173-177. Pritchard, H. 0. Chem. Rev. 1953,52, 529-563. Baluev, A. B.;Nikitina, 2.K.; Fedorova, L. I.; Rosolovskii, V. Ya. Izu. Akad. Nauk SSSR, Ser. Khim. 1980, 9, 1963-1971. (5) Wecker, D.; Christodoulides, A. A.; Schindler, R. N. Inr. J. Mass Specrrom. Ion Phys. 1981, 38, 391-406. (6) Stanbury, D. M.; Lednicky, L. A. J . Am. Chem. Soc. 1984, 106, (1) (2) (3) (4)
2847-2853. (7) Troitskaya, N. V.;Mishenko, K. P.; Flis, I. E.Russ. J . Phys. Chem. (Engl. Transl.) 1959, 33, 77-79. (8) Klining, U. K.; Sehested, K.; Holcman, J. J . Phys. Chem. 1985, 89, 76C-763. (9) Wagman, D. D.; Evans, W. H.; Parker, V. B.; Schumm, R. H.; Halow, 1.; Bailey, S. M.; Churney, K. L.; Nuttal, R. L. J. Phys. Chem. Ref. Data 1982, 11, 37-38. (10) Solomon, S.; Mount, G. H.; Sanders, W.; Jakoubek, R. 0.; Schmeltekopf, A. L. Science 1988, 242, 550-555. (1 1) McElroy, M. B.; Salawitch, R. J. Science 1989, 243, 763-770. (12) Bray, W. Z . Phys. Chem. 1906,54, 569-608. (13) Abegg, R.; Auerbach, F. Handbuch der Anorganischen Chemie; S . Hirzel: Leipzig, 1913; 4 Band, 2 Abt., pp 169-170. (14) Masschelein, W. J. Chlorine Dioxide: Chemistry and Environmental
Impact of Oxychlorine Compounds; Ann Arbor Science Publishers: Ann Arbor, MI, 1979. (15) Babcock, L. M.; Taylor, W. S.;Herd, C. R. Inr. J . Mass Specfrom. Ion Processes 1987, 81, 259-272.
0 1989 American Chemical Society
