Designed CO2-Philes Stabilize Water-in-Carbon Dioxide Microemulsions.

Angewandte
Chemie
Green Solvents
DOI: 10.1002/ange.200600397
Designed CO2-Philes Stabilize Water-in-Carbon
Dioxide Microemulsions**
Julian Eastoe,* Sarah Gold, Sarah Rogers, Paul Wyatt,
David C. Steytler, Alexandre Gurgel,
Richard K. Heenan, Xin Fan, Eric J. Beckman, and
Robert M. Enick
Herein is reported the first direct structural evidence for the
formation of nanodomains in water-in-carbon dioxide (w/c)
microemulsions by two non-fluorinated surfactants. Generation of O-surfactants, by incorporating oxygen into the
surfactant tails, significantly improves CO2-philicity, affording
stabilization of w/c structures. Tests with standard hydrocarbon surfactants (H-surfactants) demonstrate that they are
unable to support true w/c phases (see Supporting Information). This finding represents an important step forward in the
design of commercially viable and environmentally responsible CO2-philes, which has potential to expand research and
commercial applications with supercritical or liquid CO2 as a
solvent.
It is widely recognized that waste petrochemical solvents
pose a huge environmental threat; recently significant efforts
have been made to develop dense CO2 as an alternative
“green” solvent.[1] CO2 is well suited to applications in food or
pharmaceutical industries since it is environmentally benign,
biocompatible, nonflammable, nontoxic, cheap, abundant,
and crucially, unregulated by the US Environmental Protection Agency. However, there are severe limitations because
CO2 is generally a poor solvent, especially for polar and/or
high-molecular-weight solutes. Although water and carbon
[*] Prof. J. Eastoe, S. Gold, S. Rogers, Dr. P. Wyatt
School of Chemistry, University of Bristol
Bristol, BS8 1TS (UK)
Fax: (+ 44) 117-928-9180
E-mail: julian.eastoe@bris.ac.uk
Dr. D. C. Steytler, Dr. A. Gurgel
School of Chemical Sciences and Pharmacy
University of East Anglia, Norwich NR4 7TJ (UK)
Dr. R. K. Heenan
ISIS-CCLRC Rutherford Appleton Laboratories
Chilton OX11 0QX (UK)
X. Fan, Prof. E. J. Beckman, Prof. R. M. Enick
Department of Chemical and Petroleum Engineering
University of Pittsburgh, Pittsburgh, PA 15261 (USA)
[**] S.G., A.G., and S.R. thank the University of Bristol DTA, CNPq
Brazil, and EPSRC (EP/C523105/1) for studentships and a postdoctoral fellowship, respectively. CCLRC are thanked for beam time
and consumables/travel grants. Alan Pitt (Kodak (UK)) is thanked
for stimulating discussions. The University of Pittsburgh would like
to express its appreciation to the US DOE NETL for supporting this
research through contract DE-FG26-04NT-15533.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. 2006, 118, 3757 –3759
dioxide are weakly compatible (e.g. solubility of water in CO2
at 15 8C, 450 bar is approximately 0.14 %), surfactants and
polymers are needed to increase this low level, by stabilizing
the CO2–water interface. Hence, solubility levels in CO2 could
be significantly enhanced, and controlled by incorporation of
pressure-sensitive reversed micelles or w/c microemulsion
nanodroplets.
A recent review outlines the state-of-the-art in CO2
surfactants;[1] the most efficient of these contain fluorine,
making them environmentally and commercially unacceptable. It has been shown that high-pressure small-angle
neutron scattering (HP-SANS) is one of the most reliable
techniques for detecting aggregation in CO2,[1] since it
generates an unmistakable fingerprint SANS signal, which is
characteristic of micelles or water nanodroplets (see Supporting Information). Reports have emerged, mainly based on
spectral probe dye evidence, suggesting microemulsion formation in CO2 with commercially available non-ionic hydrocarbon surfactants,[2, 3] however, HP-SANS experiments do
not support these claims (see Supporting Information). HPSANS has provided direct proof of micelle formation with a
custom-designed CO2 surfactant (AOT4), an analogue of
common Aerosol-OT (AOT), but bearing tert-butyl rather
than methyl chain tips as for AOT.[4] Although AOT4 did not
disperse water, another designer surfactant comprising twin
vinyl acetate oligomeric chains AO-VAc (Figure 1) has shown
Figure 1. CO2-philic surfactants used to stabilize water-in-carbon dioxide microemulsions: sodium bis(5,5-dimethyl-4-oxo-hexyloxycarbonyl)
sulfosuccinate (Aerosol-octyl-ketone or AOK) and sodium bis(vinyl
acetate)8 sulfosuccinate (Aerosol-vinyl acetate or AO-VAc with n = 8).
promise as a stabilizer for w/c phases.[5] The design of AOVAc was based on simulations demonstrating enhanced
solubility of simple acylated sugars with CO2,[6] hence the
VAc chains are expected to promote CO2-philicity compared
to normal hydrocarbon tails. To confirm that oxygenated
chains are essential for stabilization of w/c nanodroplets the
surfactant AOK (Figure 1) was synthesized, bearing tert-butyl
and carbonyl groups in the chain tips. Furthermore, two tails
are often needed to induce reversed interfacial curvature.[1, 4, 5]
HP-SANS experiments indicated that a range of standard
commercially available hydrocarbon surfactants do not
stabilize nanodroplet water pools in CO2 (see Supporting
Information). However, with the oxygenated AOK, SANS
signals were detected: after successive additions of D2O there
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3757
Zuschriften
was a progressive increase in intensity and a shift of the peak
maximum to lower scattering vector Q, consistent with
swelling of water nanodomains. With the oxygenated surfactant AO-VAc, SANS signals were not detected until the water
loading reached wcorr = 10.9 (corresponding to an uncorrected
w = 50),[5] and the inset to Figure 2 shows this curve. Figure 3
particle interference effects this scattering law gives rise to a
Qmax peak in the SANS profile, which is the signature of a
core–shell structure (Figure 2, and Supporting Information).
This form factor[8, 9] was fitted to the I(Q) curves in Figure 2,
with the best fit parameters: core radius (r), shell thickness (t), and polydispersity (s/r) given in Table 1. The fits
Table 1: Parameters obtained from polydisperse spherical core–shell
form factor analyses of SANS data.[a]
Surfactant
wcorr
r [M]
t [M]
s/r
AOK
8.5
19.0
29.5
10.9
15
17
21
55
8
9
10
26
0.15
0.15
0.15
0.20
AO-VAc
[a] Uncertainties on r and t are 1 M.
Figure 2. SANS data (symbols) and form factor fits (lines) for surfactantstabilized D2O-in-CO2 microemulsions as a function of water loading
wcorr = {[D2O]added [D2O]CO2}/[surf]. Fitted parameters are given in Table 1.
Surfactant concentrations AOK = 50 mmol dm 3 ( 2.4 wt %) and AOVac = 6.2 mmol dm 3 ( 1 wt %). Experimental conditions 500 bar and 45 8C
for AOK, but 25 8C for AO-VAc. Example error bars are shown for one AOK
sample. Inset: SANS data for AO-Vac at wcorr = 10.9.
Figure 3. Schematic scattering length density (sld or 1 K 1010 cm 2)
profile fitted to SANS data from surfactant-stabilized D2O-in-CO2
microemulsion droplets. The values fixed in the form factor analyses
were 2.4 for the external CO2 phase (outer gray ring), 0.3 for the AOK
surfactant shell (shown as lines), and 6.4 K 1010 cm 2 for the D2O core
(small gray spheres). See text for details. For the solvent, 1CO2 was
calculated as in ref. [7].
shows the expected neutron scattering length density (1)
profile for spherical D2O nanodroplets coated by a surfactant
monolayer, and dispersed in CO2.[7] For this configuration the
I(Q) distribution should be described by a spherical core–
shell form factor,[8] modified with a Schulz polydispersity
function, as is routinely used in neutron contrast variation
studies of water-in-oil microemulsions.[9] Owing to intra-
3758
www.angewandte.de
shown in Figure 2 are reasonable given weak SANS intensities owing to poor contrast and low concentrations: for AOVAc, the peak position is well reproduced, but not the
intensity maximum. However, it is reassuring to see that the
fitted shell thickness t increases from AOK to AO-Vac, both
values being consistent with expected molecular lengths
(Table 1). Therefore, these two new CO2-philes show
common macroscopic phase behavior and core–shell nanodroplet structures.
Compared to AOK and AO-VAc, common commercial
surfactants display only weak aggregation (see Supporting
Information) highlighting the need for specialized CO2-philic
groups to stabilize the water–CO2 interface. Comparing the
behaviors of AOK and CO2-inactive Aerosol-OT,[4] shows
that only minor chemical variations at the chain-tip extremities have significant effects on CO2-philicity. This mirrors
extensive studies of hydrocarbon and fluorocarbon AOT
analogues at various model oil–water interfaces,[10] pointing to
new design approaches for CO2-philes, which could unlock
the vast potential of CO2 as a green processing, cleaning and
separation medium.[11]
Experimental Section
The surfactants were synthesized and characterized as detailed
elsewhere (Supporting Information and ref. [5]). The w/c samples
were formulated using D2O to provide contrast (Goss UK 99.9 % Datom) in a high-pressure cell on the LOQ SANS beam line at ISIS,
UK, following experimental protocols described elsewhere.[4] The
system variables were temperature T, pressure P, surfactant concentration, and water content, which was corrected for the partitioning of
added D2O into CO2 so that wcorr = {[D2O]added [D2O]CO2}/[surf]. Both
surfactants exhibited similar P-T phase behavior as seen with a
certain fluorinated CO2-philic surfactants. At low-P catastrophic
separations were observed close to CO2 vapor pressure, with
apparently biphasic regions up to the maximum operating pressure
of the HP-SANS cell (500 bar), comprising a majority transparent
CO2-rich phase in equilibrium with an approximately 2–5 % by
volume of a dense clear phase. The SANS data shown in Figure 2
were recorded after 5 min of vigorous stirring, so that the neutron
beam illuminated CO2-continuous phases. For AOK without added
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. 2006, 118, 3757 –3759
Angewandte
Chemie
water (wcorr = 0), no SANS was detected, suggesting absence of
micellization in CO2.
Received: January 30, 2006
Published online: April 28, 2006
.
Keywords: green chemistry · microemulsions ·
small-angle neutron scattering · surfactants
[1] J. Eastoe, S. Gold, Phys. Chem. Chem. Phys. 2005, 7, 1353 – 1362.
[2] a) J. Liu, B. Han, J. Zhang, G. Li, X. Zhang, J. Wang, B. Dong,
Chem. Eur. J. 2002, 8, 1356 – 1360; b) J. Liu, B. Han, G. Li, X.
Zhang, J. He, Z. Liu, Langmuir 2001, 17, 8040 – 8043; c) J. Liu, J.
Zhang, M. Tiancheng, B. Han, B. G. Li, J. Wang, B. J. Dong, J.
Supercrit. Fluids 2003, 26, 275 – 280; d) J. Liu, B. Han, Z. Wang, J.
Zhang, G. Li, G. Yang, Langmuir 2002, 18, 3086 – 3089.
[3] W. Ryoo, S. E. Webber, K. P. Johnston, Ind. Eng. Chem. Res.
2003, 42, 6348 – 6358.
[4] a) J. Eastoe, A. Paul, S. Nave, D. C. Steytler, E. Rumsey, M.
Thorpe, B. H. Robinson, R. K. Heenan, J. Am. Chem. Soc. 2001,
123, 988 – 989; b) J. Eastoe, A. Dupont, D. C. Steytler, M.
Thorpe, A. Gurgel, R. K. Heenan, J. Colloid Interface Sci. 2003,
258, 367 – 373.
[5] X. Fan, V. K. Potluri, M. C. McLeod, Y. Wang, J. Lui, R. M.
Enick, A. D. Hamilton, C. B. Roberts, J. K. Johnston, E. J.
Beckman, J. Am. Chem. Soc. 2005, 127, 11 754 – 11 762.
[6] P. Raveendran, S. L. Wallen, J. Am. Chem. Soc. 2002, 124, 7274 –
7275.
[7] J. B. McClain, D. Londono, J. R. Combes, T. J. Romack, D. A.
Canelas, L. Betts, G. D. Wignall, E. T. Samulski, J. M. DeSimone,
J. Am. Chem. Soc. 1996, 118, 917 – 918.
[8] I. Markovic, R. H. Ottewill, D. J. Cebula, I. Field, J. Marsh,
Colloid Polym. Sci. 1984, 262, 648 – 656.
[9] J. Eastoe, S. Nave, A. Paul, A. R. Pitt, R. K. Heenan, Langmuir
2005, 21, 10 021 – 10 027.
[10] J. Eastoe, S. Gold, R. Tabor, Langmuir 2006, 22, 963 – 968.
[11] “ICI enters CO2 dry cleaning”: A. Tullo, Chem. Eng. News 2002,
80(35), 12.
Angew. Chem. 2006, 118, 3757 –3759
2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.de
3759