Properties of hydrocarbon-in-water emulsions stabilized byAcinetobacter RAG-1 emulsan

Properties of Hydrocarbon-in- Water Emulsions Stabilized by Acinetobacter RAG-1 Emulsan

ZINAIDA ZOSIM, DAVID GUTNICK, and EUGENE ROSENBERG, Department of Microbiology. George S. Wise Center of Life Sciences,

Tel Aviv University, Ramat Aviv, Israel

summary

Emulsan is a polymeric extracellular emulsifying agent produced by Acbrrrobacrer RAG-1. Hydrocarbon-in-water emulsions (V, of hydrocarbon of 0.01-0.10) were stabilized by small quantities of emulsan (0.02-0.2 mg/mL). Although both aliphatic and aromatic hydrocarbon emulsions were stabilized by emulsan. mixtures containing both aliphatics and aromatics were better substrates for emulsan than the individual hydrocarbon by themselves. The emulsan remained tightly bound to the hydrocarbon even after centrifugation as determined by a) residual emulsan in the aqueous phase and b) the fact that the resulting “cream” readily dispersed in water to reform stable emulsions. With hexadecane-to-emulsan weight ratio of 39 and 155. the noncoalescing oil droplets had average droplet diameters of 2.0 and 4.0 p i . respectively. Dialy- sis studies showed that the water-soluble dye Rhodamine B adsorbed tightly to the interface of hexadecane-emulsan droplets although the dye did not bind to either hexadecane or emulsan alone. At saturating concentrations of dye. 2.2 Fmol of dye were bound per mg emulsan.

INTRODUCTION

The growth of microorganisms on hydrocarbons is often accompanied by the emulsification of the insoluble carbon source in the culture medium. In many cases, this has been attributed to the production of an extracellular bioemulsifier. Recently we reported the purification, partial chemical char- acterization,5-6 and substrate specificity’ of the emulsifier of Acinetobacter RAG-1. The polymeric substance (referred to as emulsan) had an average molecular weight of 9.9 X 10j and an intrinsic viscosity of 505 om3 per gram. Emulsan contains: 1) a polysaccharide backbone composed of N-acetyl- D-galactosamine, N-acetylaminouronic acid, and an unidentified amino sugar; 2) esterified fatty acids (0.5 pmol/mg) consisting primarily of a- and /3-hydroxydodecanoic acid; and 3) proteins which can be removed by phenol treatment without destruction of emulsifying activity or loss of viscosity. The experiments presented here were performed to investigate further the man- ner by which emulsan interacts with hydrocarbons and the properties of the emulsan-hydrocarbon complex.

Biotechnology and Bioengineering, Vol. XXIV, Pp. 281-292 (1982) 0 1982 John Wiley & Sons. Inc. CCC 0006-3592/82/020281-12$01.20

282 ZOSIM, GUTNICK. AND ROSENBERG

MATERIALS AND METHODS

Materials

Olefin-free hexadecane (99% purity) was obtained from Fluka Chemical Co., Switzerland. Other paraffins and aromatic hydrocarbons were reagent grade products of either Merck or Aldrich Chemicals. Kerosine and gas-oil were obtained from the Haifa Refinery, Haifa, Israel. Agha Jari crude oil was obtained from the Ashkelon-Eilat Pipeline Co., Israel. Rhodamine B (C28H3,- CIN203), recrystallized from ethanol, had a molar extinction coefficient of 0.9 X lo5 at 560 nm. in close agreement with reported values.8 TM buffer consisted of 0.02M Tris (hydroxymethyl), aminomethane (Tris) hydrochlo- ride buffer (pH 7.2), and lOmM MgS04.7H20. Glycerol (anhydrous) was a highly purified, redistilled product of Merck. Tritiated water (0.5 mCi/mL) was obtained from the Nuclear Research Center, Negev. Israel.

Emulsan, the extracellular emulsifying agent of Acinetobacter calcoaceticus RAG-1 (ATCC 31012), was purified from a cell free supernatant obtained from an ethanol grown culture both by ammonium sulfate precipitation4ca) and cetyltrimethyl ammonium bromide fractionation. 4(b) The emulsan used in these studies contained 17% protein. Apoemulsan, prepared by a modifi- cation5 of the hot phenol method, contained loss than 5% protein. Methyl cellulose (degree of substitution, 1.6) was a product of BDH (England). Xan- than9 (XFL-14630) was a product of XANCO Oil Field Products, Inc. The viscosities of 0.5 mg/mL solutions of emulsan, apoemulsan, methyl cellulose, and xanthan at 30°C were 0.98. 1.40, 1.20, and 2.47 centipoise, respectively.

Em ulsioii Stability Measurements

Hydrocarbon-in-water emulsions were prepared by ultrasound treatment, using the Braun Labsonic 1510 instrument. Ten milliliters of the hydrocarbon-water mixture in a cellulose nitrate tube (Beckman) was exposed to sonic oscillation for 15 s at a constant power value of 30 W. Unless otherwise stated, the emulsifiers were added prior to ultrasound treatment. The emulsions that were formed were transferred immediately to calibrated Klett tubes and turbidity measured using a Klett-Summerson photoelectric color- imeter fitted with a green filter. Reported turbidity values were corrected using a standard calibration curve prepared b y serially diluting a hexadecane- 2-methylnaphthalene emulsion. Stabilities of emulsions were determined by measuring the turbidity after standing undisturbed at 25 k 2°C for 1-120 h. The percent stability is defined as the Klett (KU) units after 24 h. divided by the turbidity immediately after sonication, times one-hundred.

Dye - Bin ding Studies

A spectrophotometric method was developed for quantitative estimation of Rhodamine B binding to emulsan-stabilized hexadecane droplets. Emulsions

EMULSION STABILIZED HYDROCARBON-IN-WATER EMULSIONS 283

were prepared in the presence of varying concentrations of the dye as described above. The emulsions were transferred to glass test tubes and allowed to stand undisturbed for 21 h (equilibrium was reached during the first 15 h). Samples of the emulsions were then centrifuged at 8 X 103g for 20 min at 25°C. Under these conditions the droplets rose to the surface and did not in- terfere with absorbance readings of the aqueous phase. The residual dye (equilibrium concentration of the dye in the aqueous phase) was determined from trichromic readings (2ASm - AM - Am) by comparison with a standard calibration curve of known concentrations of the dye in TM buffer. When Rhodamine B concentration exceeded 0.01 mg/mL, dilutions were made in the buffer before determining absorbance.

General Analytical Methods

Protein concentrations were determined by the method of Lowry et al., l o

using bovine serum albumin as a standard. Viscosity was measured in an Ostwald-Fenske microviscometer (water value, 55.1 s) at 30"C, calibrated with 20 and 50% glycerol. Emulsan concentration was determined by a standard emulsification assay4 and by the microplate modification of the en- zyme-linked immune sorbent assay ELISA test, in which antiemulsan IgG prepared against purified emulsan first was linked covalently to alkaline phosphatase with gluteraldehyde. This material was then bound to varying quantities of emulsan previously immobilized on the walls of an IgG-coated microplate. The ELISA test yielded the same quantitative results regardless of whether the antigen was emulsan or apoemulsan. I I ( b ) Size distribution of hexadecane droplets stabilized by emulsan was determined both by microscopy (see the Results section) and using a Coulter counter model ZB.

RESULTS

RAG-1 emulsan enhanced both the formation and stability of hexadecane- in-water emulsions prepared by ultrasonic treatment (Fig. 1). Using a volume fraction (V,) of hexadecane of 0.01, the initial turbidities varied from 520 Klett units (no emulsan) to 2600 Klett units (0.1 mg of emulsan/mL). On standing, the hexadecane droplets in the control coalesced and rose rapidly to the surface; by 10 h the control emulsion was completely broken. In the presence of either 0.05 and 0.1 mg/mL emulsan (hexadecane:emulsan weight ratios of 155 and 77, respectively) there was a small decrease in turbidity for 2 h, after which the emulsions remained relatively stable for the next 40 h. With a higher ratio of hexadecane-to-emulsan of about 400 (0.02 mg emulsan/mL) the turbidity dropped sharply during the initial 17 h of standing and then remained relatively stable for the remainder of the experiment. Slightly more stable emulsions were obtained when emulsan was added immediately after, rather than prior to, ultrasonic treatment. (This

284 ZOSIM, GUTNICK. AND ROSENBERG *

T I M E , HRS

Fig. 1. Stability of hexadecane-in-water emulsions as a function of emulsan concentration. Hexadecane (0.1 mL) in 10 mL of TM buffer was emulsified by exposure to sonic oscillation for 15 s in the presence of: ( 2 ) 0.0, (A) 0.2 mg, (B) 0.5 mg, or ( 0 ) 1.0 mg of emulsan.

might be due to an approximately 10% decrease in viscosity of the emulsan during the sonic treatment).

The effect of emulsan on the formation and stability of different hydrocarbon-in-water emulsions is summarized in Table I. Although the stability of all hydrocarbon emulsions tested was enhanced in the presence of 0.05-0.1 mg emulsan/mL, the effect varied significantly with the particular hydrocarbon examined. In general, the higher the molecular weight of the liquid hydrocarbon, the more effective emulsan was in inducing and stabilizing the emulsion. This was true both for aliphatic and aromatic hydrocarbons, even though the initial turbidities of aromatic hydrocarbon emulsions were much higher than aliphatic hydrocarbon emulsions. Mixtures containing both aliphatics and aromatics were invariably better substrates for emulsan than the individual hydrocarbons by themselves. For example, initial turbidities of 2-methylnaphthalene and hexadecane emulsions (formed in the presence of 0.05 mg emulsan/mL) were 7600 and 3150, respectively, whereas for the 1 : 1 mixture of the two hydrocarbons the initial turbidity was 1.7 X lo4; similarly the stabilities of the individual hydrocarbons were 60 and 53% after 24 h, while the stability of the emulsified mixture was 71%. Emulsions prepared from each of the complex hydrocarbon mixtures-crude oil, gas-oil, and kerosine-were stabilized by 0.05 mg/mL emulsan.

Table 11 and Figure 2 compare two RAG-1 emulsifier preparations with two polymeric emulsifiers, methylcellulose and xanthan. Both emdsan and apmmulsan (deproteinked emulsan) were more effective than the two

EMULSION STABILIZED HYDROCARBON-IN-WATER EMULSIONS 28.5

TABLE I Various Hydrocarbon-in-Water Emulsions Stabilized by Emukana

Apparent Emulsm T u r b i d i t y (Kle t t u n i t s 1 stability

Ilyd1.oc.:l 1.11011 (iiig/nil) t - o hr t:1 111. (%)

A l i p h a t i c s

Octane

Dodccclnc

Tetratlccane

I-lexadccnne

Aromatics

To 1 uene

p-Xylene

2 -Met hylnapht ha1 ene

Te t rahydronapht ha 1 ene

bI i x t tires

llexndecnne : 2 methyl - naphthalene

Kerosine

Gas-oi l

Crude o i l

0 . 1 1200 1 1 2

0 . 1 1700 850

0 . 1 2 7 2 0 1500

0 . 1 5150 1900

0 .05

0.1

0.05

0.05

0 . 0 5

0 . 0 5

0.05

0.05

0.05

3600

4400

3080

7600

7600

; 7000

6700

.>-so 0 2 K

7 , -

165

475

x n 4000

3800

12000

3400

7 h 5 0

1400

9 . 3

50

55

60

4.6

11

8.4

53c

50

71

51

83

'1

"Emulsions were prepared as described in Fig. 1 with V, of hydrocarbon of 0.01. bApparent stability is defined as the turbidity of the emulsion at 24 h divided by turbidity at

t = 0 X 100. Controls for each hydrocarbon (without emulsan) yielded 24-h turbidity values that were less than 3% of the corresponding values for the emulsan experiment.

'This experiment was performed at 35°C to avoid solidification of the hydrocarbon.

TABLE I1 Stabilization of Hexadecane-in-Water Emulsions by Different Emulsifiers"

Apparent S t a b i l i t y (%) Emu1 s i f i e r 0.02 mg/ml 0.05 mg/ml 0 .1 mg/ml

Emulsan 1 3 63 74

Apoemul san 14 62 75

Methylcel l u l o s e 4 13 27

Xan t han 10 19 21

aExperiments were performed as described in Fig. 1 with a V, of hexadecane of 0.01. Apparent stabilities of emulsions prepared without emulsifier were less than 2%.

286

I I I I I I

I

ZOSIM, GUTNICK. AND ROSENBERG

I I 1

c \ \ - I \

IN IT IAL P E R I O D T I M E H R S S E C O N D P E R I O D

Fig. 2. Stability and reversibility of hexadecane-in-water emulsions. The emulsions were prepared as described in Fig.] using: ( 0 ) 1 mg of emulsan. (B) 1 mg of methylcellulose. or (A) 1 mg of xanthan. After measuring the decrease in turbidities of the emulsions for 120 h. the samples were mixed gently by hand and turbidities determined for an additional 120 h (second period).

mercial emulsifiers in stabilizing hexadecane-in-water emulsions over the entire concentration range studied. It should be noted that native emulsan (containing approximately 20% protein) is more effective than apoemulsan in forming emulsions. However. if the hydrocarbon-in-water emulsion is preformed by ultrasound treatment, the protein does not play a crucial role in emulsion stabilization. At a V, of hexadecane of 0.01 and an emulsifier concentration of 0.1 mg/mL, the turbidities of emulsions (after standing for 120 h) stabilized by emulsan, xanthan. and methylcellulose were 1420, 160, and 155 KU. respectively (Fig. 2, initial period).

The lowering of turbidity may be the result of: a) coalescence of the hydrocarbon droplets giving rise to a complete phase separation; or b) creaming of stabilized hydrocarbon droplets which float to the surface. In the case of creaming, the hydrocarbon droplets would be expected to reform a homoge- nous emulsion with similar stability characteristics upon gentle agitation. The results summarized in Figure 2 (second period) illustrate the turbidities of partially broken emulsan mixtures which were agitated by hand and sub- sequently allowed to stand for an additional 120 h. Breakage of xanthan- stabilized emulsions was brought about primarily by coalescence of oil droplets since breakage could not be reversed. Upon gentle mixing the turbidity rose only to 305 Klett units and then dropped rapidly to 150 Klett units. Mi- croscopic examination provided further evidence that breakage of xanthan- stabilized emulsions was due to coalescence rather than by flocculation. In contrast, breakage of emulsan-stabilized emulsions was due to creaming, since gentle mixing totally reversed the process (2250 Klett units initially


compared to 2300 Klett units after mixing at 120 h). It can be seen that breakage during the second period was similar to that which was observed in the initial 120 h. Furthermore, cream obtained by centrifugation of emulsan- stabilized emulsions readily became suspended in aqueous solution to reform emulsions which did not differ significantly from that obtained initially by ultrasonic treatment. The methylcellulose-stabilized emulsion was an interme- diate case; breakage was due to both creaming and coalescence. The hydrocarbon upper layer reformed an emulsion when mixed with water to yield a turbidity similar to that produced by ultrasonic treatment. However, emulsion breakage was much more rapid during the second period than during the initial 120 h.

A photomicrograph of an emulsan-stabilized hexadecane-in-water emulsion is shown in Figure 3 (lower field). Analysis of a number of such micrographs yielded an average droplet diameter of 4.0 pm when the weight ratio

Fig. 3. Light microscopy of hexadecane-in-water emulsions stabilized by emulsan. Hexa- decane (0.2 d) in 10 d of TM buffer was emulsified by sonic oscillation for 15 s in the presence of 0.2 mg of emulsan. The lower field shows the emulsion immediately after sonic treatment; the upper field shows the cream formed after several days standing. Magnification in both cases is 389 X .

288 ZOSIM, GUTNICK, AND ROSENBERG

of hexadecane-to-emulsan was 155 (Fig. 4). Emulsions produced under similar conditions yielded an average droplet diameter of 3.6 pm when measured with a Coulter counter. Ninety percent of the droplets had a diameter of less than 6.1 pm. Upon prolonged standing the droplets rose and concentrated at the surface forming a cream (Fig. 3, upper field). After standing for one week, the cream was separated from the aqueous phase. Less than 10% of the emulsan was found in the aqueous phase as determined both immuno- logically (see the Materials and Methods section) and by the standard emulsifier assay.4

The water content of the different samples of cream, calculated from the density of the cream at 20°C compared to the known densities of water and hexadecane, was found to be 40-60%. Following centrifugation at 8 X 103g for 40 min, more viscous creams were obtained which consisted of about 65% hexadecane and 35% water. Even after the centrifugation procedure there was no coalescence of oil droplets. The water content of the cream was also determined using tritiated water as a tracer. With a V, of hexadecane of 0.1 and an emulsan concentration of 0.5 mg/mL, the resulting cream contained 43% water after standing three days and 30% water after standing ten days and then centrifuging at 8 X 103g for 30 min.

A qualitative experiment demonstrating binding of the water-soluble dye Rhodamine B to emulsan-stabilized hexadecane droplets is shown in Table 111. When the hexadecane emulsion was prepared in the presence of both emulsan and Rhodamine B, the dye was rendered nondialiiable for at least 24 h. However, the dye was released during the first hour of dialysis if either emulsan or hexadecane was omitted, or if the emulsion was stabilized by

DROPLET DIAMETER, M I C R O N S

Fig. 4. Sue distribution of hexadecane-in-water droplets stabilized by emulsan. Emulsions were prepared in 10 mL of TM buffer containing 0.1 mL of hexadecane and 0.5 mg of emulsan. Diameters of 1724 droplets were determined from enlargements of micrographs.

EMULSION STABILIZED HYDROCARBON-IN- WATER EMULSIONS 289

TABLE 111 Binding of Rhodamine B to Emulsified Hexadecane Stabilized by RAG-1 Emulsan

Observation a f t e r 24 h of d i a lys i sb Reaction mixturea Insidc (10 mL) Dialysate (800mL)

1. Complcte Intcnsc p i n k , tu rb id Colorlcss

2. Complete minus hexadecane Colorless, c l e a r Faint pink

3. Complete minus emulsan Colorless, two phases Faint pink

4. Complete minus hexadecnne Colorless, c l e a r Faint pink and emulsan

5 . Complete minus emulsan Colorless, t u rb id Faint pink p lus 1 .0 mg methyl- ce l lu lose

'The complete reaction mixture consisted of 0.1 mL of hexadecane, 1 .O mg of emulsan, and 0.1 mg of Rhodamine B in a final volume of 10 mL of TM buffer.

bAfter exposing the mixture to sonic oscillation as described in the Materials and Methods section, the entire volume was placed in a cellophane bag and dialyzed for 24 h against 800 mL. of distilled water.

methylcellulose instead of emulsan. Even when much higher concentrations of methylcellulose were used to stabilize the emulsions, the dye was not re- tained in the dialysis bag. Thus, Rhodamine binding was a property of the emulsan-hydrocarbon complex and not due simply to the greater hydrocarbon-water interface.

Adsorption of Rhodamine B to hexadecane-in-water droplets stabilized by emulsan was further investigated by determining residual Rhodamine spectrophotometrically following separation of the cream from the aqueous phase (Fig. 5). At low concentrations of Rhodamine, more than 80% of the dye was adsorbed. Since the cream phase represented less than 5% of the total volume of liquid, the effective concentration of dye into the cream was greater than 75-fold. At saturating concentrations of dye, about 2.2 pmol of dye were bound per 1.0 mg of emulsan.

The data of Figure 5 were replotted according to the Freundlich empirical adsorption equation1* in logarithmic coordinates (Fig. 6 ) . With initial dye concentrations from 0.2-6 pmol, a straight line was obtained, indicating a single type of binding. From the logarithmic isotherm curve, the adsorption constants k and n were estimated to be 0.29 and 2.8, respectively.

DISCUSSION

The data presented here demonstrate that emulsan, the water-soluble, extracellular bioemulsifier of Acinetobacter calcoaceticus RAG- 1, stabilizes a


0.45

y1 z 2 4 0 K n 0.30 z 2 0 m - 0

* p1 - E“

8 a 0.15

I I I I

1 I I I 1.0 2 .O

p mcles RHODAMINE ADDED

Fig. 5. Adsorption of Rhodamine B to hexadecane-in-water droplets stabilized by emulsan. Experiments were performed with: a) Vf of hexadecane of 0.01 and emulsan concentration of 0.1 mg per mL ( X ) or, b) V f of hexadecane of 0.02 and an emulsan concentration of 0.02 mg per mL (C). Rhodamine binding was measured spectrophotometrically following separation of the “cream” from the aqueous phase (see the Materials and Methods section).

wide variety of hydrocarbon-in-water emulsions. The emulsan binds tightly to the surface of hydrocarbon droplets, presumably forming a strong polymeric film on the droplet surface which prevents coalescence. This provides a rationale for the previously reported finding that the average size of hydrocarbon droplets formed in water by mechanical agitation depended upon the weight ratio of hydrocarbon-to-emulsan, rather than simply on emulsan concentration. Furthermore, the tight binding of emulsan to hexadecane-water interfaces explains the failure to find emulsan in clarified culture broths of hexadecane-grown Acinetobacter RAG-1.

The stability-reversibility phenomenon, as summarized in Figure 2, may be considered to reflect the affinity of the various emulsifiers for hexadecane droplets; xanthan < methylcellulose < emulsan. This relationship may be due to the presence of hydrophobic groups (or sites) on the structure of the three polysaccharides. Emulsan is the most hydrophobic of the gel polymers, containing C12 fatty acid side chains, in addition to N-acyl groups.

From the weight ratio of hydrocarbon-to-bound-emulsan and the average droplet diameter (as determined by two independent methods: microscopy and Coulter-counter techniques), it is possible to estimate the thickness of the emulsan film around the oil droplet. For a weight ratio of 155, the mean


-08

I ’ I I I

-

1 I I I I

LOG Ceq

Fig. 6. The logarithmic plot of the adsorption isotherm of Rhodamine onto the hexadecane droplet interface (stabilized by emulsan). The Freundlich equation (ref. 12) applied to adsorption from solution takes the form of x/m = kc l /n , where x is the amount of solute (Rhodamine) adsorbed by a mass m of adsorbent (emulsan), and R and n are constants. Data of Fig. 5 are replotted.

droplet diameter was 3.8 pm. The volume of hexadecane per droplet, V H , is 4/3 rr$, where rH is the radius of the inner hexadecane sphere; the volume of emulsan per droplet, V,, is ?3 r r i - 4/3 rr$= 4/3 r(i-$- I;), where rT = 1.9 pm is the radius of the total droplet (hexadecane sphere plus emulsan shell).

Dividing, 3 v,- - r H

V , 6.86 - r;

also,

where the ratio of masses m H / m E = 193 (corrected for fraction of emulsan bound), and the densities of emulsan and hexadecane are 1.4MS and 0.773 g/cm3, respectively.

VH - 1 350.5 = VE 6.86 - r$

or rH = 1.898 pm, and the thickness of the emulsan film, rT - rH, equals 0.002 pm or 20 A. In the same way, for a weight ratio of hexadecane-to- emulsan of 39 which gave a mean droplet diameter of 2.0 pm, it was found


that the film thickness was 42 A. These values are minimum estimates since they assume the tightest possible packing of the emulsan molecules. The data are consistent with early results showing that polysaccharides tend to lie flat on the oil-water interface. l3 In general, coalescence is inversely related to the thickness of the film formed on the droplet; strong polymeric films with a thickness greater than 5 A have been shown to prevent coalescence. l4

Hexadecane droplets stabilized with emulsan can be separated from the bulk water phase by centrifugation. We refer to this cream phase, consisting of small hydrocarbon droplets coated with emulsan and water, as “emulsanasol.” With a weight ratio of hexadecane-to-emulsan of 155, the resulting emulsanasol contained 30-50% water. Not surprisingly, emulsanasols have properties characteristic of neither the pure hydrocarbon nor aqueous solutions of emulsan. One such interesting property is the ability to bind and thereby concentrate the water-soluble dye Rhodamine B. At saturating concentrations of the dye, 2.2 pmol of dye were bound per mg of emulsan in the emulsanasol. Since emulsan contains only 1 .5 microequivalents of carboxyl residues per mg,5 simple ionic binding between the tertiary amine groups of Rhodamine and the carboxyl groups could not explain the phenomenon. Rather, emulsan on the interfacial film in contact with the hydrocarbon may explain better the binding properties of emulsanasols.

References

1. L. E. Erickson and T. Nakahara, Roc. Biochem.. 10, 9 (1975). 2. J. E. Zajic and C. J. Panchal, Crit. Rev. Microbiol.. 5 , 39 (1976). 3. D. L. Gutnick and E. Rosenberg, Annu. Rev. Microbiol.. 31, 379 (1977). 4. (a) E. Rosenberg, A. Zuckerberg, H. Rubinowitz, and D. L. Gutnick, Appl. Environ.

5. A. Zuckerberg, A. Diver, Z. Peen, D. L. Gutnick, and E. Rosenberg, Appl. Environ.

6. I. Belsky, D. L. Gutnick, and E. Rosenberg, FEES Lett.. 101, 175 (1979). 7. E. Rosenberg, A. Peny, D. T. Gibson, and D. L. Gutnick, Appl. Environ. Microbiol..

8. C . R. Singleteny and L. A. Weinberger, J. Am. Chem. Soc.. 73, 4574 (1951). 9. E. R. Moms, in Extracellular Microbial Polysaccharides. P. A. Sandford and A. Laskin,

10. 0. H. L o w j , N. J. Rosebrough, A. L. Farr, and R. J. Randall, J. Biol. Chem.. 193, 265

11. (a) E. Engrall and P. Perlmann,J. Zmmunol., 109, 129, (1972); (b) S. Goldman, in prep-

12. D. J. Shaw, in Introduction to Colloid and Surface Chemistry. 2nd ed. (Butterworths,

13. N. K. Adams, Trans. Faraday Soc.. 29, 90 (1933). 14. B. Biswas and D. A. Haydon, Kolloid-Z. Z. Polym.. 185, 31 (1962).

Microbiol.. 37, 402 (1979); (b) I. Belsky, unpublished.

Microbiol.. 37, 414 (1979).

37, 409 (1979).

Eds. (ACS Symp. 45, Washington, DC, 1977), pp. 81-89.

(1951).

aration.

London, 1970), pp. 120-132.

Accepted for Publication June 18, 1981

Documents

Properties of hydrocarbon-in-water emulsions stabilized byAcinetobacter RAG-1 emulsan