Biotechnology Letters Vol 5 No ii

725-730

(1983)

BACTERIA-INDUCED DE-EMULSIFICATION OF WATER-IN-OIL PETROLEUM EMULSIONS A.L. Stewart, N.C.C. Gray, WLL. Cairns and N. Kosaric Chemical and Biochemical Engineering Faculty of Engineering Science The University of Western Ontario London, Ontario, Canada N6A 5B9 SUMMARY: The de-emulsifying a b i l i t y of two bacteria ( i . e . , N. o~narae and C. p e t r o p h i l u m ) has been demonstrated on w a t e r - i n - o i l petroleum f i e l d emulsions. De-emulsification by whole cultures and by washed cells was correlated with changes in the culture during growth. For each species, there was a p a r t i c u l a r growth phase which produced a species-dependent de-emulsification optimum. INTRODUCTION Petroleum f i e l d emulsions generated during conventional, combustion, .and steam-stimulated recovery processes, are very diverse in composition and properties. In order to break such a v a r i e t y of complex emulsions, a de-emulsifier must have properties which are compatible with a broad range of emulsions, be modifiable to possess the desired properties f o r a given emulsion, or be blended with other de-emulsifiers to achieve the desired properties. B a c t e r i a l cell surfaces can act as de-emulsifying agents of simple, defined model w a t e r - i n - o i l emulsions (Cooper e t a l . , 1980 a, b; Cairns e t a l . , 1981, 1982). The de-emulsifying a c t i v i t y of these surfaces depends on the species, growth medium, culture age and post-harvest treatment. This paper reports on the a b i l i t y of bacteria to de-emulsify a variety of complex wateri n - o i l petroleum f i e l d emulsions. MATERIALS AND METHODS Organisms: The two bacteria used in t h i s study were Nooardia opnarae LL Se6 ATCC 27808 and Corynebacterium petrophilum ATCC 21404. Previous studies (Cairns e t a l . , 1982, 1981; Cooper e t a l . , 1980 a, b) have shown these species to be e f f e c t i v e de-emulsifiers of simple, defined model emulsions. The cultures were maintained on Difco n u t r i e n t agar slants (Difco Laboratories, D e t r o i t , Michigan). Batch Cultures: Bacteria were grown ( 1 8 - 1 i t e r s ) for 20 days without pH control in a 2 4 - 1 i t e r fermentor (model No. CMF-128S; New Brunswick S c i e n t i f i c Co., New Brunswick, New Jersey) at 25 ° C, using an a i r flow of 15 l i t e r / m i n , and an a g i t a t i o n at 200 rpm. N. c~narae was grown on a 4% hexadecane (99% pure; Fisher S c i e n t i f i c Co., Toronto, Ontario) and 0.8% yeast e x t r a c t (BBL L t d . , a d i v i s i o n of Becton, Dickinson & Co., Cockeysville, MD) medium. C. petrophiZuJn was grown on a 2% glucose (Fisher S c i e n t i f i c Co., Toronto, Ontario) and 0.8% yeast extract medium. A l - l i t e r inoculum was used in both runs with a minimum of 4 g m / l i t e r . T w o - l i t e r samples were removed p e r i o d i c a l l y f o r analysis (biomass, pH, e t c . ) . Bacterial Growth Measurements: Biomass was determined by vacuum f i l t e r ing I0.0 ml of broth through dried and preweighed 0.45 ~m micropore f i l t e r s (Amicon Corp., Lexington, MA), drying the residue at 105 ° C f o r 24 hours and reweighing. Surface Tension: The surface tension of I0 ml aliquots of whole bacterial broth was monitored at room temperature in I00 ml acid-washed glass beakers using an Autotensiomat (Fisher S c i e n t i f i c Co., Toronto, Ontario). This is a modified 725

deNuoy surfa¢e tensiometer with a motorized sample stage and a s t r a i n gauge connected to a platinum ring. Surface tension was recorded d i r e c t l y on a linear chart recorder in mN-m-} Electrophoretic M o b i l i t y : The e l e c t r o p h o r e t i c m o b i l i t y of the bacterial cells was determined using a Zeta-Meter CZeta-Meter I n c . , New York, NY), set at 200 mV. Bacteria were centrifuged from the culture and suspended (130 mg/lO ml) in d i s t i l l e d water p r i o r to the determination. Bacteria-lnduced De-emulsification: Experiments c o r r e l a t i n g de-emulsif i c a t i o n with culture properties (Figures 1 and 2) were carried out using an o i l well-head w a t e r - i n - o i l (W/O) emulsion (#02-12-12: 20.5% water, 16.8% "clay" with a density of 1.310). "Clay" refers to the residual material l e f t a f t e r heating the emulsion at 600 ° C f o r 18-36 hours to obtain a constant weight. In these studies, I00 g of emulsion (wet weight) was added to 400 ml toluene ( a n a l y t i c a l grade: Fisher S c i e n t i f i c Co., Toronto, Ontario) and mixed using a magnetic s t i r r e r at ca. 200 rpm. Subsequent experiments used day 1 cells of N. amarae and C. p e t r o p h i l w n (Figures 3, 4 and 5) to examine the influence on de-emulsification of (a) type of de-emulsifier, (b) emulsion composition, and (c) the concentration of the de-emulsifier. In these l a t e r studies, 200 g of emulsion was mixed with 300 ml toluene. The bacteria (0.312 gm dry weight of cells in 20.0 ml 95% ethanol) were added, and mixing was continued f o r 2 rain. Aliquots (30.0 ml) of the mixed dispersion were placed into I00.0 ml glass beakers covered with parafilm and l e f t to stand. At various time i n t e r v a l s , the top I0.0 ml of the solution was pipetted o f f and the water content determined by Dean-Stark d i s t i l l a t i o n for a minimum of 3 hrs. to ensure q u a n t i t a t i v e determination of water. The extent of de-emulsification was indicated by the percentage of i n i t i a l amount of water remaining in the upper layer. Emulsion h a l f - l i f e (t½) was determined as outlined in Cooper e t a l . , 1980 a. Controls, using the above procedure without the addition of bacteria, were carried out. All controls showed no de-emulsification over the experimental time period. The c e l l - f r e e broth of bacterial cultures was also tested f o r de-emulsifying a c t i v i t y . Bacteria were removed from the whole culture using a Sorval RC-28 centrifuge (500 g for 20 min.) and 0.5 ml of the r e s u l t a n t supernatant ( c e l l - f r e e broth) was tested using the above methodology. Commercial De-emulsifiers: Two commercial de-emulsifiers ( T r e t o l i t e E-3453 and F-46) were obtained from the P e t r o l i t e Corp. of Canada Ltd., Calgary, Alberta. RESULTS Correlations were sought between the de-emulsifying a b i l i t y of cells or c e l l - f r e e broth and a number of properties of the cells or broth which were changing during batch c u l t u r i n g of the bacteria. The studied properties of cells included the growth phase (as r e f l e c t e d in biomass p r o f i l e s ) and the electrophoretic mobility. The studied properties of the broth included surface tension and pH. The culture age dependent changes in these properties and in the de-emulsifying a b i l i t i e s are shown f o r N. ornarae and C. p e t r o p h i l u m in Figures 1 and 2, respectively. More de-emulsification of the W/O petroleum f i e l d emulsion (#02-12-12) was achieved during a 24-hr. contact period with cells of N. amarae and C. p e t r o p h i l u m than with the c e l l - f r e e broths from which the cells were obtained. Cells from the e a r l i e r phases of growth (exponential and early s t a t i o n a r y phase) were more e f f e c t i v e than late s t a t i o n a r y phase c e l l s . The dependence of de-emulsification on concentration of bacterial cell de-emulsifiers was studied with day 1 cells of N. amarae and C. p e t r o p h i l u m (Fig. 3). Complete de-emulsification of the #02-12-12 emulsion was accomplished by 24 hrs. using I00 ppm of c e l l s . The time required f o r complete de-emulsif i c a t i o n was approximately inversely proportional to the concentration of c e l l s . 726

Several w a t e r - i n - o i l f i e l d emulsions (supplied by AOSTRA) were tested against day 1 cells of C. petrophiZuJn to determine the f l e x i b i l i t y of bacterial de-emulsifiers in coping w~th the v a r i a b i l i t y encountered in petroleum f i e l d emulsions. All of the emulsions tested could be de-emulsified by day 1 cells o f c. petrophil~n. The rates of de-emulsification varied depending on the emulsion. Attempts were made to correlate the physical c h a r a c t e r i s t i c s of the W/O f i e l d emulsions with t h e i r r e l a t i v e rates of d e - e m u l s i f i c a t i o n , using day 1 cells of C. petrophilum (500 ppm). Fig. 4a is a p l o t of the i n i t i a l % water in each of these f i e l d emulsions against the emulsion t½. Although emulsions with a high % water generally broke more r a p i d l y , there was no c o r r e l a t i o n of t½ with water content when the i n i t i a l % water was low. On the other hand, a more l i n e a r r e l a t i o n s h i p is suggested when the % clay in the original emulsion is plotted against the emulsion t½ (Fig. 4b). I t is seen from Fig. 4b that the rapid breaking, of the high-water content emulsions in Fig. 4b correlates with the low-clay content of these emulsions. Emulsion t½ in the presence of a given amount of b a c t e r i a l de-emulsifiers appeared best to correlate with the amount of clay in the emulsion (Fig. 4b). A bacterial de-emulsifier (day 1 cells of C. petrophilu3n) was compared to available commercial de-emulsifiers ( T r e t o l i t e F3453 and F-46). Of the two commercial de-emulsifiers, only T r e t o l i t e E-3453 had any a b i l i t y to break the W/O f i e l d emulsions. This d e - e m u l s i f i e r (500 ppm dry wt./volume basis) was compared to the day 1 cells of C. petrophilu~n (500 ppm dry wt./volume basis) using three f i e l d emulsions (Fig. 5). In all cases, the bacterial de-emulsifier was as e f f e c t i v e as the T r e t o l i t e E-3453, i f not b e t t e r ( i . e . , for I-JB-15-9). DISCUSSION This laboratory has now demonstrated that a number of bacterial species have the a b i l i t y to act as solid-phase de-emulsifiers of complex W/O petroleum f i e l d emulsions as well as simple, defined model emulsions (Cairns et a l . , 1982, 1981). In the present study, the a c t i v i t y of the microbial de-emulsifier was comparable to that of a petrochemical-based de-emulsifier. De-emulsification by the most active bacteria (N. c~narae and C. petrophilum) was associated with the cell surface. For W/O petroleum emulsions, cells from younger cultures were most e f f e c t i v e . For O/W model emulsions, N. oTnarae cells from older cultures were more e f f e c t i v e (Cairns et a l . , 1983). This l a t t e r observation was demonstrated as being due to the increasing hydrophobicity of cell surfaces which occurs with aging of nocardial cultures. Conceivably, the adhesion of an o i l droplet in an O/W emulsion to a bacterial surface, the spreading of the o i l phase on that surface, and the coalescence of that droplet with another s i m i l a r l y adhering and spreading o i l droplet is f a c i l i t a t e d i f the bacterial surface is hydrophobic. In W/O petroleum emulsions, the adhesion, spreading and coalescence of water droplets is therefore l i k e l y f a c i l i t a t e d by the h y d r o p h i l i c surfaces of cells from younger cultures. Preliminary evidence has been presented that microbial de-emulsifiers might be optimized f o r breaking of complex W/O petroleum emulsions by attention to such factors as cell surface hydrophobicity and cell surface charge. The commercial use of microbial de-emulsifiers which operate by any of the above mechanisms can be expected to become increasingly more favourable as (a) the costs of petrochemical-based de-emulsifiers increase f u r t h e r and (b) the costs of producing bio-de-emulsifiers is decreased by development of appropriate c u l t u r i n g techniques to produce b i o - d e - e m u l s i f i e r s by growing microbes on inexpensive a g r i c u l t u r a l , municipal or i n d u s t r i a l wastes.

ACKNOWLEDGEMENTS This work was supported by the Alberta Oil Sands Technology and Research A u t h o r i t y (AOSTRA) under AOSTRA Agreement #211.

727

REFERENCES Cairns, W.L., Cooper, D.G. and KosarCc, N. (1983). Bacteria-lnduced De-emulsification, In Microbial Enhanced Oil Recovery, eds. Zajic, J.E., Cooper, D.G., Jack, T. and Kosaric, N., pp. 106-113, Penn Well Publishing Company, Tulsa, OK, U.S.A. Cairns, W.L., Cooper, D.G., Zajic, J.E., Wood, J.M. and Kosaric, N. (1982). Characterization of N o o a r d i a c~narae as a potent biological coalescing agent of water-oil emulsions. Applied and Environmental Microbiology, 43: 362-366. Cooper, D.G., Zajic, J.E., Cairns, W.L. and Kosaric, N. (1980a). De-emulsification and Microbes. Biochemical Engineering Research Reports, Vol. VI. The University of Western Ontario, London, Canada. Cooper, D.G., Zajic, J.E., Cairns, W.L. and Kosaric, N. (1980b). De-emulsification and Microbes. Part I I . Biochemical Engineering Research Reports, Vol. VIII. The University of Western Ontario, London, Canada. Kosaric, N., Gray, N.C.C., Stewart, A.L. and Cairns, W.L. (1981). BacteriaInduced De-emulsification of Complex Petroleum Emulsions. Part I. Biochemical Engineering Research Reports, Vol. IX. The University of Western Ontario, London, Canada.

79A

t-cOnl~ ,~ c- q-- .I-I

!

I

i"

I

•

l

!

<1

I

I> *'~ ~ . , - -

0 t..~ . r ,~

~'l

m

I...)

-t.J r - -

"-I-.5

r"-

~-,o . ~

><

~ j . ~ --

U

4-~ rr~ ~ - ~

,

.[

\

• .

,I

I'O " ~

g

@

~

R

I

I

.~-

c-

c"

~1

O,J

0

.X=

OJ I

H d

I

(,.i w

0,.; ~ ,e4-.~ CO ~ I.~ ~ZI • t-" r---

/ @

I

0

a

/

o -I-~ ~>~

I,

=

I

. ~tu) S S V P J O I 8

"

•

g

1..335

• S

~

)

m

-~

v

ol

I

'

I

'= i

i"

el

i

i

• "i

I

. =

I

<1,

1

i

I

•

8

I

~- -o ""~ 4-) 4-~ L } tO

I>

X

c'4-> ro .,~

c-

0..)

.< e

~a

IX:?,---

Or'-

o

i

// ;'

(1

/

,,°

I

i

I

" O . ~

~ Y ] ~ a a a n i~'ommvic-aa ~l~zv~ 40

• Hd

-

I

I

( I.|i&l

. 8~I)

I

-r'.~

0

I

,

I

(:$Yl,~O|i

I

,

i.l~

1

I

/

u~

"~

¢:

p-

E

4J

~

~

c" .r--°r-

~ ~I c-

•

I

0

4J

1

--

ro

•~

~ •

.l'lOh

•,"--

X nr- ,~. (iI ~I_ 0 c" c*"

c"

¢~ 4-~ " r I=:: ~.) m

g ~1~:

I::T~ C> .-.-.~ ~

1.~_ I l ; ~ v v

l

729

ID'I ~-

I -[~

I

~u ]00

~Jk

I

I

I

(A)

* ~ . ~0

z 0 D

BO

l,.i

L~ II

7

.2

|

o "

•

_z

6o p.

30

\ Ua I--

\

40 I

Io

<

\

• \ \

5

2g

10

]2 ~.

9

.13

7 4

16

12

8

CONTACT

TIME

2/}

24

I

(B)

1

Z

(HP3)

2 ~

13 4

~mroph~

N. am ~-a e

/

Z

•

20 ppm

0

I 0 0 ppm

&

2D0 ppm

•

5 0 0 ppm

u

1 0 0 0 ppm

A Z~

]2

500 ppm < .-a u

| 0 0 0 ppm

2

3

4

./

Q

i

i

i

i

100

LE;END EMULSION

C., pc| rophilLU~

E-3453

I-JB-I5-7

•

o

I-JB-IS--8

•

u

I-JB-I$-9

80

•

8

*

oil

Figure 3" The d e - e m u l s i f y i n g a b i l i t y of d i f f e r e n t c o n c e n t r a t i o n s of day 1 c e l l s from 1 8 - 1 i t e r batch fermentations of N. c~no_~ae and C. p e t r o p h i l u m against the w e l l - h e a d emulsion #02-12-12

90

~"

/

/

!

I

I

2

4

6 t~

i

I

•

i

8

IO

( H ILS )

Figure 4: Emulsion h a l f - l i f e (b~) in the presence of 500 ppm, day l c e l l s of C. p e t r o p h i l ~ n as a f u n c t i o n of (A) the i n i t i a l amount of water and (B) the amount of c l a y , in Esso well-head emulsions ( 1 , 2 , 3 , 5 , 9 - 1 2 ) , Athabasca steam flood emulsions (4,6) and Lloydminster steam d r i v e emulsions (7,8,13).

A

)n,

70

L~

~c _z

60

50

40

I\. \ \ \ "~-A . \ ~~._

3(]

~\\

110

\

"A

\

•

"~

~

"~

"~Z~

~

20

10

" rv'~rAPT

T I L ~ 114Re&

:o

12

730

Figure 5: A comparison of the deemulsifying a b i l i t i e s of C. petrophiZwn (day 1 c e l l s ; 500 ppm) and the commercial de-em ul si fi er E-3453 (500 ppm) against Lloydminster steam drive ( I - J B - 1 5 - 7 ) and combustion ( I - J B - 1 5 - 8 and I - j B - 1 5 - 9 ) W/O emulsions.