COLLOIDS

AND
A
Colloids and Surfaces SURFACES
ELSEVIER A: Physicochemicaland Engineering Aspects97 (1995) 119 128

Water-in-crude oil emulsions from the Norwegian continental shelf

11. Ageing of crude oils and its influence on the emulsion stability
Hans Petter Ronningsen a,,, Johan Sj6blom b, Li Mingyuan b
a Reservoir Technology, Fluid laboratory, Statoil A/S, Forus, N-4001 Stavanger, Norway
b Department of Chemistry, University of Bergen, N-5007 Bergen, Norway
Received 27 May 1994; accepted 11 December 1994

Abstract

The stability of water-in-oil emulsions based on some North Sea crude oils has been studied with particular
emphasis on the effects of ageing, i.e. exposure of the crude oils to air (and light). It is clearly demonstrated that the
interfacial tension of a crude oil towards formation water decreases substantially as a result of ageing. This seems to
be caused by the formation of various oxidation products, mainly carbonyl compounds. In addition, the water-in-oil
emulsions in general become more stable. However, in some cases a reversal of this trend has been observed, indicating
that oxidation products still with high interfacial activity but having less beneficial film properties are formed in the
later stages of the ageing process.

Keywords: Ageing; Emulsion stability; Water-in-crude oil emulsions

1. Introduction with its content of indigenous polar compounds,

found among the resins and asphaltenes, is well
It is frequently observed in studies of the forma- documented [1]. With this background it would
tion and breaking of water-in-crude oil emulsions not be surprising that processes which change the
that the results may be time-dependent, in the polarity of a crude oil may change its ability to
sense that the measured emulsion stability depends form emulsions.
on the age of the crude oil sample, i.e. the extent Possible changes that may occur during ageing
to which it has been subjected to air and light of a stabilized (atmospheric) crude oil exposed to
exposure. This may apply not only on a long time- air and light, are; (1) loss of light ends (evapora-
scale (months), but also on a short time-scale tion); (2) precipitation of heavy components such
(hours), and may thus introduce errors and poor as wax and asphaltenes (these processes occur, of
reproducibility even within a single series of succes- course, even without exposure to air and light);
sive measurements. Consequently, the direct rele- and (3) chemical alteration (oxidation/photolysis).
vance and reliability of such tests for process design Such changes may in turn lead to changes in
purposes may be highly questionable.
physical properties such as density, viscosity, sur-
The idea that the tendency of a crude oil to
face/interfacial tension and dissolving capacity.
form stable water-in-oil emulsions is correlated
Finally, the stability of water-in-crude oil emul-
sions depends to some extent on each of these
* Corresponding author. properties of the oil phase. Processes like those

0927-7757/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved

SSDI 0927-7757(94)03039-1
120 tL P. Ronningsen et al./Colloids Surfaces A: Physicochem. Eng. A,spects 97 (1995) 119 128

mentioned above have been described in connec- Some oils were also stored under nitrogen for
tion with studies of the behaviour of oil-spill reference.
on the sea, then called weathering processes.
Evaporation is then most often the dominating
effect, but photo-oxidation has also been shown to 3. Experimental
contribute [2].
To our knowledge, ageing effects occurring in For the mixing of synthetic water-in-oil emul-
crude oils and their influence on the emulsion sions, a Buhler HO4 propeller-type homogenizer
formation tendency or other crude oil properties was used. It was operated at 8000 rev min-1 for
have not been studied systematically before; at 1 rain. The sample volume was 11 ml. The water
least, only a few published results appear to exist. cut was 50% and the temperature was 50 or 70 °C.
Tort and Andersen [3] studied interracial behavi- These mixing conditions have previously been
our and chemical changes taking place in an shown by Johansen et al. [5] to create droplet size
asphalt sample subjected to progressively stronger distributions that are fairly realistic compared to
exposure to oxygen. In parallel with increased that of naturally occurring emulsions, i.e. giving
levels of oxidation, they observed enhanced inter- mean droplet sizes of around 10 p,m with a stan-
facial activity. In fact, an increased content of dard deviation of about 5 ~tm. The emulsions were
asphaltenes was observed as well. These experi- transferred from the mixing vessel to a graduated
ments are, to some extent, similar to those con- cylinder and were kept in a thermostatted bath at
ducted by Sj6blom et al. [4] with interfacially the specified test temperature for an observation
active fractions separated from some of the same period of at least 30 min, during which the water
North Sea crude oils that are used in the present break-out (free water as a percentage of the original
paper. water) was recorded.
With this background, a project was initiated in In tests conducted with demulsifier, this was
order to study the changes that occur in some added as a 0.1% solution in xylene by shaking for
North Sea crude oils as a result of exposure to air 30 s immediately after mixing of the emulsion.
and light, and their implications for the examina- The total acid numbers of the crude oils were
tion of emulsion formation and stability. measured according to ASTM D-664/89.
The density was measured at 15°C using a
frequency density meter ( D M A 4 0 with a
DMA 602M measuring cell; Anton Paar).
The kinematic viscosity was measured at 50°C
2. Samples using Ubbelohde glass capillary tubes.
The dynamic viscosity was measured using a
North Sea crude oils from five different fields, Haake RV12 concentric cylinder viscometer equ-
all sampled at the separator outlet during the ipped with an NV double-gap measuring system.
testing of exploration wells and stored in 201 The cooling rate was 12°C h -1.
pressurized bottles, have been used. The bottles The surface tension and interfacial tension
were heated to 60°C overnight in the laboratory, towards synthetic formation water were measured
whereafter the oils were depressurized in one stage at 50°C using a KSV Sigma 70 interfacial tensiom-
to atmospheric pressure. The oils had been stored eter (Du Noi~ying method).
from 1 month to 3 years in the pressurized bottles The IR spectra of whole crude oils were recorded
prior to testing. The first test (i.e. no ageing) was using a Perkin-Elmer 1720 X Fouriertransform-
started as soon as practically possible after depres- infrared spectrophotometer.
surization, i.e. within a few minutes. All crude oil The total wax content was measured by a slightly
samples were aged at 50°C in transparent or dark modified U.O.P. method 46 64, the so-called ace-
bottles with air naturally available, i.e. the exposure tone precipitation technique (at - 2 5 °C) described
to air was not forced or accelerated in any way. by Burger et al. [6].
 H.P. Ronningsen et aL/Colloids Surfaces A: Physicochem. Eng. Aspects 97 (1995) 119-128 121

The asphaltene content was determined by pre- from pressurized bottles) exhibited the slowest
cipitation with pentane at room temperature. dehydration; no observable water break-out took
The wax formation temperature was determined place within 24h without use of demulsifier
by polarization microscopy. (Fig. l(a)). These were rather waxy oils (particu-
More details about methods for determination larly oil E) with the highest asphaltene contents,
of the wax content, asphaltene content, wax forma- i.e. 1.2-1.7 wt.%. However, oil B, which was a
tion temperature, viscosity, pour point and density waxy oil with very low asphaltene content, also
are found in the paper of Ronningsen et al. [7]. formed a very stable emulsion (only 20% of the
More details about interfacial tension measure- original water breaking out within 24 h). Like oils
ments and IR spectra are given by Sj6blom et al. D and E, oil B had a very low acid number, i.e. a
E43. low content of organic acids. It should be empha-
sized that the tests were conducted at a temperature
far above the wax formation temperature; thus no
4. Results contribution from wax particle stabilization is
likely. Neither was the viscosity of these oils high;
4.1. Emulsion stability tests --fresh oils (no at 50°C it was in fact the lowest of all the oils
ageing) investigated, and the interfacial tension towards
formation water was not particularly low.
Some relevant chemical and physical data of the Fresh oil C also exhibited slow dehydration
crude oils investigated are summarized in Table 1. during the first 30 rain after emulsion formation,
Emulsions based on fresh oils D and E (directly but, in fact, complete separation took place within

Table 1
Chemical and physical data of the fresh North Sea crude oils investigated

Characteristic Oil A Oil B Oil C Oil D Oil E

Molecular weight 244 241 233 192 263

Density at 15°C (kg m 3) 883.3 856.3 887.0 847.2 865.7
Clo + (wt.%) 91.20 89.12 88.44 79.03 90.15
C10 + aromatics (wt.%) 42.4 29.0 41.3 36.2 28.8
Total wax (wt.%) 2.2 14.8 6.7 11.4 15.6
Asphaltenesa (wt.%) 0.3 0.27 0.50 1.7 1.2
Sulphur (wt.%) 0.42 0.1 0.48 0.20
Total acid number (mg K O H g-1) 0.22 0.07 2.60 0.07 0.09
Basic nitrogen (wt.%) 0.031 0.014 0.023 0.031 0.013
Wax formation temp. (°C) 30 40 41 42 40
Wax dissolution temp. (°C) 50 53 55 - 50
M i n i m u m pour point (°C) -45 13 -48 12 16
Surface tension b (raN m - 1) 27.3 26.3 27.7 25.7 28.4
Interracial tension bc (mN m - 1) 28.8 26.7 24.6 22.3 26.1
Kinematic viscosity (cSt)
At 80~C 3.65 2.86 3.31 2.40 3.84
At 5 0 ' C 7.17 5.12 6.40 3.30 7.07
Dynamic viscosityd (mPa s)
At 20°C 16.9 46.7 29.1 19.5 155
At 5~C 41.1 295 56 78.6 900

a Pentane insolubles.
b At 50 °C.
c Towards synthetic formation water.
d At shear rate 100 s i.
122 H.P. Ronningsen et al./Colloids Surfaces A: Physicochem. Eng. Aspects 97 (1995) 119 128

100 4.2. Emulsion stability tests - - aged oils

l Oils D and E: No separation
"O
= 8oJ
; J A significant increase of e m u l s i o n stability was
~. 6o o b s e r v e d a l r e a d y after 24 h e x p o s u r e of oils A
; •• Oil
o,, CB
AJ (Table 2) a n d B (Table 3) to air (and light). This
~ 40, [ ] ~i'°il ! trend c o n t i n u e d for a b o u t 1 m o n t h . Thereafter, a
reversal of the trend t o w a r d s m o r e stable emulsions
o~ 202
t o o k place. F o r oil B the same t r e n d was also
o b s e r v e d with e x p o s u r e to air only (i.e. in a d a r k
1 10 100 1000 10000 bottle) (Table 4). F o r oil C (Tables 5 a n d 6) the
(a) Time (min) emulsions were originally very stable a n d no
1oo ~ o b s e r v a b l e change seemed to take place in the first
weeks. F r o m a b o u t 3 m o n t h s ( p e r h a p s s o m e w h a t
'o 80] earlier) there was a substantial change in the
• Oii direction of less stable emulsions, a trend that
~. 60 /J I eoil B
c o n t i n u e d at 6 m o n t h s as well. I n d i c a t i o n s of the
ffl ~0, C
~ 40' ~0iI D same k i n d of reversal has been observed with
~o E m o d e l emulsions stabilized with aged interfacially
o~ 201 active fractions from oil C [ 4 ] .
Even m o r e p r o n o u n c e d effects of ageing were
1 10 1oo 1ooo 1o0oo observed with oils D a n d E. Some emulsion sta-
(b) Time (min) bility d a t a for oil D are s u m m a r i z e d in Fig. 2. This
crude, which was rather asphaltenic (1.7 wt.%
Fig. 1. Separation of water as a function of time from w/o
emulsions based on fresh North Sea crude oils: a) Without p e n t a n e insolubles) formed extremely stable emul-
demulsifier and b) With 10 ppm demulsifier. sions. All e x p e r i m e n t s therefore h a d to be per-
formed at higher t e m p e r a t u r e (80 ° C) a n d b y using
24 h (Fig. l(a)). This oil h a d r e m a r k a b l y high acid demulsifier to b r e a k the emulsions. It is seen in
n u m b e r a n d a m o d e r a t e a s p h a l t e n e content. Fresh Fig. 2 that emulsions based on fresh oil (directly
oil A, which was a low-wax, l o w - a s p h a l t e n e b i o d e - from a pressurized bottle) were quite easy to b r e a k
g r a d e d oil, showed the fastest s e p a r a t i o n with 60% with demulsifier (94% within 6 mins). A l r e a d y after
of the original w a t e r b r e a k i n g out within 30 min. 3 h e x p o s u r e to air a n d light (at 50°C), a substan-

Table 2
Emulsion stability data for oil A (aged with air and light at 50°C) 50% water

Ageing Free water separated (%)

time
After 1 min After 3 min After 6 min After 10 min After 15 min After 20 min After 30 min After 24hrs

0 11 (16) 20 (90) 38 (94) 40 (95) 45 (95) 48 (95) 60 (95) G)

2h (8) (88) (88) (90) (90) (90) (90) ()
24 h 1 (10) 8 (92) 14 (94) 14 (94) 16 (94) 20 (94) 30 (96} (-)
1 week (14) (90) (92) (92) (94) (94) (95) - (-)
1 month 0 (5) 0 (54) 0 (92) 0 (92) 0 (94) 0 (94) 1 (94) 42 (98)
1 month" 1 (24) 2 (92) 3 (96) 5 (96) 8 (96) 12 (98) 16 (98) 76 (98)
3months 0 (34) 0 (80) 2 (90) 8 (92) 24 (92) 32 (94) 60 (94) 86 (98)
3months a 0 0 0 2 3 4 4 46

Results obtained with 10 ppm demulsifier are given in parentheses.

a Crude sample stored under nitrogen in the dark at 50°C.
 H.P. Ronningsen et al./Colloids Surfaces A: Physicochem. Eng. Aspects 97 (1995) 119 128 123

Table 3
Emulsion stability data for oil B (aged with air and light at 50°C) 50% water

Ageing Free water separated (%)

time
After 1 min After 3 min After 6 min After 10 min After 15 min After 20 min After 30 min After 24 h

0 1 (6) 2 (46) 4 (98) 6 (98) 8 (98) 8 (98) 12 (98) 22 (100)

2h 0 (0) 1 (18) 2 (90) 3 (98) 3 (100) 4 (100) 6 (100) 14 (100)
24 h 0 (0) 0 (68) 1 (98) 1 (100) 1 (100) 1 (100) 1 (100) (100)
1 week 0(0) 0 (10) 0 (32) 0 (68) 0 (90) 0 (94) 0 (98) 1(100)
1 month 0 (0) 0 (2) 0 (60) 0 (90) 0 (98) 0 (100) 0 (100) 0 (100)
3months 0 (0) 0 (8) 0 (16) 0 (56) 1 (84) 1 (86) 2 (90) 24 (100)

Results obtained with 10 ppm demulsifier are given in parentheses.

Table 4
Emulsion stability data for oil B (aged with air (dark) at 50°C)-50% water

Ageing Free water separated (%)

time
After 1 min After 3 min After 6 min After 10 min After 15 min After 20 min After 30 min After 24 h

0 1 (6) 2 (46) 4 (98) 6 (98) 8 (98) 8 (98) 12 (98) 22 (100)

2h . . . . .
24 h 0 (0) 0 (40) 0 (76) 0 (98) 0 (98) 0 (98) 0 (98) 0 (100)
1 week 0 (0) 0 (60) 0 (96) 0 (96) 0 (96) 0 (98) 0 (98) 0 (100)
1 month 0 (0) 0 (10) 0 (90) 0 (98) 0 (98) 0 (98) 0 (100) 0 (100)
3 months 0 (0) 0 (0) 0 (94) 1 (96) 1 (96) 2 (96) 14 (96) 32 (96)

Results obtained with 10 ppm demulsifier are given in parentheses.

Table 5
Emulsion stability data for oil C (aged with air and light at 50°C)-50% water

Ageing Free water separated (%)

time
After 1 min After 3 min After 6 min After 10 min After 15 min After 20 min After 30 min After 24 h

0 0 (44) 2 (98) 0 (98) 0 (100) 0 (100) 0 (100) 0 (100) 100

2h 0 (32) 0 (98) 0 (98) 0 (98) 0 (98) 0 (98) 0 (100) 100
24 h 0 (28) 0 (96) 0 (98) 0 (100) 0 (100) 2 (100) 15 (100) 96
1 week 0 (58) 0 (92) 0 (96) 0 (96) 0 (98) 0 (98) 5 (100) 98
1 month (3) (90) (92) (94) (94) (94) (100)
3months 0 (6) 2 (84) 16 (88) 70 (90) 90 (90) 90 (90) 90 (98) 100
3months 0 (12) 0 (92) 0 (94) 0 (96) 0 (96j 0 (96) 0 (96) (98)
6months 0 2 16 60 90 90 90 94

Results obtained with 10 ppm demulsifier are given in parentheses.

tial i n c r e a s e o f s t a b i l i t y w a s o b s e r v e d , a n d a f t e r 1 months ageing, the behaviour had nearly returned

month no water break-out occurred within the to the original!
first 2 h a f t e r m i x i n g . A f t e r 7 m o n t h s a g e i n g , a E m u l s i o n s b a s e d o n t h e w a x y oil E w e r e h a r d
r e v e r s a l o f t h e t r e n d h a d t a k e n p l a c e a n d a f t e r 10 t o b r e a k e v e n w i t h d e m u l s i f i e r , a n d as s e e n i n
124 H.P. Ronningsen et al./Colloids Surfaces A." Physicochem. Eng. Aspects 97 (1995) 119 128

Table 6
Emulsion stability data for oil C (aged with air (dark) at 50°C) 50% water

Ageing Free water separated (%)

time
After 1 min After 3 min After 6 min After 10 min After 15 min After 20 min After 30 min After 24 h

0 0 (44) 0 (98) 0 (98) 0 (100) 0 (100) 0 (100) 0 (100) 100 (100)

2h . . . . .
24 h (8) (88) (88) (90) (90) (90) (92)
1 week (1) (90) (94) (96) (98) (98) (100)
1 month 1 (1) (86} (88) (88) (90) (90) (92)
3months 0 (6) 5 (88) 14 (96) 90 (96) 92 (96) 92 (96) 92 (98) 98
6months (1) 2 (80) 18 (92) 76 (92) 96 (94) 98 (94) 100 (96) 100 (96)

Results obtained with 10 ppm demulsifier are given in parentheses.

100
Ageing 1 ON . . . .
time: Ageing time:
'10 80
IINo "o 8 •
,= ! 0'3 hrs BBNo
60 i AZ4hrs
U) I O 1 week ~ 6 ' : '2:rhrs
40 @ 1 month • 1 week
t~ -~ 7 months 4 ~ 1 month
20 10 months I ~ 3 months
o~ 2
i V 6 months
0 [ )~t 9 months
0 10 20 30
1 10 100 1000
Separation time (min) Separation time (min)
Fig. 2. Effect of ageing with air and light at 50°C on water Fig. 3. Effect of ageing with air and light at 50°C on water
separation from emulsions based on crude oil D. Water cut; separation from emulsions based on crude oil E. Water cut;
50%; temperature; 80°C; demulsifier; 10 ppm NH-601 IX. 50%; temperature; 70°C; demulsifier; 10 ppm NH-6011X.

Fig. 3, a gradual change towards more stable emul-

sions took place. After 6 months, no water break- 100 i
Ageing time:
out was achieved within 24 h after mixing. So far, "o 80
there was no evidence of any reversal of the trend
P F INo
for this oil. ~. 60!
• 24 hrs
For oils D and E a very slight increase in i A 11 days
stability was observed even when the oils had been 40
~ • 40 days
stored under nitrogen (for several months). This is
20
illustrated for oil E in Fig. 4. For oil A as well,
slightly enhanced emulsion stability was evident 0
when the crude oil was stored in the dark under a 0 10 20 30
nitrogen atmosphere (Table 2). The same reversal Separation time (min)
after about 3 months ageing, as in the case with Fig. 4. Effect of ageing in nitrogen (in the dark) at 50°C on
exposure to air, is n o t seen to occur. This indicates water separation from emulsions based on crude oil E. Water
that although a gradual change, in fact, may occur cut; 50%; temperature; 70°C; Demulsifier; 10 ppm NH-601 IX.
 H.P. Ronningsen et al./Colloids Surfaces A." Physicochem. Eng. Aspects 97 (1995) 119-128 125

even in a nitrogen atmosphere, p r o b a b l y owing to 4.4. Density and viscosity

small a m o u n t s of oxygen dissolved in the oil during
depressurization, the ageing process is slowed For all the oils, a slight increase in density and
down, as one would expect. viscosity was observed during 3 m o n t h s ageing at
50 ° C. D a t a for crude oil A are shown in Table 7
4.3. Interracial tension as a typical example. This increase can mainly be
attributed to a certain (though small) loss of light
The interfacial tension (at 50 ° C) between crude components. The density in general increased by
oil A and synthetic formation water was measured 0.2-0.4% and the viscosity by 2-4%. These changes
regularly for a period of 50 days after depressuriza- are too small to have any significant effect on the
tion of the live separator oil. The results are emulsion stabilities.
summarized in Fig. 5. A more or less linear reduc-
tion from about 28 to about 1 6 m N m -~ was 4.5. Acid number
observed. There is a tendency of the curve to level
out at a value of about 15 m N m -1. N o significant In Table 1 it can be noted that oil C had a very
effect of using U V instead of ordinary light was high acid number, 2.60, i.e. a b o u t ten times higher
observed in this case. than n o r m a l for N o r t h Sea crude oils. This is
nicely correlated with a dominating carbonyl peak
a r o u n d 1725 cm -~ in the infrared spectrum of the
interfacially active fractions (Fig. 4 in Ref. 4), com-
3O pared to oils A and B (particularly B), which had
Temperature: 50 °C low acid numbers. Also oils D and E had very low
+uv acid numbers.
"
0
25 I
I As seen in Table 8, no significant increase in the
= -UV
-I total acid n u m b e r of oils A and C was observed
as a result of exposure to air and light for 3
months, only a very slight increase for oil C, almost
within experimental error.
C

-- 15 .... i L i i i i i

0 10 20 30 40 50 60 5. D i s c u s s i o n
Storage time (days)

Fig. 5. Interfacial tension between crude oil A and synthetic Low interfacial tension towards the water phase
formation water as a function of ageing time at 50°C. is beneficial with regard to the formation of small

Table 7
Density (15°C) and kinematic viscosity (50°C) vs. ageing time (at 50°C) for oil A

Ageing Ageing with air and light Ageing with air (dark) Ageing under nitrogen (dark)
time
Density Viscosity Density Viscosity Density Viscosity

0 881.6 6.71 881.6 6.71 882.5 6.68

2h 881.9 6.73 - -
24 h 883.2 6.91 881.7 6.68 883.6 6.72
I week 884.0 6.83 883.8 6.74 883.4 6.67
1 month 882.7 6.78 882.7 6.78 882.9 6.78
3 months 882.7 6.80 883.1 6.85 883.5 6.88
8 months 883.0 6.98 6.96 -
126 H.P. Ronningsen et al./Colloids Surjaces A. Physicochem. Eng. A~pects 97 (1995) 119-128

Table 8 pounds, ranging from relatively strong carboxylic

Total acid number (mg KOH per gram of oil) of oils A and C
acids to weak phenolic and naphthenic acids [ 12].
as a function of ageing in air and light
The total acid number of a crude oil is a measure
Ageing time Aged in air and light Aged in air (dark} of the overall content of acidic compounds and is
determined by alkali titration. Such compounds
Oil A Oil C Oil A Oil C are known to exhibit interfacial activity [13] and
the acid number could thus be suspected to give
No ageing 0.25 2.51 0.25 2.51
24 h 0.24 2.55 0.24 2.50 an indication of the tendency of a crude oil to
1 week 0.22 2.56 0.22 2.55 form stable water-in-oil emulsions, like for instance
1 month 0.22 2.52 0.22 2.59 the polarity index proposed by Brianing [14].
3 months 2.61 2.64 Moreover, the acid number might be used as an
indicator of oxidative changes taking place during
the ageing of a crude oil.
water droplets in the continuous oil phase. Oils B, D and E, which had the lowest acid
However, low interfacial tension itself does not numbers, formed equally stable or more stable
ensure stable emulsions. The main mechanism emulsions than oil C. Other examples from our
behind the stabilization of water droplets in a crude laboratory also indicate that oils with low acid
oil is believed to be the formation of rigid interfacial numbers form very stable emulsions. We have seen
films (see Refs. [ 8 - 1 0 ] and others). In addition, no examples, so far, of high-acid-number crudes
particle and polymer (steric) stabilization may con- that do not form stable emulsions. No definite
tribute. Hence, interfacially active components pre- correlation can thus be identified between total
sent in a crude oil must also possess beneficial film acid number and tendency of a crude oil to form
properties in order to promote stabilization of stable emulsions. It appears that high-acid-number
water-in-oil emulsions. Asphaltenes are known to crudes, in general, form rather stable emulsions,
possess both these properties and a correlation is whereas low-acid-number crudes may or may not
thus normally seen between the asphaltene content form stable emulsions. It was surprising to notice
and the tendency of a crude oil to form stable that no increase in acid number took place as a
emulsions [ 1,11 ]. result of ageing, even though it was clearly demon-
The rather high asphaltene content of oils D strated by Sj6blom et al. [4] that ageing of inter-
and E (compared to most North Sea oils) may facially active fractions was accompanied by a
thus partly explain the stable emulsions formed by substantial increase in the same carbonyl peak in
these oils. However, like oil B, which also formed the IR spectrum that distinguishes oil C from the
stable emulsions, oil D and E were waxy (wax other oils. However, since carboxylic acids and
contents between 11.4 and 15.6 wt.%). The total other carbonyl compounds (ketones, aldehydes,
wax content has previously been identified (to- etc.) are not very well separated in the IR spectrum,
gether with the oil viscosity) by Johansen et al. this fact may be taken as an indication that the
[5] as an important parameter for the prediction main oxidation products formed as a result of
of emulsion stability. Waxes may promote stabili- ageing, although being highly interfacially active,
zation either as particles, i.e. below the wax forma- are not of the acid type.
tion temperature, or on the molecular level as The reduction of interfacial tension between
lipids. The experiments in this study were con- oil A and formation water is not very surprising
ducted at 50°C which is about 10°C above the in view of the results of Sj6blom et al. [4] with a
wax formation temperatures of the oils. Hence, model system with interfacially active fractions
although the presence of submicron wax crystals added. They found even more pronounced reduc-
cannot be completely excluded, the molecular effect tions of interracial tensions with ageing of the
of the waxes is probably the most important in fractions. The trend seen in Figure 5, is likely to
these cases. be representative for most crude oils. This means
Crude oils are known to contain acidic corn- that in the study of phenomena that depend on
 H.P. Ronningsen et al./Colloids Surfaces A: Physicochem. Eng. Aspects 97 (1995) 119-128 127

the interfacial tension, such as emulsion formation, duces less stable emulsions than corresponding oil
aged crude oil samples may give a highly erroneous obtained from an ordinary bottom-hole sample
impression of the real properties of that fresh (which had been below its bubble point and repres-
crude oil. surized into the single-phase state) and separator
The clear demonstrations of ageing effects shown samples, each collected and stored under pressure.
in this study mean that oxidation products formed This could be related to differences in the state of
as a result of the ageing of crude oil may not only dissolved asphaltenes, and is also a factor that has
give rise to enhanced interfacial activity and thus to be considered seriously. This will be examined
easier droplet formation. They also are able to in more detail together with possible ageing effects
promote the long-term stabilization of dispersed taking place immediately after depressurization of
water droplets by the mechanisms mentioned live reservoir oil, and published in a subsequent
previously. paper.
The mechanism behind the reversal of the trend
towards more stable emulsions with ageing
observed in several cases is not understood in 6. Conclusions
detail yet. Sj/Sblom et al. [4] found no such trend
in the interfacial activity of the pure oxidized The stability of water-in-oil emulsions based on
fractions. It is well known that surface active North Sea crudes appears to be mainly correlated
compounds added to a surfactant solution can with the wax and asphaltene content, in agreement
have a film-stabilizing effect when present in low with previous studies. There appears to be no
concentrations. At higher concentrations they can definite correspondence between the total acid
have a reverse effect. A possible explanation could number and the tendency of a crude oil to form
be that even though more interfacially active com- stable emulsions.
pounds are formed all the time, the late oxidation Crude oils are subjected to chemical alteration
products have less beneficial film properties, possi- as a result of exposure to air. In particular, car-
bly due to "wrong" molecular weights. Hence, bonyl compounds and certain condensation pro-
these compounds, due to their high interfacial ducts have been identified. Even when stored under
activity, tend to replace other compounds present nitrogen, gradual changes seem to occur in the
at the oil-water interface and thereby creating a direction of more stable emulsions, probably due
less stable film. to small amounts of dissolved oxygen. When
In many (perhaps most) cases, emulsion stability exposed to air for several months, a reversal of the
will be overestimated using bottle tests conducted trend towards more stable emulsions has been
with aged crude oil samples. However, owing to observed in several cases. The reason for this
the possible reversal, such tests need not necessarily behaviour is not clearly understood, but may be
be conservative. Whether they are, may depend on related to differences in the film properties of
the particular oil. The only way to obtain reliable oxidation products formed in earlier and later
results is to conduct tests on fresh crude oil samples stages of the ageing process. Ageing effects tend to
as soon as possible after the flash of pressurized oil. be less pronounced in tests where demulsifier is
One should be aware of the possibility that not used to break the emulsions. This is actually reas-
only the time a crude oil has been in contact with suring, since most emulsion tests in practice will
air, but also its total age (including storage time be conducted with demulsifiers. There are, how-
in pressurized bottles) and the location of the ever, examples of substantial ageing effects also in
sampling point during well testing (i.e. bottom experiments in which demulsifier has been used.
hole, wellhead, separator or stock tank) might Slight increases in the density and viscosity of
influence the results of the emulsion formation and the oils as a result of light end loss have been
stability tests. We have, in fact, observed indic- observed. These changes are far too small, however,
ations that crude oil obtained from a single-phase to influence the emulsion stability to any great
bottom-hole sample from a North Sea field pro- extent.
128 H.P. Ronningsen et al./Colloids Surfaces A. Physicochem. Eng. Aspects 97 (1995) 119-128

The results emphasize the need to use oil samples [2] P.S. Daling, P.J. Brandvik, D. Mackay and O. Johansen,
as fresh as possible when evaluating the emulsion Oil Chemi. Pollut., 7 (1990) 199.
[3] F. Tort and S.I. Andersen, Proc. 1st World Congr. on
formation tendency of a crude oil and when evalu- Emulsions, Paris, 1993, Paper 1-22-194.
ating the performance of chemical demulsifiers. [4] J. Sj6blom, L. Mingyuan, A.A. Christy and H.P.
Preferably, tests should be conducted in such a Ronningsen, Colloids Surfaces A: Physiochem. Eng.
way that contact with air is completely avoided Aspects, COL 3038.
or, at least kept to a minimum. [5] E.J. Johansen, I.M. Skj~ervo, T. Lund, J. Sj6blom,
H. Soderlund and G. Bostrom, Colloids Surfaces, 34
(1988/89) 353.
I-6] E.D. Burger, T.K. Perkins and J.H. Striegler, J. Pet.
Acknowledgements Technol, 33 (1981) 1075.
[7] H.P. Ronningsen, B. Bjorndak A.B. Hansen and W.B.
Pedersen, Energy Fuels, 5(6)(1991) 895.
Technical assistance by Brit Bjorndal and
[8] C.M. Blair, Chem. Ind. (London), (1960) 538.
Gunbjorg Fagerland, both of Statoil RESTEK- [9] J.E. Strassner, J. Pet. Technol., 20 (1968) 303.
Fluidlab, with the emulsion stability tests, and the [101 Berridge, S.A., Thew, M.T. and A.G. Loriston-Clarke,
density and viscosity measurements, and by J. Inst. Pet., London, 54(539) (1968) 333.
Kjartan Sletten, Statoil SFTK at Mongstad, with [11] J. Sj6blom, Li Mingyuan, A.A. Christy and T. Gu,
acid number measurements, is gratefully Colloids Surfaces, 66 (1992) 55.
[12] R.J. Parker and E.S.N. Chung, J. Can. Pet. Technol.,
acknowledged. 25(4) (1986) 72.
[13] V. Hornhof, G.H. Neale and A. Yu, J. Pet. Sci. Eng., 10
(1994) 291.
References [14] I.R.M. de A. Brt~ning, Oil Gas J., August (1991) 38.

[1] D. Kessel, H.J. Neumann and I. Rahiman. Proc. 1st

World Congr. on Emulsions, Paris, 1993, Paper 1 22 071.