DIELECTRIC STRENGTH AND ELECTRIC CONDUCTIVITY

OF LIQUID DIELECTRICS CONTAINING WATER

M. V. K u r l i n , A. A. S p i r i n , UDC 537.311 : 665.5

a n d Y a . M. F r a n k f u r t

The existing theories on the e l e c t r i c a l breakdown of liquid dielectrics containing water, which were proposed
by Hemant and Boning [1, 2], are based on r e l a t i v e l y old e x p e r i m e n t a l studies. Of the new experimental and t h e o -
r e t i c a l studies in this field, mention should be made of [3, 4, 6-9]. The lack of any practical confirmation of these
theories for many liquid dielectrics, as sell as the major disagreement in results from e x p e r i m e n t a l determinations
of the relation Ebr = f(% H20) for i d e n t i c a l dielectrics by different authors [1-5], is explained in particular by dif-
ferences in the temperature of the dielectric during the experiment, since increases in temperature lead to conver-
sion of water from the emulsified state to the dissolved state.

The aim of the present work is a further development of the existing views on the physics of e l e c t r i c a l b r e a k -
down of liquid dielectrics containing water. The studies were conducted on Balakhany heavy crude oils. The fol-
lowing experiment was performed to study the relation between dielectric strength and dissolved water content.
Water was placed in the bottom of a flask, which was then filled with crude oil that had been dehydrated in a Dean
and Stark apparatus. The flask was heated while stirring the crude oil continuously, but without allowing any m i x -
ing of the oil with the water. The crude oil, at the temperatures indicated in Fig. 1, was poured into the cup of an
AIMTI-60 oil breakdown apparatus, where Ebr was measured, and a sample was taken for determination of water
content by the Karl Fischer method. The measurements showed the independence of Ebr or dissolved water content
(Fig. 1, curve 1). In order to determine the dependence of Ebr of the crude oil on the amount of emulsified water
in the oil with a definite quantity of dissolved water, i.e., saturated with water at the given temperature, a natural
w a t e r - i n - o i l emulsion (taken from c o m m e r c i a l crude oil tanks) was added. The resulting emulsion, with a known
concentration, was mixed thoroughly. The measurements showed that the relation between Ebr and the amount of
emulsified water was n e a r - e x p o n e n t i a l (Fig. 1, curve 2). As a result of the experimental study of the relation b e -
tween breakdown strength of the crude oil and the amount of water in the oil, taking into account the state of the
water, the curve shown in Fig. 1 was plotted by combining the two relationships for dissolved and emulsified water
It should be noted that an analogous representation of the dependence of Ebr of liquid dielectrics on the amount of
water present, taking into account the state of the water, was given in [9] somewhat later than in our work [12].
Our measurements showed that, for an emulsion, Ebr is independent of d i a m e t e r of the dispersed water drops over the
range from 0.5 to 50 g, this range being obtained by adding emulsions drawn from valves located at different heights
in a tank containing a polydisperse crude oil emulsion that had been allowed to settle for a prolonged period of time.

This conclusion, which also agrees with that of Fries [1] and Brounshtein [10], points out the incorrectness of
the theory of Hemant and Boning, according to which Ebr depends on the diameter of the water globules.

The independence of Ebr of water drop diameter can be explained by dispersion of large drops into fine
emulsified droplets under the influence of the electric field when it reaches a certain c r i t i c a l strength thus forming
an emulsion that is close to one and the same degree of dispersity. These conclusions of ours are confirmed by the
work of V. G. Ben'kovskii [7] and also by research conducted at the V. I. Ul'yanov (Lenin), Leningrad Electrical
Engineering Institute and at the Vrotslavskii Polytechnic Institute (Poland).

When the crude oil was cooled during the t i m e of making the measurements, transition of the water from the
dissolved state to a microemulsion led to a sharp reduction in Ebr (Fig. 1, curve 3).

The differences between emulsified and microemulsified water in their effect on the d i e l e c t r i c strength of
the liquid dielectric is explained by their high cataphoretic mobility and also by the fact that nonuniformity of the
e l e c t r i c field caused by surface roughness of the electrodes (which according to [4] serves as the source of "bridge"
formation) is significant to microparticles with size of the same order as the magnitudes of the roughness, whereas

V. I. Ulyanov (Lenin) Leningrad Electrical Engineering Institute; AziNeftekhim. Translated from Khimiya i
Tekhnologiya Topliv i Masel, No. 12, pp. 46-50, December, 1966.

868
 Z6 9 1
80
f
o
gO ~ , g
;>
v /+0
s-~ , /z
z., /8-

i I I I I
0,0, o,0e o, lz o,m o,z
Water content, %
'r
0/ I I I I I I
Fig. 1. Dielectric strength of crude oil as a 1 g 3 4~ 5 6
function of water content: 1) Ebr ; f(% H20 Pgage, kg/cm2
dissolved); 2) Bor = f(% H20 emulsified); S)
Ebr = f(% H20 microemulsified); A) solubili-
Fig. 2. Breakdown voltage of crude
ty l i m i t of water in crude oil (100~ Tem-
oil with different water contents as a
peratures, ~ O) 15; IX) 88; [3) 75; e) 100.
function of pressure, water content,
%: 1) 0.02; 2) 0.04; S) 0.2; 4)
0.6; 5) 0.7; 6) 0.85; 7) 1; 8) 3; 9) 8.
s*~v9 e Z,

g
3
j_2-
0 g

30

tl/i//z • o
~>

g., l

0
J 0,5
I
z,

f
I I

Q, c m a / m m 2. sec

Fig. 4. Effect of bulk velocity in

4 ~ Ig 16 20 ~4
inter-electrode space on dielectric
U, kV strength with different amounts of
Fig. 8. Conduction current as a water in crude oil (%): 1) 0.S; 2)
function of applied voltage with 2.4; 3) 6.4; 4) 20.
different contents of water in crude
such nonuniformity is of no signifance in the case of emulsified par-
o i l ( w t . % ) : 1)0.02; 2)0.07; S)
ticles that are larger by three to four orders of magnitude.
0.2; 4) 0.45; 5) 0.6; 7) 1; 8) S;
9) 8; 10) 16. Thus, the disagreement among the results of various authors
should be explained by their use of different methods of adding
water to the dielectric; in some cases the water was present in the form of a microemulsion [5, 6] and
in other cases in the form of an emulsion [8].

We carried out further investigations in a test stand [11] with special devices for static and dynamic measure-
ments. Standard electrodes were installed in the device for static measurements. Coaxial cylindrical electrodes
(14 m m O.D., t0 m m I.D., 22 m m long) were used in the device for dynamic measurements. Data are shown in
Fig. 2 relating the breakdown voltage of crude oils with different water contents as a function of pressure, at a t e m -
perature of 22~ these data indicate that an increase in Ubr with increasing pressure is observed with water c o n -
tents of 0-1%.

Data are shown in Fig. 3 relating conduction current to applied voltage with different contents of water, at a
temperature of 22~ and a pressure of 6 k g / c m 2. The final points on the curves of I = f(U) correspond to the pre-
breakdown values.

Analysis of these curves shows that, with water contents from 0 to 0.85%, the relation between electric conduc-
tivity of the crude oil and the eIectric field voltage has a bend in the curve, analogous to the region of saturation

869