Power Systems

Vasily Y. Ushakov et al.

Impulse Breakdown of Liquids
Vasily Y. Ushakov et al.

Impulse Breakdown
of Liquids

With 202 Figures and 23 Tables

Vasily Y. Ushakov
Tomsk Polytechnik University
Lenin Ave. 30
TOMSK
russische föderation 634034
icd@tpu.ru

Library of Congress Control Number: 2007929737

ISBN 978-3-540-72759-0 Springer Berlin Heidelberg New York

This work is subject to copyright. All rights are reserved, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilm or in any other way, and storage in data
banks. Duplication of this publication or parts thereof is permitted only under the provisions
of the German Copyright Law of September 9, 1965, in its current version, and permission
for use must always be obtained from Springer. Violations are liable for prosecution under
the German Copyright Law.
Springer is a part of Springer Science+Business Media
springer.com
c Springer-Verlag Berlin Heidelberg 2007
The use of general descriptive names, registered names, trademarks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
Production: Integra Software Services Pvt. Ltd., Puducherry, India
Cover design: deblik, Berlin
Printed on acid-free paper SPIN: 11685579 42/3100/Integra 5 4 3 2 1 0
The main physical processes and phenomena in pulsed electric breakdown of
liquids are described. The nature of charge carriers, mechanisms of formation
and evolution of the gas phase, and their role in charge ignition (initiation)
and development are considered. Spatiotemporal laws of propagation and pa-
rameters of charge channels are described. The basic mechanisms of liquid
breakdowns and boundary conditions of their initiation are formulated. The
monograph will be useful to experts in high-voltage pulsed technology, physics
of dielectrics, and electrical insulation as well as to students of the correspond-
ing speciality.
Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI

Introduction. Brief Historical Review . . . . . . . . . . . . . . . . . . . . . . . . . XV

References to Preface and Introduction . . . . . . . . . . . . . . . . . . . . . . XXIV

1 Behavior of Liquids in Strong Electric Fields . . . . . . . . . . . . . . 1

1.1 Liquid Pressure under an Electric Field . . . . . . . . . . . . . . . . . . . . 1
1.1.1 Liquid pressure under conditions of
impulse breakdown . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1.2 Nature of the electrostriction pressure . . . . . . . . . . . . . . . . 3
1.1.3 Pressure in electro-hydrostatic equilibrium . . . . . . . . . . . . 3
1.2 Transient Processes of Establishing Pressure . . . . . . . . . . . . . . . . 4
1.2.1 Electrostriction wave near the spherical electrode
under impulse voltage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.2.2 Dipole saturation effect in water and its influence
on pre-breakdown processes in the system of tip
electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.2.3 Electrostriction waves in a coaxial line . . . . . . . . . . . . . . . 13
1.3 High-Voltage Electric Conduction of Liquids . . . . . . . . . . . . . . . . 20
1.3.1 Generation of charge carriers . . . . . . . . . . . . . . . . . . . . . . . . 20
1.3.2 Motion of charge carriers . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
1.4 Phenomena at the Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.4.1 Double electric layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
1.4.2 Near-surface charge generation . . . . . . . . . . . . . . . . . . . . . . 27
1.4.3 Emission of charge carriers and electrode reactions . . . . . 35
1.4.4 Field distribution at the electrodes . . . . . . . . . . . . . . . . . . 37
1.5 Pressure Distribution during Charge Carrier Emission . . . . . . . . 41
1.5.1 Uniform field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
1.5.2 Sharply nonuniform field . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
VIII Contents

2 Behaviour of the Gas Phase in Liquids

in Strong Electric Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
2.1 Nature and State of the Gas Phase (Bubbles) in Liquids . . . . . . 54
2.1.1 Bubbles in Liquid before the Electric Field Application . 54
2.1.2 Deformation in an Electric Field . . . . . . . . . . . . . . . . . . . . 56
2.1.3 Deformation and Stability . . . . . . . . . . . . . . . . . . . . . . . . . . 57
2.1.4 Deformation in the Linear Approximation . . . . . . . . . . . . 58
2.1.5 Stability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
2.2 Influence of the Electric Field on the Bubble Formation . . . . . . 67
2.2.1 Electrostatic Mechanism of Bubble Formation . . . . . . . . . 67
2.2.2 Influence of Pressure Variations Inside
the Bubble on Phase Equilibrium . . . . . . . . . . . . . . . . . . . . 69
2.2.3 Calculations of the Pressure Inside
the Bubble upon Small Deformations . . . . . . . . . . . . . . . . 71
2.2.4 Influence of the Bubble Shape Instability
on Boiling of Liquids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
2.2.5 Influence of the Electric Field on the Boiling Point of
Liquids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
2.2.6 Influence of the Electric Field on Metastable Liquids . . . 75
2.2.7 Analysis of the Feasibility of Forming Pre-breakdown
Bubbles Under the Effect of Electric Field . . . . . . . . . . . . 75
2.3 Analysis of Pre-breakdown Bubbles . . . . . . . . . . . . . . . . . . . . . . . . 78
2.3.1 Deformation of a Moving Bubble in an Electric Field . . 79
2.3.2 Influence of Electrohydrodynamic Flows . . . . . . . . . . . . . . 80
2.4 Experimental Investigations of Gas Bubble
Behavior in an Electric Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2.4.1 Methods and Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
2.4.2 Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86
2.4.3 Formation of Long-living Bubbles and their Behavior
in Pre-breakdown Fields . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
2.4.4 Discussion of Experimental Results . . . . . . . . . . . . . . . . . . 92
2.4.5 Discharge Ignition in Water with the Help of Bubbles . . 99
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104

3 Discharge Propagation in Short Gaps with a Quasiuniform

Field . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
3.1 Discharge Propagation from Anode . . . . . . . . . . . . . . . . . . . . . . . . 110
3.2 Discharge Phenomena in Submillimeter Distilled-Water Gaps . 121
3.3 Time Boundaries of Anode and Cathode Discharges
in n-Hexane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
3.4 Influence of the Pressure on the Electric Discharge
Mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
3.5 On the Nature of Electric Hardening of Liquids in
Micron Gaps . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
 Contents IX

4 Discharge Propagation in Gaps with a Non-Uniform Field . 133

4.1 Spatial Structure of a Discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
4.1.1 Appearance of Discharge Figures . . . . . . . . . . . . . . . . . . . . 135
4.1.2 Electrostatic Properties of Discharge Figures . . . . . . . . . . 139
4.1.3 Dependence of the Spatial Structure of Discharge
Figures on the Main Influencing Factors . . . . . . . . . . . . . . 144
4.2 Discharge Kinetics under Microsecond Impulse Voltage . . . . . . . 156
4.2.1 Highly Non-uniform Fields (Small r0 ) . . . . . . . . . . . . . . . . 157
4.2.2 Weakly Non-uniform Fields (Large r0 ) . . . . . . . . . . . . . . . 162
4.2.3 Discharge in Purified Water and Ethyl Alcohol . . . . . . . . 164
4.2.4 Discharge in Liquids with Increased Electrical
Conduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
4.2.5 Special Features of the Discharge
on the Voltage Impulse Front . . . . . . . . . . . . . . . . . . . . . . . 173
4.2.6 Summary and Discussion of Results . . . . . . . . . . . . . . . . . . 175
4.3 Kinetics of Nanosecond Discharges . . . . . . . . . . . . . . . . . . . . . . . . . 178
4.3.1 Discharge in Distilled Water . . . . . . . . . . . . . . . . . . . . . . . . 178
4.3.2 Discharge in Transformer Oil . . . . . . . . . . . . . . . . . . . . . . . 182
4.4 Discharge in a Liquid with a Solid Dielectric Interface . . . . . . . . 183
4.4.1 Barrier Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
4.4.2 Creepage Discharge(Frashover). . . . . . . . . . . . . . . . . . . . . . 192
4.5 Ionization Region and its Parameters . . . . . . . . . . . . . . . . . . . . . . 204
4.6 Parameters of the Leader Channel . . . . . . . . . . . . . . . . . . . . . . . . . 207
4.6.1 Channel Parameters in the Stage of Conversion
into a Leader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
4.6.2 Steady-state Values of the Leader Channel
Parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
4.7 Processes in the Stage of Transition from the Leader
to an Arc . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

5 Statistical Investigations of the Electrical Breakdown . . . . . . 239

5.1 Possibilities and Methods of Statistical Investigations . . . . . . . . 239
5.1.1 Relationship between the Statistical Characteristics
and Nature of the Discharge Processes . . . . . . . . . . . . . . . 239
5.1.2 Automated System for Statistical Investigations . . . . . . . 241
5.1.3 Methods of Analyzing Statistical Distributions
of the Discharge Time Lag . . . . . . . . . . . . . . . . . . . . . . . . . . 243
5.2 Dependence of the Statistical Characteristics on the Discharge
Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
5.3 Discharge Mechanisms in Liquids from the Data
of Statistical Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
X Contents

6 Basic Laws Describing of the Impulse Electric Strength

of Liquids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
6.1 Duration, Shape, and Frequency of Voltage Pulses . . . . . . . . . . . 263
6.2 Voltage Polarity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
6.3 External Pressure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
6.4 Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
6.5 Chemical Nature and Composition of Liquids . . . . . . . . . . . . . . . 281
6.5.1 Chemical Nature of Liquids . . . . . . . . . . . . . . . . . . . . . . . . . 281
6.5.2 Electrical Conduction and Permittivity . . . . . . . . . . . . . . . 282
6.5.3 Soluble Additives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
6.5.4 Insoluble Impurities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
6.5.5 Moisture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
6.6 Discharge Gap Geometry and Conditions on the Electrodes . . . 301
6.6.1 Macro- and Microgeometry of the Electrodes . . . . . . . . . . 301
6.6.2 Dielectric Electrode Coatings . . . . . . . . . . . . . . . . . . . . . . . 306
6.6.3 Interelectrode Distance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
6.6.4 Electrode Area . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
6.6.5 Liquid Dielectric Volume in the Interelectrode Gap . . . . 311
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314

7 Physical Discharge Initiation (Ignition) Mechanisms . . . . . . . 319

7.1 Bubble Discharge Initiation Mechanism . . . . . . . . . . . . . . . . . . . . 320
7.1.1 Physical Model of Discharge Initiation . . . . . . . . . . . . . . . 321
7.1.2 Comparison with Experimental Results . . . . . . . . . . . . . . 328
7.2 Microexplosive Discharge Initiation Mechanism . . . . . . . . . . . . . . 338
7.2.1 Anode Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
7.2.2 Cathode Initiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
7.3 Ionization Mechanism of Discharge Initiation . . . . . . . . . . . . . . . . 363
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 365

8 Mechanisms of Discharge Propagation

and their Applicability Limits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 369
8.1 Propagation of the Fast (Anode) Channel . . . . . . . . . . . . . . . . . . 370
8.2 Propagation of the Slow (Cathode) Channel . . . . . . . . . . . . . . . . 378
8.3 Boundaries of the Regions of Existence
of Slow and Fast Discharges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 384
8.4 Conversion of the Initial Channel into the Leader . . . . . . . . . . . . 386
8.5 Conditions of Realization of the Leader Process . . . . . . . . . . . . . 392
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 393
Preface

Recently there has been a remarkable surge of interest in the phenomenon of

pulsed electric breakdown in liquids in connection with the rapid development
of high-voltage pulsed power. High-voltage micro- and nanosecond pulses find
wide application in experimental physics, electrodischarge technology, physics
of dielectrics, radar detection and ranging, high-speed photography, etc.
In the last decades, a new direction of science and technology – pulsed
power – has been originated from electrical power engineering and physics.
Sources of high pulsed currents (105 –106 A) and voltages (106 –107 V) with
nanosecond front duration are required to carry out the experiments on super-
fast plasma heating in laser and accelerating technologies and in a number of
other cases. In such generators, a high-voltage pulse is first generated without
stringent requirements on its shape; then this pulse is transformed with the
use of low-inductive energy storages and dischargers. Extremely high values of
the parameters of generated pulses impose a number of stringent requirements
on the insulation medium of storages and switches and on configurations of
these elements in actual circuits. The requirements on insulation can be briefly
formulated as follows: 1) high electric strength, 2) high permittivity, 3) low
electrical conduction, 4) low rate of decomposition in electric discharges, and
5) fast restoration of the dielectric properties after breakdowns. The experi-
ence on application of various dielectric media as insulation in high-voltage
pulsed power accumulated by the present time demonstrates that purified wa-
ter and dielectric liquids, in particular, transformer oil meet the above-listed
requirements. An increase in working electric field gradients required of in-
sulation of pulse generators with the state-of-the-art parameters obligatory
calls for reduction of their overall dimensions and hence parasitic parameters
in order to shape and use nanosecond pulses.
To simplify the design of high-voltage high-current nanosecond generators
with liquid insulation, to reduce their inductance, and to make a solution
of the insulation problem easier, it is expedient to switch currents using liq-
uid dielectric dischargers. Moreover, in many cases controllable dischargers,
which allow a current to be switched to several parallel spark channels thereby
XII Preface

increasing the power transferred to a load (a vacuum diode, solenoid, etc.),

must be used because of the requirements imposed on generated pulse power.
Thus, two problems – the problem of insulation and creation of high-voltage
high-current high-speed dischargers – determine in many respects an increased
interest in the phenomenon of pulsed electric breakdowns in liquids.
The start of the intensive development of pulsed power have coincided
with the expansion of works on the application of high-voltage spark dis-
charge energy in industrial technologies. When spark is a working tool, the
development of the discharge in a liquid is the first stage of the complex
technological process, and the liquid itself plays the role of an insulating and
working medium. In this case, the pulsed discharge characteristics influence
the laws of energy liberation in the spark channel and finally, the formation
of various pulsed disturbances (hydrodynamic, thermal, acoustic, light, etc.)
used in the technology.
At the same time, interest in the liquid breakdown under lightning pulses
and in particular, internal surges has quickened in connection with an expan-
sion of working voltages of electric power transmission lines and substations
and the tendency to a decrease in the insulation level of high-power electrical
equipment systems.
The knowledge of the electric breakdown of liquids has not kept pace with
this increasing interest and more and more stringent requirements on liquid
dielectric insulation design. In the literature, 20 or so different models of liquid
breakdown were considered equivalent, thereby testifying to the absence of
reliable experimental data on the laws of discharge propagation and their
dependence on the liquid properties and breakdown conditions.
The foregoing gave impetus to investigations of the electric breakdown in
liquids in the laboratories of the leading countries in the early 60s.
The main results of these investigations obtained till 1972–73 have been
generalized in monograph [1]. In the next 30 years, investigations of the elec-
trical conduction and breakdown in liquid in the USSR (later in Russia),
France, the USA, Great Britain, Norway, Japan, Poland, India, China, and
in other countries were characterized not only by a wide variety of works but
also by the most advanced experimental technology and modern methodical
procedures.
The present book is devoted to a description of physical mechanisms of
initiation and propagation of pulsed discharges in liquids as well as to the
basic laws describing impulse electric strength of liquids. The book is based
mainly on the investigations carried out at Tomsk Polytechnic University,
Novosibirsk State University, Siberian Scientific-Research Institute of Power
Engineering, Novosibirsk State Technical University, Saint-Petersburg State
Technical University, Nuclear Physics Institute of the Siberian Branch of the
Russian Academy of Sciences (Novosibirsk), and All-Russian Electrical Engi-
neering Institute. The results obtained in foreign laboratories and institutes
and necessary for the integrated presentation of the problem formulated in
the title of the monograph have also been included.
 Preface XIII

In the process of writing, we have generalized the results of modern re-

search and re-analyzed and re-examined a large volume of data on liquid
dielectric breakdown obtained in the last decades. This has provided their
deeper understanding and interpretation.
We especially wish to thank the Rector of Tomsk Polytechnic University
Prof. Yury P. Pokholkov for financial and moral support in preparation of the
English version of this monograph.
We would like to acknowledge all our colleagues who took part in investi-
gations that provided the basis of the monograph. Their names are presented
in References.
The authors are grateful to E. V. Bogdanova and S.V. Barabin for help in
preparation of the manuscript for printing.
The Foreword, Introduction, Conclusions, Chapters 4, 6, and 8, and the
introductory part of Chapter 7 have been written by V. Y. Ushakov; Sections
7.3 and 8.2 has been written by V. Y. Ushakov and V. F. Klimkin.
Chapters 3 and 5 and Section 7.2 have been written by V. F. Klimkin.
Chapters 1 and 2 and Section 7.1 have been written by S. M. Korobeynikov.
The translation of monograph have been made by Ph.D. L.G. Shamonaeva,
Institute of Atmosphere Optic of the Siberian Branch of Russian Academy of
Sciences, Tomsk.
Introduction. Brief Historical Review

The concept of the physical mechanism of liquid electric breakdown in has

run a very complex course. Even in the early 20th century, two radically dif-
ferent approaches to the phenomenon of liquid electric discharge were clearly
distinguished. According to one of them, the electric discharge in a liquid is
the discharge in gas cavities either already presented in the liquid and on the
electrodes or newly formed under voltage exposure (by electrolysis, boiling,
cavitation, and decomposition under electron bombardment). This discharge
mechanism is conventionally called the bubble discharge. According to the
other approach, the discharge is a consequence of avalanche multiplication of
free charge carriers in the liquid and is described by the gas discharge model
generalized to the liquid phase. It implies that electrons in strong fields are
accelerated in liquids and ionize molecules and atoms. The electric strength
of the liquid is related to the molecular liquid structure through the elec-
tron free path, interaction cross section, and ionization energy. This discharge
mechanism is further called the ionization discharge.
Macfarlane [2] pointed out in 1880 that under long-term voltage exposure,
four processes simultaneously participate in the liquid breakdown: formation
of filaments from suspended particles, liquid flow, formation and motion of
gas bubbles, and finally, spark discharge. This point of view was confirmed
by experimental investigations that provided the basis for a number of non-
ionization theories of liquid breakdown: thermal, gas-thermal, voltaic, polar-
ization, etc. A detailed review of these theories can be found in [3–6]. This
point of view prevailed till the middle 40s.
At the same time, external analogs in the phenomena accompanying the
electric discharge in liquids and gases as well as similarity in the laws describ-
ing the electric strength of liquids and gases provided the basis for the at-
tempts to develop the theory of liquid breakdown starting from the Townsend
theory [3, 7–9]. In more recent investigations it was established that the liquid
impurity, temperature, and gas pressure above the liquid affect the breakdown
under impulse voltage much weaker than under long-term voltage exposure.
This fact together with small delay times of breakdowns in liquids (∼10–8 s)
XVI Introduction. Brief Historical Review

established by A. F. Valter and L. D. Inge [10] and other authors were in-
terpreted as the proof of the ionization breakdown mechanism. These results
have demonstrated that the Townsend breakdown mechanism has only limited
application to a description of the discharge phenomena in liquids.
A. F. Valter and L. D. Inge put forward a hypothesis that initial electrons
causing the collision ionization in the ionization breakdown mechanism are
emitted into the liquid from the cathode. Baker and Boltz [11] accepted this
hypothesis, and more recently it provided the basis for a number of theories
of liquid breakdown [12–17].
Goodwin and Macfadyen [12] tried to find an analytical expression for the
criterion of liquid breakdown with allowance for the field emission from the
cathode described by the Fowler–Norgdheim equation and for the collision
ionization described by the Löbe equation [18]. The field strength at which
a current instability arises due to emission and ionization is taken to be the
breakdown field strength. In [6, 19] it was demonstrated that this model has
many mistakes and incorrect initial premises; therefore, the coincidence of
theoretical and experimental values of the electric strength was only random.
In the Bragg–Sharbaugh–Crowe model [15], the relationship of the emis-
sion current density, space charge, and field intensities at the cathode and
anode is established in the absence of ionization. Assuming that the break-
down occurs when the field strength in the entire liquid volume is no less than
a certain fixed value, the authors derive the breakdown criterion from the
condition of electron emission from the cathode. Three cases corresponding
to the cathodes with strong, intermediate, and weak emissivities were consid-
ered. In the first case, the current obeys the emission law with allowance for
the space charge; this dependence is observed until the average field strength
in the gap reached 2/3Ebr =2/3Eav . AtEbr , the field at the anode is 1.5 times
greater than Eav , that is, it reaches the breakdown value. In the second case,
the field at the anode reaches Ebr forEav lying between 2/3Ebr andEbr . Fi-
nally, the space charge is not formed for the weak emitter, and the breakdown
is initiated atEav = Ebr . Only in the third case the process depends on the
conditions on the cathode.
According to the Swan hypothesis [16], the breakdown is formed mainly
due to the field emitted from the cathode and amplified by positive ions accu-
mulated on the dielectric or semiconductor film on the cathode surface. The
collision ionization is considered only as a reason of field distortion at the cath-
ode. The field strength at which positive ions at the cathode cause the emission
current to increase continuously until the breakdown initiation in the liquid is
taken to be the breakdown field strength. For well-fitted empirical coefficients
in the equation for the criterion, the calculated dependenceEbr = f (d) for
liquid argon was in good agreement with the experimental dependence.
On the contrary, in the Lewis model [13] the electron emission from the
cathode was not considered at all. It was assumed that for high enough elec-
tric field strengths, a number of electrons acquire more energy from the field
than lose in the process of non-ionization collisions with molecules of the
 Introduction. Brief Historical Review XVII

liquid. These electrons are accelerated up to the ionization energy and give
rise to electron avalanches. From all collision types causing the electron en-
ergy losses, only the energy losses on the excitation of molecular vibrations in
the medium were taken into account. The breakdown condition was derived
from the Hippel criterion [20] for solid dielectrics based on the energy balance
condition:

[qEλ = C · h · ν, ] (1)
whereE is the electric field intensity, λ is the electron free path,C is an arbi-
trary constant,hν the quantum of energy the electron loses by excitation of
molecules. The satisfactory agreement between the calculated and experimen-
tal Ebr was obtained for a number of hydrocarbons.
In the Kuchinsky model [17] developed for thin layers (10–4 –10–2 cm) of
mineral oil, the initial low-intensive partial discharges are attributed to the
processes of collision ionization, and the occurrence of critical discharges is
attributed to a sharp increase in the gassing intensity and formation of gas
bubbles. The condition of the non-stationary mode in the gap leading to an
increase in the positive charge at the cathode that causes a fast increase in
the emission current density thereby progressively increasing the space charge
density until a breakdown of the oil layer is initiated is taken as the break-
down criterion. Based on the study of the characteristics of individual partial
discharge pulses, Kuchinsky [17] estimated the collision ionization coefficient
α and found a dependence of α on E. This allowed him to calculateEbr of
thin mineral oil layers.
Not dwelling on a detailed analysis of the above-mentioned theories con-
sidered in [5, 6, 19, 21], we must note the following. In many careful investi-
gations of the dependence of static and pulsed electric strength of very pure
liquids and conduction currents in strong electric fields on cathode material,
gap length, molecular structure of the liquid, etc. [5, 6, 19–22], the main points
of these models were not confirmed. The cathode material and temperature,
that is, the cathode emissivity had essentially no effect on the electric strength
of even micron gaps, though emission played the main role in the high-voltage
electrical conduction of dielectric liquids. Moreover, the initial concept of ion-
ization breakdown models for liquids that the discharge in a uniform field
always develops from the cathode due to initiation by emitted electrons was
violated for pulsed breakdowns.
The collision ionization as a basis of the ionization mechanism of break-
down in liquids was not convincingly confirmed with experiments as well. A
decrease in Ebr and an increase in the pre-breakdown current with increasing
interelectrode distance pointed out in a number of early experimental works
and used to prove the collision ionization mechanism in liquids were inter-
preted in more recent works as a consequence of imperfection of the experi-
ments and, in particular, of application of spherical rather than flat electrodes.
A decrease inEbr and an increase in current with increasing distance between
XVIII Introduction. Brief Historical Review

the hemispherical electrodes can be caused by an increased area of electrodes

located in the strong field and hence, an increased number of effective weak
sections and emitting centers [21].
Thus, the authors would have to consider the coincidence of the calculated
and experimental data on the electric strength the only criterion for model
reliability. However, the experimental data on the electric strength of even so-
called extremely pure liquids are highly diversified; therefore, their coincidence
with the calculated results is mostly random.
Because of lacking of the direct experimental proof of the correctness of
ionization liquid breakdown models developed in [12, 13, 23, 24] and the de-
pendence of electric strength on hydrostatic pressure established in a number
of works even for degassed liquids under short-term voltage exposure, a re-
birth of interest in the non-ionization (bubble) liquid breakdown mechanism
reveals itself in models [21, 23–25] developed in the 60s. Moreover, conditions
of realization of this discharge model were generalized to short-term voltage
exposure down to pulses of nanosecond duration.
The Kao model [25] suggests that a gas bubble that already existed or was
newly formed in the liquid under voltage exposure is extended along the field
lines. To find a mathematical expression for the breakdown criterion, Kao has
made the following assumptions: a) the volume of the deformed bubble re-
mains unchanged and b) the breakdown occurs when the voltage drop inside
the bubble is equal to the minimum of the Paschen curve (for the gas inside
the bubble). The theoretical dependence of electric strength of the liquid on
hydrostatic pressure was analogous to the experimental one. However, calcu-
lations of the electric strength for this model are difficult, because the model
does not describe the process of gas bubble formation, and it is necessary to
know the initial bubble radius, the state of the gas in the bubble, and some
other parameters that are difficult to measure. In addition, the model is in-
applicable when the condition of constancy of the deformed bubble volume
is violated or the interelectrode distance is much greater than the deformed
bubble length.
Gassing due to local (near geometrical cathode inhomogeneities) liquid
heating by emission currents provides the basis for the Watson–Scharbaugh
model [21]. The breakdown condition is the violation of equality between the
energy necessary for boiling up of a certain liquid mass and the energyW lib-
erated in the liquid as well as the gas bubble extension up to the critical length
or complete bubble crossover of the interelectrode distance. The energyW is
determined as

[W = A · E n · τr , ] (2)
where E is the field intensity, n is the power determined by the emission cur-
rent, space charge, and applied field strength, and τr is the time during which
the liquid mass can be located near the microtip without its replacement by a
colder liquid. If voltage pulse duration is longer than τr , the electric strength
 Introduction. Brief Historical Review XIX

of the liquid must be independent of the pulse duration. The uncertainty of

almost all factors entering into Eq. (2) gives no way of quantitative calcu-
lations for this model, and the experimentally established fact of preferable
pulsed discharge ignition on the anode rather than on the cathode calls into
question the correctness of the initial premises of this model.
The well-known phenomenon of the decreased electric strength of liquids
in the presence of suspended solid particles provides the basis for the Kok–
Corbey model [23]. It suggests that the suspended particles – polarized spheres
whose permittivity is much greater than the permittivity of the liquid – move
toward the strong field region under voltage exposure. The diffusion and vis-
cous friction hinder the particle motion. The field strength at which parti-
cles are progressively accumulated near a field amplification center forming a
bridge along which a breakdown occurs is taken as a breakdown field strength.
The model suggests an increase inEbr with increasing temperature, which con-
tradicts the available experimental data. Moreover, the formation of bridges
from suspended particles seems unrealistic under impulse voltages.
Thus, the non-ionization breakdown models are also based on premises
that in most cases were not confirmed experimentally.
In conclusion of our brief analysis of theoretical liquid breakdown mod-
els, it should be noted that no one of these models considers the discharge
evolution which in many respects determines the breakdown characteristics,
especially under short-term voltage exposure. The ionization theories consider
only the conditions of the build-up of collision ionization or electron emission
that are identified with the breakdown criterion. Skanavi [5] emphasized that
the condition of the build-up of collision ionization is insufficient criterion of
electric breakdown even if its nature is purely ionization. In the non-ionization
models, conditions of progressive gassing, gas bubble deformations, formation
of the bridge from impurity particles, etc. are identified with the breakdown
criterion.
The history of gas discharge physics demonstrates that a reliable break-
down theory can be elaborated only based on a detailed physical pattern
of the phenomenon, including a description of all discharge phases develop-
ing in time and space. This approach to the study of breakdown is especially
promising for liquids, taking into account a wide variety and complexity of the
discharge formation and development processes. Even in the most favorable
case of breakdown in a simple carefully purified liquid, it is impossible to es-
tablish a reliable relationship between the electric strength and the molecular
structure of the liquid that could be used as a basis of the theory. Despite care-
ful purification by different methods and decontamination, the liquids used in
experiments on breakdown were still far from being ideal. For example, in [12]
it was demonstrated that the electrical conduction of the liquid continuously
purified and decontaminated for five months decreased but did not reach the
limit which could be considered as the physical-chemical constant of this liq-
uid. In addition, the situation is complicated by lacking of a well-developed
theory of material in the liquid phase.
XX Introduction. Brief Historical Review

Komelkov [26, 27] first described the propagation of the discharge in a

liquid using a high-speed camera for photographic registration of the discharge
channel and simultaneously an oscillograph for registration of voltage and
current accompanying the discharge propagation.
The discharge in transformer oil and distilled water in tip-plane and tip-tip
gaps of length d =12–20 cm was photographed with the Boys rotating lens-
type camera. Satisfactory streak photograph were obtained with the help of ar-
tificial discharge retardation using a high ohmic resistanceRret (104 –1.5·106 Ω)
connected into the discharge circuit. It was established that the breakdown
of the gap was similar to the leader process in long air gaps followed by a
bright burst considered by Komelkov as a main (return) discharge in long air
gaps; after the burst, an arc was observed. At highRret , repeated breakdowns
were observed after the main discharge. It was pointed out that breaks in
leader propagation were caused by voltage relaxation whose time was deter-
mined by the circuit parameters (R,L,C), electron capture by molecules and
atoms of the liquid that led to aging of the leader channel and decreasing
its electrical conduction and electric field gradient in the region of ionization,
and the leader deceleration by the space charge barrier formed in the gap by
the ionization zone. The length of the ionization zone was 1–2 cm, and the
effective velocities of the leader channel propagation in transformer oil were
1.7·105 and 6·104 cm/s for positive and negative leaders. Based on the results
of these investigations, Komelkov has made a very important conclusion that
the main concept that provided the basis for the classical theories, namely,
the concept that the breakdown is the build up of pre-breakdown currents
in the gap or the single-avalanche process was incorrect. Actually, a quali-
tatively new phenomenon, involving the formation of a conduction channel
and an ionization zone, takes place, and the processes of collision ionization
proceed in a local moving region. He also pointed out that the development
of the leader from the tip electrode of positive polarity demonstrated that
the electron emission from the cathode is not the governing factor of the dis-
charge mechanism in long gaps. He emphasized that the discharge in liquids
was similar to the breakdown of long rather than short gas gaps similar to a
lightning and hence it can be used to model the lightning.
More recently, Liao and Andersen [28, 29] carried out analogous investi-
gations. To clip currents and glow in the final stage of the discharge, they
put an insulation plate on the flat electrode. The linear film framing speed in
the Boys camera was about 4·103 cm/s. The discharge parameters (velocity,
current, and duration of pauses between bursts) were the same as in [26, 27],
but the discharge evolution was treated differently. According to Liao and
Andersen, it includes five rather than three stages: 1) initiating streamer, 2)
equivalent pilot-streamer, 3) stepped streamer, 4) return discharge, and 5)
arc. At the moment of voltage application to the discharge gap, the initiat-
ing streamer develops from the tip electrode. It passes half the interelectrode
distance (3–4 cm) in 1 μs. From the end of the initial negative streamer, con-
tinuous diffusion glow – the pilot-streamer – is observed. It is reached by a
 Introduction. Brief Historical Review XXI

series of streamers with intervals between bursts 1–5 μs. After the next burst
(step), a pause in streamer propagation is observed during which the streamer
channel does not glow. For the negative discharge, the pilot-streamer devel-
ops after the stepped streamer from its head; it propagates with a velocity
of 2.5·105 cm/s. For the positive streamer, jumps in the potential and bursts
in the streamer channel corresponding to them are also observed; however, a
smooth increase of the potential in this case is absent. Based on these results,
the authors concluded that the pilot-streamer was absent in this case. When
the stepped streamers reached the opposite electrode, the inception of the
main discharge stage was observed followed by the arc.
Because of artificial clipping of the discharge current (the discharge re-
tardation), the lack of the data on the discharge channel parameters (cross
section, longitudinal electric field gradients, current density, electrical con-
duction, etc.), and different discharge schemes used in [26–29], the leader
liquid breakdown mechanism in long intervals has been recognized not by
all researchers of this phenomenon. For example, Balygin [30] on the basis
of investigations of the discharge in liquids with an oscillograph concluded
that the leader process in the liquid reported in [26] was due to the presence
of the high retarding resistance in the discharge circuit. Without resistance,
the discharge is developed as a single avalanche. The unjustified character of
these conclusions was discussed in [31], and their inaccuracy was demonstrated
in [32, 33].
The first experimental investigations of the discharge in liquids by the
methods of high-speed photography have demonstrated with all evidence that
they are promising. However, technical imperfection of cameras available at
that time with mechanical framing did not allow one to develop this direction
of research further and to detail the physical pattern of the discharge in liquids.
In the next decade (approximately from 1953 until 1963), the great number
of efforts had been made to create a breakdown pattern and theory based
only on the dependence of the electric strength and electrical conduction of
carefully purified liquids on various factors.
In the foregoing historical review, we have briefly considered the works
published approximately until the middle 60s.
By the early 60s, the opportunity appeared to study the discharge in liq-
uids using high-speed photographic devices equipped with electron-optical
image intensifiers (Image Converter Camera-ICC) having time resolution of
∼10–12 s with light amplification coefficient of ∼106 together with the methods
and means of nanosecond pulsed power, including high-speed oscillographs.
The first studies of electric breakdown in liquids with the use of ICC were
carried out by Ushakov [1] in the Laboratory of High-Voltage Gas Discharge
and Lightning Protection of the G. M. Krzhizhanovskii Institute of Power
Engineering in 1962–1964 (Moscow); then they were continued at Tomsk
Polytechnic Institute (University). In these studies of the physical processes
initiating pulsed breakdown of liquids, the main methodical procedure was
integrated synchronous registration of spatiotemporal patterns and electric
XXII Introduction. Brief Historical Review

parameters of discharges with high-speed electronic, electron-optical, and op-

tical equipment by methods of high-voltage pulsed power including nanosec-
ond one.
The application of these methods allowed the basic laws of formation and
propagation of pulsed discharges in liquids to be investigated for a wide range
of pulse durations and discharge gaps covering a significant part of the range
of their variations in actual designs.
Results of these investigations were generalized in [1]. The hypotheses
about two main breakdown mechanisms in liquids, namely, bubble and ion-
ization ones were confirmed, and their basic characteristics were elucidated.
The approximate boundary conditions of realization of these mechanisms were
also indicated.
In the bubble model, the main sources of charge carriers creating the
plasma discharge channel are the processes inside of the gas bubbles (voids)
that have already existed in the liquid or have been newly formed in a strong
electric field.
In the second model, it is accepted that plasma in the initial stages of
channel development (in “fresh” sections of the channel) is created as a result
of collision or autoionization (field ionization) processes in the liquid. The
charge carriers that are formed in this case and move in the electric field cause
warming and boiling up of the liquid and forming gas channels. Ionization can
develop in these channels in accordance with the gas discharge laws, converting
them into highly conductive channels (into the leader channels in long gaps).
In [1], the voltage exposure time and the electrical conduction of liquids
were chosen as the key parameters of one or another breakdown mechanism.
Investigations carried out in the leading laboratories all over the world in
the next 2.5–3 decades have demonstrated that the actual situation is much
more complicated. It was found that the nature and sequence of the discharge
processes depend not only on these actually most important factors but also
on a number of other factors including the curvature radius of the electrode
initiating the discharge, discharge polarity, viscosity of the liquid, interelec-
trode gap length, and hydrostatic pressure.
Further progress in the study of the electric discharge mechanism in liquids
was mainly due to application of laser technology. Laser light sources used in
optical methods (schlieren photography, interpherometry, etc.) allowed one to
obtain high temporal and spatial resolution. This is especially important for
studying the initial discharge stages. It should be noted that the high temporal
resolution provided by ICC was accompanied by deterioration of the image
quality at nanosecond exposure times. This limits their application for the
study of microobjects changing with high rates.
In the USSR, the first studies of the physical pattern of the electric dis-
charge in liquids with the use of a ruby laser as an illumination source in
high-speed schlieren photography were carried out at the Nuclear Physics
Institute of the Siberian Branch of the Russian Academy of Sciences (NPI
SB RAN) in 1970–1971 [34, 35]. They allowed some important details of the
 Introduction. Brief Historical Review XXIII

discharge initiation and development in water to be elucidated. The results

obtained demonstrated that the given direction of research, intensively devel-
oped over the next few years, was promising. The application of multiframe
systems allowed one to study the dynamics of the processes more reliably.
Further development of the new experimental approach to the study of the
electric discharge mechanism in liquids was connected with the refinement of
methods and modernization of means for ultrahigh-speed optical registration.
The special features of the electric discharge in liquids (a great variety and
complexity of the phenomena itself and its small characteristic dimensions
≤ 10 μm, high propagation rates ∼105–107 cm/s allowed us to formulate a
number of requirements on the methods of high-speed optical measurements:
1) illumination pulse duration no longer than ∼10–0.1 ns, 2) frequency of
shooting in the frame-by-frame mode ∼109 –108 frames/s, 3) time interval
between the frames changing in a wide range (∼1–100 ns), 4) high accuracy
of frame synchronization and the desirable frame number no less than 4–7,
5) light beam energy sufficient for registration of subsequent frames with the
necessary time delay, and 6) high beam quality to obtain reliable quantitative
data. Systems of multiframe photography based on light delay lines with a
laser illumination source were developed and used at the Institute of Applied
Mechanics of the Siberian Branch of the Russian Academy of Sciences and
Novosibirsk State University [36, 37].
In the early 80s, the interference and schlieren methods of laser diagnos-
tics of the phenomena from changes in the complex refractive index of the
examined medium, including algorithms of reconstruction of the permittivity
profiles in perturbed media from the amplitude and phase of a probing wave
with allowance for the diffraction, refraction, and absorption were developed
at the High-Voltage Research Institute at Tomsk Polytechnic University. The
diagnostic complex comprised:
– a laser with active mode synchronization providing the following param-
eters per pulse: duration of 1 ns, energy of 40 mJ, wavelength of 0.69 μm, and
instability of pulse of discharger triggering < 1 ns
– a synchronization system that allows gap sensing to be performed within
0.1–10000 μs of the start of the high-voltage pulse
– a two-frame system with spatial division of probing beams
– a two-frame system with a common camera angle and polarization divi-
sion of beams
– a Sagnac interferometer with triangular beam path outside of the exam-
ined object [38].
Approximately at the same time, two other research groups from Siberian
Scientific Research Institute of Power Engineering (Novosibirsk) [39, 40] and
Saint Petersburg State Technical University [41] developed optical methods of
registration of the dynamics of electric field re-distribution and (or) changes
in the liquid density based on the Kerr effect and used these methods to study
the pre-breakdown and breakdown processes in liquids.
XXIV Introduction. Brief Historical Review

In [39,40], optical registration was carried out using a SFR ultrahigh-speed

photodetector operating in the chronographic mode. Multiframe photographs
were obtained using a pulsed ruby laser in the Q-switched mode. In this case,
a series of light pulses with 50–80 ns duration was generated with time interval
between pulses changing from 1 to 40 μs. During the pause between pulses
(exposure time), the image moved along the film by the frame width, and
the frame was illuminated by the next light pulse. The number of subsequent
photographs was 10–20, that is, the method allowed the events that lasted
from a few tens to a few hundreds of microseconds to be registered.
The Mach–Zehnder laser interferometer with registration of interference
patterns in the photochronographic mode on an electron-optical photochrono-
graph was used in [41]. It was capable of registering short-term phenomena,
including nanosecond high-voltage pulses.
The experimental data obtained by more sophisticated experimental meth-
ods enabled us to address again the problem of integrated study of the nature
of electric breakdown in liquids on a new higher level.

References to Preface and Introduction

1. V. Y. Ushakov, Pulsed Electric Breakdown in Liquids [in Russian], Publishing
House of Tomsk State University, Tomsk (1975).
2. Mac Farlane, Philos. Mag., No. 10, 389 (1880).
3. A. Nikuradze, Das Flüssige Dielktrikum (Isolierende Flüssigkeiten).
Berlin,Verlag von Julius Springer, 1934.
4. F. F. Volkenshtein, Liquid Dielectric Breakdown [in Russian], Moscow-
Leningrad, (1934).
5. G. I. Skanavi, Physics of Dielectrics (Area of the Strong Fields). [in Russian],
State Ph. and Math. Press, Moscow (1958).
6. I. E. Balygin, Electric Strength of Liquid Dielectrics [in Russian], Energiya,
Leningrad (1964).
7. S. Whitehead, Dielectric Phenomena. Vol. 2. Electrical Discharges in Liquids,
E. B. Wedmore, ed., London (1928).
8. W. Rogowski, Der elektrische Durchschlag von gasen, festen und fluessigen
Isolatoren, Arch. Elektr., 23, 569–578 (1930).
9. F. W. Peeck, Dielektric Phenomena in High-Voltage Engineering, New York,
1930.
10. A. F.Valter and L. L. Inge, Electric breakdown in liquid dielectrics, Zh. Tekh.
Fiz., 4, No. 9, 1669–1687 (1934).
11. E. B. Baker and H. A. Boltz, Thermionic emission into dielectric liquids, Phys.
Rev., 51, No. 11, 989 (1937).
12. D. W. Goodwin and K. A. Macfadyen, Electrical conduction and breakdown in
liquid dielectrics, Proc. Phys. Soc. London, 66, No. 398, 85–96 (1953).
13. T. J. Lewis, The dependence of the dielectric strength of pure liquids on cathode
material, Proc. Phys. Soc. London, B66, No. 5, 425–428 (1953).
14. T. J. Lewis, The electric strength and molecular structure of liquids, Br. J. Appl.
Phys., 9, No. 1, 30–33 (1958).
 Introduction. Brief Historical Review XXV

15. J. K. Bragg, A. H. Sharbaugh, and R. W. Crowe, Cathode effects in the dielectric

breakdown of liquids, Appl. Phys., 25, No. 3, 382–391 (1954).
16. D. W. Swan, Electrical breakdown of liquid dielectrics, Proc. Roy. Soc. London,
No. 3, 423–432 (1961).
17. G. S. Kuchinskii, Theory of breakdown of thin liquid dielectric layers, Zh. Tekh.
Fiz., 36, No. 7, 1297–1304 (1966).
18. L. B. Loebe, Fundamental Pracesses of Electrical Discharge in Gases, N.Y. ,
1939.
19. T. Lewis, in: Progress in Dielectrics, Part 1, J. Berks and J. Shulman, 6, (1965).
20. A. Hippel, Electric breakdown of solid and liquid insulators, Physics, 8, No. 12,
816–832 (1937).
21. A. H. Scharbaugh and P. K. Watson, in: Progress in Dielectrics, Claredon Press,
Oxford (1962), pp. 201–248.
22. I. Adamczewski, Ionization, Conductivity and Breakdown in Dielectric Liquids,
Taylor and Francis Ltd. London, (1969).
23. J. A. Kok and M. M. G. Corbey, Testing the electric strength of liquid dielectric
or insulating material, Appl. Sci. Res., 6, No. 4, 285–295 (1957).
24. K. D. Metzmacher and J. E. Brignell, Application of the statistical model for
liquid breakdown, Colloq. Int. CNRS, No. 179, 145–159 (1970).
25. K. C. Kao, Theory of high-field electrical conduction and breakdown in dielectric
liquids, IEEE Trans., EI-11, 121–128 (1976).
26. V. S. Komelkov, Mechanism of pulsed breakdown in liquids, Dokl. Akad. Nauk
SSSR, 47, No. 4, 269–272 (1945).
27. V. S. Komelkov, Propagation of pulsed discharge in a liquid, Zh. Tekh. Fiz., 31,
No. 8, 948–960 (1961).
28. T. W. Liao and I. G. Anderson, Propagation mechanism of impulse corona and
breakdown in oil, AIEE Trans., 72, No. 1, 641–648 (1953).
29. I. G. Anderson and T. W. Liao, The propagation mechanism of impulse creepage
discharges over oil immersed surfaces, AIEE Trans., No.1, 74 (1955).
30. I. E. Balygin, On the discharge duration in liquid insulating materials, Zh. Tekh.
Fiz., 24, No. 2, 337–347 (1954).
31. G. A. Vorobjev, Comments to the paper of I. E. Balygin “Pre-breakdown cur-
rents in a liquid,” Zh. Tekh. Fiz., 27, No. 6, 761 (1954).
32. I. S. Stekolnikov and V. Y. Ushakov, Study of the discharge phenomena in
liquids, Zh. Tekh. Fiz., 35, No. 9, 1692–1700 (1965).
33. V. Y. Ushakov, Optical and oscillographic studies of a pulsed discharge in liq-
uids, Candidate’s Dissertation, Tomsk Polytechnic Institute, Tomsk (1965).
34. A. P. Alkhimov, V. V. Vorobjev, V. F. Klimkin, et al., On the propagation
of electric discharge in water, Dokl. Akad. Nauk SSSR, 194, No. 5, 1052–1054
(1970).
35. E. A. Abramyan, V. A. Kornilov, V. M. Lagunov, et al., Megavolt Energy Con-
centrator, Dokl. Akad. Nauk SSSR, 201, No. 1, 56–59 (1971).
36. V. F. Klimkin and A. G. Ponomarenko, Study of pulsed electric breakdown
in liquids by the method of optical interferometry, Zh. Tekh. Fiz., 49, No. 9,
1896–1904 (1979).
37. V. F. Klimkin, A multiframe ultraspeed laser shlieren photography system for
observation of nanosecond pre-breakdown phenomena in liquids, Zh. Tekh. Fiz.,
61, No. 9, 15–19 (1991).
XXVI Introduction. Brief Historical Review

38. V. R. Kukhta, V. V. Lopatin, and P. G. Petrov, Reconstruction of permittivity

profiles for symmetric microobjects from the interferometric data, Opt. Spek-
trosk., 56, No. 1, 179–181 (1984); A setup for investigation of the initial stage
of the electric discharge in dielectrics, Elektronn. Obrab. Mater., No. 3, 66–68
(1986).
39. I. T. Ovchinnikov, K. V. Yanshin, and E. V. Yanshin, Study of electric field
distributions in water using the Kerr effect, Zh. Tekh. Fiz., 44, No. 2, 472–474
(1974); Experimental investigations of pulsed electric fields in water at the tip
electrode using the Kerr effect, Zh. Tekh. Fiz., 48, No. 2, 2596–2598 (1978).
40. S. M. Korobeynikov, K. V. Yanshin, and E. V. Yanshin, Study of pre-breakdown
fields in nitrobenzene using the Kerr effect, in: Abstracts of Reports at the All-
Union Conf. on Physics of Dielectrics and New Areas of Their Application, Ser. 2
[in Russian], Karaganda (1978), pp. 18–19.
41. G. S. Kuchinskii and E. A. Morozov, Study of the physical phenomena in wa-
ter in pre-discharge electric fields, Pisma Zh. Tekh. Fiz., 8, No. 24, 1526–1531
(1982); Registration of electric fields in liquid dielectrics with the Mach–Zehnder
interferometer using the Kerr effect, Zh. Tekh. Fiz., 53, No. 6, 1215–1220 (1983).