Capacity Building in Salt cavern

Prepared by
Chandan Kumar (Rock Mechanics Expert)
Gopi Kannan L. (Hydro Geological Expert)
Saikat Pal (Geological Expert)
U/G Storage Group, ETD
June 2016

1
 TABLE OF CONTENTS
Executive Summary 4

1. Rock salt : Geo -material 5

1.1 Introduction 5
1.2 Engineering Application of Rock Salt 6
1.3 Salt Cavern Storage Technology 11
1.4 References 14

2. Geological Formation of Rock Salt 15

2.1. Rock Salt Deposits 15
2.2 Rock Salt Distribution 19
2.3 References 24

3. Investigation 25
3.1 Stages of Investigation 25
3.2 Geological Investigation 28
3.3 Core Drilling 30
3.4 Geological logging of cores 31
3.5 Geophysical Investigation 32
3.6 Geotechnical Investigation 34
3.7 Hydro-geological Investigation 37
3.8 In-situ testing During Construction 39
3.9 References 40

4 Geo-mechanical Properties of Rock Salt 43

4.1 Physical Properties 43
4.2 Rock salt behaviour 44
4.3 Mechanical Properties of Rock Salt 45
4.4 Rheological Behaviour 51
4.5 Constitutive Laws for Rock Salt 54
4.6 Failure Strength of Rock Salt 60
4.7 Dilatancy of rock salt 61
4.8 Dilatancy criterion of rock salt 63
4.9 References 64

5 Design and Geo-mechanical Analysis of Salt Cavern 65

5.1 Rock Mechanics Cavern design 65
5.2 Rock Mass Model 69
5.3 Numerical Analysis 72
5.4 References 75

6 Evaporation 76
6.1 Introduction 76
6.2 Measurement of evaporation 79

2
 6.3 Evaporation Pond Design 81
6.4 References 84

7 Bikaner Rock Salt: A Case Study 85

7.1 Project Description 85
7.2 Geological Description of Project Site 87
7.3 Strength and Rheological Properties 88

Acknowledgement 107

3
Executive Summary

In view of the unique physical and mechanical behaviour of rock salt, solution mined salt
caverns are considered as an excellent storage medium for the inventory stockpiling of
hydrocarbons. Therefore, a total envisaged storage capacity of around 3.75 MM T (or 4.4 MM
m³) of crude oil in underground salt cavern is proposed near Bikaner, India under phase II
storage program. The site was selected due to the availability of favourable geological setting,
i.e. a bedded salt formation of appropriate depth and thickness and a shallow aquifer to provide
sufficient yield of a brackish groundwater. Salt caverns are engineered according to rock-
mechanical specifications utilizing the solution mining (or leaching) technology. The proposed
storage caverns near Bikaner are located at a depth ranging from 570m to 750m depth.

As a first part of this capacity building exercise in salt cavern technology, this paper discusses
in brief the salt cavern storage technology and critically reviews rock salt as a geo-material, its
engineering application, its physical and mechanical behaviour, geotechnical investigations
performed at different stages of project development, design and geo-mechanical analysis of
caverns and evaporation basin for the disposal of brine produced from cavern leaching. The
present paper also discusses geological setting of the project site located near Bikaner and the
physical & mechanical properties of the bedded salt formations within the Nagaur-Ganganagar
Basin as a case study. Constitutive model encapsulating the strength and rheological behaviour
of Bikaner Rock salt is also presented.

The next stages of capacity building shall include drilling and completion of around 1000m
deep boreholes, leaching technology including data requirements , simulation and procedures,
salt cavern stress and stability analysis, draft specifications for investigations, design of
evaporation basin as well as the project execution plan. An interaction with drilling contractor
of EIL oil / gas block is also envisaged as part of this capacity building exercise. Interaction
with IIT (Delhi) on the testing and study of geo-mechanical behaviour of rock salt shall
continue as a part of PhD program.

4
 1. Rock salt : Geo -material

1.1 Introduction

Rock Salt also commonly known as Halite, is the mineral form of sodium chloride (NaCl).
Halite is a very delicate mineral that way and physically weak compared to other rocks in
general. It does not last long on the Earth's surface, unless water never touches it. Over geologic
time, back to billion years of earth’s history, several enormous rock salt deposits were formed
in dry and arid climatic condition all over the world including India when repeated evaporation
occurred in restricted basins of lakes, playas, and seas. Under quite moderate pressure, rock
salt flows much like ice.

Though rock salt is dominantly halite, it can have some other minerals such as calcite, dolomite,
anhydrite, gypsum, and potash as well as some impurities of gypsum (CaSO4) and sylvite (KCl)
and occasionally polyhalite (K2Ca2Mg (SO4)4.2H2O) etc. It is a coarsely crystalline,
sedimentary rock containing massive, fibrous, or granular crystal aggregates. The mineral is
typically colorless or white as shown in Figure 1.1 (a), but may also be light blue, dark blue,
purple, pink, beef red, orange, yellow or grey depending on the amount and type of impurities
as shown in Figure 1.1 (b). It commonly occurs with other evaporite deposit minerals such as
several of the sulfates, halides, and borates. Some of the physical properties of Rock Salt are
outlined in the Table 1.1.

Fig. 1.1(a) Pure Rock Salt Fig 1.1(b) Rock Salt with impurity

5
 Table 1.1 Physical Properties of Rock Salt

Chemical Classification Halide

Colorless or white when pure; impurities produce any color
Color
but usually yellow, gray, black, brown or red
Streak White
Luster Vitreous
Diaphaneity Transparent to Translucent
Cleavage Perfect, cubic, three directions at right angles
Mohs Hardness 2.5
Specific Gravity 2
Diagnostic Properties Cleavage, solubility, salty taste
Chemical Composition NaCl
Crystal System Isometric

1.2 Engineering Application of Rock Salt

Commercially, most of the salt produced is crushed and used in the winter on roads to control
the accumulation of snow and ice. Because brine (a solution of water and salt) has a lower
freezing point than pure water, putting salt or saltwater on ice that is near 0°C (32 °F) will
cause it to melt. Salt is also used extensively used by the chemical industry. Salt is an essential
nutrient for humans and most animals (natural cattle lick) and it is also a favorite seasoning for
many types of food. Interestingly, Rock Salt is the only mineral that people eat.

Salt has been mined over the millennia for consumption and food preservation. Underground
mining of salt in Austria and Romania may have begun in the New Stone Age. By contrast, the
use of underground caverns (or cavities) formed by solution mining of salt has occurred over
only about the last five to six decades. Broadly, due to distinct inherent properties rock salt
possess, its engineering application goes manifold such as oil and natural gas reservoirs, source
of sulfur, sources of industry salt, underground storage sites for oil and natural gas, and disposal
sites for hazardous waste.

6
(a) Oil and natural gas reservoirs

As the salt in form of dome grows, the cap rock above it is arched upwards. This cap rock can
serve as an oil or natural gas reservoir as illustrated in Figure 1.2. As a dome grows the rocks
that it penetrates are arched upwards along the sides of the dome. This upward arch allows oil
and natural gas to migrate toward the salt dome where it can accumulate in a structural trap.

Fig. 1.2 Accumulation of hydrocarbon near salt dome

The rising salt can also cause faulting. Sometimes these faults allow a permeable rock unit to
be sealed against an impermeable rock unit. This structure can also serve as an oil and gas
reservoir. A single salt dome can have many associated reservoirs at a variety of depths and
locations around the dome.

(b) Source of sulfur

Salt domes are sometimes overlain by a cap rock that contains significant amounts of
elemental sulfur. The sulfur occurs as a crystalline material filling fractures, intergranular
pores, and in some cases replacing the cap rock. The sulfur is thought to have formed
from anhydrite and gypsum associated with the salt by bacterial activity.

Some salt domes have enough sulfur in the cap rock that it can be economically recovered. It
is recovered by drilling a well into the sulfur and pumping superheated water and air down the
well. The superheated water is hot enough to melt the sulfur. The hot air converts the molten
sulfur into a froth that is buoyant enough to rise up a well to the surface.

7
Today most sulfur is produced as a byproduct from crude oil refining and natural gas
processing. The production of sulfur from salt domes is generally not cost competitive with
sulfur produced from oil and natural gas.

(c) Salt production

Salt have been exploited by underground mining. These mines produce salt that is used as a
raw material by the chemical industry and as salt for treating snow-covered highways. Salt are
also extracted by solution mining process. Water is pumped down a well into the salt which
dissolves the salt and is brought back to the surface through production wells. At the surface
the water is evaporated to recover the salt, or the salty water is used in a chemical process.

(d) Underground storage reservoirs

Man-made solution mined salt caverns basically constitute very large underground openings
that provide secure containment for materials that do not dissolve salt. Utilizing the solution
mining (or leaching) technology, caverns are constructed below ground with a minimum
footprint on the surface. The process is done through drilling a well down into the formation
and cycling water through the completed well. The water dissolve and extract the salt from
deposit, leaving a large artificial cavity filled with brine. The hydrocarbons intended to be
stored are then pumped into the cavern thereby displacing the brine. The salt cavern have no
lining and are only confined by the rock salt formation itself.

Storage of both liquids and gases in solution mined salt caverns was reportedly first conceived
in Canada in the early 1940’s, during World War II. Storage in salt caverns of liquid petroleum
gas (LPG), and other “light hydrocarbons” spread rapidly in the early 1950’s in North America
and several European countries. These Light hydrocarbons include propane, butane, ethane,
ethylene, natural gasoline, and other products extracted from refineries and natural gas that are
transported and stored as liquids.

Storage of crude oil reportedly occurred first in England, also in the early 1950’s, during the
“Suez Crisis” and then spread to various other European countries and United States.The
largest underground storage operations in the United States are part of the U.S Department of
Energy’s (DOE’s) Strategic Petroleum Reserve (SPR). The SPR currently stores about 600

8
million barrels of crude oil in 62 caverns located at four sites in Louisiana. A general
arrangement for storage of crude oil in salt cavern is shown in Figure 1.3. The caverns are
designed to be located at a depth ranging from 600m to around 1 km below ground with a
diameter of around 100 m, cavern height of around 150 m and volume of approx. 500,000 m3.

Fig. 1.3 General layout of storage in salt leached cavern

In recent year natural gas has become a prime source of energy worldwide. Once “flared” as a
by-product in oil fields, natural gas was thought to be of short supply in the mid 1970’s. It has
since been discovered in relative abundance, and is considered to be a “clean burning” fuel.
Consequently, most of the recent storage projects in salt caverns have involved natural gas.
Also number of caverns used for crude oil storage in Germany have now been converted to gas
storage.

(e) Waste disposal

Salt is an impermeable rock that has the ability to flow and seal fractures that might develop
within it. For this reason salt have been used as disposal sites for both hazardous and non-
hazardous waste. Caverns in salt as shown in Figure 1.4 have been used as repositories for oil
field drilling waste and other types of hazardous waste in the United States and other countries.
In U.S., the State of Texas alone so far have permitted six salt caverns for disposal of non-
hazardous oilfield wastes (NOW), and one cavern for naturally occurring radioactive materials
(NORM).

9
The potential for disposal of high-level nuclear waste in salt caverns has also been recognized,
however, no site in the United States has received that type of waste. Application of disposal
of hazardous wastes in salt domes in the U.S. and elsewhere have been subject to advance
analyses and detailed examination.

Fig. 1.4 Storage of nuclear waste deposits in nuclear repository tunnel

(f) Unconventional uses

Some unconventional concepts have also been proposed for the use of salt caverns. Some of
the mines developed in salt in the United States and Russia have been carefully sealed and then
serve as a national repositories for government reserves of hydrogen, anhydrous ammonia and
helium gas. Salt is the only type of rock that has a permeability so low that it can hold the tiny
helium atoms.

Compressed air energy storage (CAES) as shown in Figure 1.5 in salt cavern were implemented
in Germany in 1978 and U.S. in 1991. CAES plants are used to meet peak power needs. Storage
of liquid natural gas (LNG) in salt caverns is also appealing, because volume requirements are
reduced by a factor of about 600 from the gas phase. However, cryogenic temperatures are
required, which caused extensive fracturing of salt in an LNG storage test in German mine.

One of the unconventional use of salt formations in the U.S. are for tests of nuclear devices.
Two 5-kiloton devices were detonated in salt as events within the “Plowshare” and “Project
Dribble” programs, in 1961 and 1964 respectively. Three additional tests involving explosions

10
of natural gas were performed in the cavity formed in the Tatum salt stock by Salmon. To date
no leakage of radioactivity has been reported over the Tatum dome, which can be cited as an
example of the exceptional containment properties of salt caverns, given adequate confinement
and salt cover.

Fig. 1.5 Storage of compressed air in salt cavern

1.3 Salt Cavern Storage Technology

Man-made solution mined salt caverns generally constitute very large underground cavities
that provide secure containment for materials that do not dissolve salt (Bays, 1963). Utilizing
the solution mining (or leaching) technology, caverns are constructed below ground with a
minimum footprint on the surface (Richner et al., 1992). The process of leaching is carried
away through drilling a well down into the formation and cycling water through the completed
well. The cycled water dissolves the salt leaving a large artificial cavity filled with brine. The
hydrocarbons intended to be stored are then pumped into the cavern displacing the brine.

These storage caverns do not have lining and are only confined by the rock salt formation itself.
Rock salt is considered intrinsically tight when subject to the overburden pressure of an
overlying rock column of some hundreds of meters thickness.

Cavern construction for the storage of hydrocarbons, compressed air, or hydrogen takes
advantage of the natural sealing properties of rock salt against gaseous media and non-aqueous
liquids. However, the rock salt deposits essentially require certain thickness and depth range to

11
be suitable for cavern construction and storage operation. The intended use of a cavern and the
nature of the salt formation in which it is formed, determine a cavern’s shape and size. Some
caverns in salt domes are very tall and narrow, while some caverns in bedded salt formations
may be short and wide or long cavities.

Salt caverns are engineered according to rock-mechanical specifications. Storage cavern design
consider long-term mechanical stability and tightness along with the realization of the desired
volume and operational pressure (as a function of overburden pressure, i.e. depth). Thus, the
starting point of the cavern design and prime constraint for cavern construction is the ‘rock-
mechanical envelope’, i.e. the maximum theoretical boundary that guarantees the long-term
mechanical stability of the cavern void. The rock-mechanical envelope is not to be exceeded
at any time during the leaching process and cavern lifetime.

To achieve the designed cavern shape, the cavern is developed from the borehole by circulating
water through the leaching tubings as shown in Figure 1.6. These are concentrically installed
into one another to provide different independent flow pathways for water and brine. Two
different flow directions or ‘leaching modes’ are applied that are referred to as ‘direct mode’
(or ‘bottom injection’, i.e. injection through the inner leaching tubing) and ‘reverse mode’ (or
‘top injection’, i.e. injection through the annulus between outer and inner leaching tubing),
respectively. In order to protect the cavern roof from getting dissolved, a fluid, non-aqueous
or gaseous blanket medium which is less dense than brine (nitrogen, crude or diesel oil) is
pumped into the well via the annulus between the outer leaching tubing and the last cemented
casing or wellbore, respectively.

By switching between both leaching modes, changing the setting depth of the leaching tubings
as well as by moving the blanket level upward according to a pre-defined leaching program,
the cavern is shaped within the boundaries of the rock-mechanical envelope. The development
of the cavern is monitored by constantly analyzing the produced brine and by performing
echometric measurements such as acoustic sonar surveys, which provide a three-dimensional
image of the actual cavern shape.

12
 A B

Fig. 1.6 Flow directions of water and brine in direct (A) and reverse leaching mode (B)

Typical storage caverns for liquid or gaseous hydrocarbons in domal salt are 300 to 500 m high
and measure 70 to 90 m in diameter. In bedded salt formations, where the thickness of
individual layers is limited, caverns are designed up to, 100 m in diameter and a designed height
which is governed by the thickness of the formation. Geometrical volumes of storage caverns
may range from 500,000 to almost 1,000,000 m³. The construction time of a cavern of this size
typically is four years.

The caverns are always maintained as filled with liquid as it is never allowed to be emptied to
atmospheric pressure for stability reasons. During oil withdrawal, saturated brine is used for
compensation to avoid that additional cavern volume is not created by further leaching and to
prevent that the caverns grow beyond their rock-mechanically defined boundaries.

Due to the advanced technology, in combination with the material properties of deeply buried
rock salt, salt cavern storage is considered the safest way of storing hydrocarbons. Also
compared to conventional surface and underground rock cavern storages, the underground
storage of crude oil in salt caverns offers significant economic advantages.

13
1.4 References

1 Baar, C. A., (1977) Applied Salt-Rock Mechanics-l The in-situ behaviour of salt rocks .
Elsevier Scientific Publishing Company Amsterdam - Oxford - New York.
2 Borchert, H., Muir, R.O., (1964) Salt Deposits: The Origin, Metamorphism, and
Deformation of Evaporites, 338 pp., D. Van Nostrand Co. Ltd., New York.
3 Bays, C.A., (1963) Use of Salt Solution Cavities for Underground Storage, Symp. On
Salt, Northern Ohio Geol Soc. 564.
4 European Commission, (2004) “Geological Disposal of Radioactive wastes produced by
Nuclear Power, from concept to commissioning”, Euratom, EUR 21224.
5 Fairhurst, C., (2004) “Nuclear waste Disposal and Rock Mechanics: contributions of the
Underground Research Laboratory (URL), Pinawa, Manitoba, Canada”, International
journal of Rock Mechanics and Mining sciences 41, 1221- 1227.
6 Gentry, H.L., (1963) Storage of High Pressure Natural Gas in Underground Salt or Rock
Caverns, Symp. On Salt 604.
7 Gevantman, L. H., (1981) Physical Properties Data for Rock Salt, U.S. Department of
Commerce; National Bureau of Standards monographs.
8 Hsu, K.J., (1972) Origin of Saline Giants: A critical Review after the Discovery of the
Mediterranean Evaporite: Earth Science Reviews 8, 371-396.
9 ISPRL, March (2013) , Detailed Project Report for Phase-II of Strategic Storage
Program for Crude Oil (Volume IV)
10 Linn, J.K. , Culbert, J., (1999) Experience in Underground Storage of Crude Oil in Salt,
Sp. Pub. 90, Geo-Inst. ASCE, 810.
11 Scruton, P.C., (1953) Deposition of Evaporites: American Association of Petroleum
Geologists Bulletin 37, 2498-2512.

14
 2. Geological Formation of Rock Salt

2.1. Rock Salt Deposits

Halite, the chloride of sodium (NaCl), can form rare, pure beds as much as 200 meters thick.
Once deposited and buried by succeeding layers of sediments, halite and associated minerals
are susceptible to a variety of changes brought about by the effects of the elevated temperatures
and pressures that are found at depths. Beside actual changes in mineralogy of the accessory
minerals, the most dramatic changes are due to the lowered viscosity of halite and the increased
contrast between its density and that of the surrounding sediments. Deeply buried salt
commonly forms diapirs (domes or anticlinal folds, the overlying rocks of which have been
bent or ruptured by the squeezing out of the plastic core) 1-6 kilometers in diameter by flowing
plastically from original bedding horizons up thousands of meters through the sediments and
sometimes to the surface.

Some of the characteristics of rock salt deposits change when the rock salt flows into domes.
Domes contain generally purer halite, and rock salts in domes lose most of their primary
sedimentary structures. Fluid inclusions in domed rock salt are more apt to contain foreign
substances, such as oil. Other characteristics of rock salt, specifically porosity and
permeability, do not necessarily change on intrusion. Both porosity and permeability are
minimal in bedded and diapiric rock salt deposits.

2.1.1 Bedded rock salt deposits

The most widely accepted theory to explain formations of bedded rock salt is “bar theory"
originally proposed by Ochsenius. This model in its simplest form hypothesizes an arm of the
sea in which circulation is restricted (by a bar, ridge, or reef across the mouth); water in the
arm is concurrently evaporated and replenished, and its dissolved salt is precipitated into basin-
wide beds as shown in Figure 2.1. A large number of observed primary sedimentary structures
in evaporites may be explained by this model: clay-filled desiccation cracks, ripple marks, and
crossbedding in halite, the presence of preferentially oriented hopper crystals, dissolution
horizons, unconformities, and halite conglomerates and cyclic deposition.

15
A variation on this model that has gained many adherents is Scruton’s hypothesis of constant
water volume in a basin and reflux of final-stage brines to the sea (Figure 2.2). This theory,
which may have a modern analog in the Gulf of Kara Bogaz off the Caspian Sea, involves a
subsea-level restriction that allows partial circulation in and out of the basin. The sea adds
water over the restriction as it is evaporated from the basin. This restricted addition of seawater
creates a salinity gradient; the water in the basin becomes progressively concentrated away
from the sea and precipitates its various components in a lateral continuum of minerals. The
final-stage brines either precipitate their contents at the far end of the basin or are cycled back
out of the basin along the sea floor. This theory accounts for the large-scale lateral facies
variations seen in many evaporite deposits.

Fig. 2.1 Ochsenius bar theory of rock-salt deposition as illustrated by Hsu

Fig. 2.2 Scruton’s reflux hypothesis

16
2.1.2 Salt domes

Besides flowing downward as glaciers, salt can rise upward into overlying rock beds as
buoyant, balloon-shaped bodies often known as salt dome or salt diapir. Figure 2.3 shows cross-
section of rocks of the East Texas Basin between the Oklahoma-Texas border (on the left) and
the Gulf of Mexico coastline (on the right). The purple rock unit is the Middle Jurassic salt, a
rock unit that has the ability to flow under pressure. The salt is overlain by thousands of feet of
sediment which place enormous pressure on the surface of the salt and cause it to flow. At
numerous locations the salt has intruded upwards into overlying sediments. This has produced
small mounds or towering columns of salt that can be thousands of feet tall. The salt columns
and smaller mounds are called "salt domes." The development of salt domes can deform rock
units into traps that hold oil and natural gas. They are often mined as sources of salt and sulphur.
Where conditions allow, salt domes can rise thousands of feet above the layer of salt from
which they began growing thus forming a very large structures. The parent rock units that serve
as a source of salt are usually several hundred to a few thousand feet thick. The salt domes
ascend from depths of between 150 and 2000 metre (or more) below the surface. Among the
largest salt dome deposits are provinces of northern Germany (which actually extend into
Poland, the Netherland and Denmark) and of the US Gulf Coast and Gulf of Mexico.

Domes may be circular to broadly elliptical, and their diameter may either increase or decrease
downwards. Some domes may even have detached from their source bed. Domes commonly
have a mushroom shape near the top, on any or all sides, termed "overhang". The average
diameter at the top of Gulf Coast domes is about 3.2 km, though they range from 1 to 6 km in
diameter . The upper surfaces may be flat, convex upward, or irregular and are usually but not
always covered by "caprock,"

Fig. 2.3 Columns of salt that intrude through overlying sediment units.
17
2.1.3 Caprock

Some salt domes are covered by "caprock," a sequence of a basal, granular anhydrite, an
intermediate layer of gypsum, and a covering of calcite (which in many places contains minable
quantities of native sulfur) as shown in Figure 2.4. Around 181 of the some 330 known salt
domes along the Gulf Coast are known to have caprock and that it is usually thicker on domes
that are closer to the surface. Uncapped domes are also known. The thickness of caprock ranges
from less than 3 m to greater than 500 m and averages 130 m.

The origin of caprock is probably a combination of a solution lag of anhydrite, as the anhydrite-
bearing rock salt of the dome intruded into the freshwater horizon, and an in-place alteration
of the upper zones of that anhydrite to gypsum and calcite. Given the low percentage of
anhydrite in dome rock salt, a large volume of the rock salt must have been dissolved in order
to leave some of the thicker layers of caprock.

Fig. 2.4 Idealized stratigraphic cross section of upper part of an intermediate depth Gulf Coast
salt dome

2.1.4 Salt glacier

In the Zagros Mountains of Iran, salt domes break through the surface to produce flowing
glaciers of salt as shown in Figure 2.5. The arid climate does not produce enough rain to
dissolve the salt and carry it away. Salt features, which in other parts of the world had to be
pieced together during decades of toilsome work in dark mine shafts, are magnificently
displayed in full daylight in the Middle East. Particularly in Iran, hundreds of outcropping salt

18
domes reveal salt tectonics and salt movements. The spectacular agglomeration of exposed salt
domes in Iran and the Persian Gulf area is a unique case in the world and has attracted many
observers." From a rock-mechanical point of view, the numerous salt glaciers, magnificently
displayed in full daylight as shown by air photos are the most spectacular and important
features.

The salt glaciers in Iran are flowing down from salt domes which rose to about 1200 m above
the surrounding plains; although considerable winter rains carry away salt in solution, and the
glaciers remove solid salt, the domes are maintaining their height. As the salt dome mountains
represent the highest elevations in some areas, it may be concluded that such domes are still
rising, the losses by dissolution and glacier flowage being at least compensated by salt flowage
from depth. In some cases, salt flowage from sources at unknown depths apparently is no longer
sufficient to make up for losses at surface. As a result, a large crater is left, with a salt glacier
flowing out of the crater at one side.

Fig. 2.5 Salt glacier flowing out of crater, Iran, The black color is caused by clay minerals
included in the salt along with airborne dust that sticks to the salt.(Image by NASA)

2.2 Rock Salt Distribution

2.2.1 Across the World

Over geologic time, back to billion years of earth’s history, several enormous rock salt deposits
were formed in dry and arid climatic condition all over the world including India when repeated
evaporation occurred in restricted basins of lakes, playas, and seas. However, many inland

19
lakes such as the Great Salt Lake of North America and the Dead Sea between Jordan and Israel
are also locations where rock salt is even forming today. Across the world, the rock salts are
found prominently in United States, Canada, Germany, Poland, Russia, Denmark, France,
Britain, Australia, Iran, Pakistan, Mexico, Italy, Portugal, Spain, Switzerland and many more
as shown in Figure 2.6 in the form of extensive beds, or in domes and plugs. They are known
to have existed in almost every geological period and are compiled in Table 2.1. Among the
largest salt deposits worldwide are the salt dome provinces of northern Germany (which
actually extend into Poland, the Netherlands, and Denmark) and of the US Gulf Coast and Gulf
of Mexico. Both are being intensively used for cavern construction either for salt production
or for storage. More than 1,000 salt caverns have been intentionally created in these salt
deposits in the United States.

Fig. 2.6 Worlds’ salt distribution

20
 Table 2.1 Salt Deposits of Every Geological Period

Geological Period Location of Salt Deposits

Cenozoic Era Cenozoic Era

a. Recent a. Solar salt operations, Playa Lakes
b. Pleistocene b. California, Nevada, Russia, Mexico, Israel
c. Pliocene c. Nevada, Utah, Italy, Jordan
d. Miocene d. Dominican Republic, Cyprus, Czechoslovakia, Poland, Spain.
Turkey, Russia, Algeria, Egypt, Sudan, Iran, Iraq, Syria, Trucial
States
e. Oligocene e. France, Germany, Spain, Turkey, Iran, Iraq
f. Eocene f. Green River Basin, Wyoming, British, Honduras, Pakistan (?)
g. Paleocene g. There are no known Paleocene salt deposits
Mesozoic Era
a. Cretaceous a. Florida, BoliVia, Brazil, Colombia, Russia, Angola, Congo
Republic (Kinshasa), Congo Republic (Leopoldville), Gabon,
Morocco
b. Jurassic b. Gulf Coast Area, Idaho, Cuba, Chile.Germany. Aden. Kuwait
c. Triassic c. Isthmus of Tehuantepec-Mexico, Bulgaria (?), France, Greece (?),
Germany. Netherlands. Portugal, Spain. Switzerland, United
Kingdom, Algeria. Libya, Morocco, Tunisia
Paleozoic Era
a. Permian a. Permian Basin - U.S., Supai Basin,Williston Basin, Mexico,
Peru, Germany, Greece (?), Netherlands, Poland, United
Kingdom, Russia, Denmark, Australia (?)
b. Pennsylvanian b. Colorado, Paradox Basin,Brazil
c. Mississippian c. Viriginia, Williston Basin, New Brunswick, Nova Scotia
d. Devonian d. Williston Basin, Russia. Australia
e. Silurian. e. Salina Basin - U.S., and Canada
f. Ordovician f. Williston Basin
g. Cambrian g. Northwest, Territories, Australia, Russia, Iran, Pakistan (?)
Precambrian Iran, Pakistan (?), Australia

21
2.2.2 Rock salt deposits in India

In India, known salt deposits are restricted to a small but highly tectonized Lower Shali
Formation (Lower Precambrian) exposed around Guma, District Mandi, and the extensive
bedded formation of marine evaporates in north-western Rajasthan of the so-called Nagaur-
Ganganagar Basin.

(a) Bedded salt deposits of Mandi, Himachal Pradesh

The Mandi salt deposits occur in a zone of limestone, shale, and sandstone believed by some
to belong to the Krol Series and by others to the Subathu (Nummultic) group. Three salt plugs
are known i.e. Drang (21°49'-77°1'), Guma (31°58'-76°55 '), and Megal. The salt is of a dark
purplish hue and quite opaque. It contains a large admixture of earthy impurities. Occasionally,
small bodies of pure crystalline salt are found. Details such as depth, extent, etc., however, of
the individual salt deposits are not known. The salt reserves are estimated at 100 million tons.
However, the mines was closed in 2008 after the state government detected several lapses on
account of environmental issues and financial overdue to the state own enterprises Hindustan
Salts Limited.

(b) Bedded salt deposits of Bikaner, Rajasthan

With a surface area of approx. 50,000 km², the salt basin in India covers much of north-western
Rajasthan state and extends from near the city of Jodhpur in the South into Haryana and Punjab
states in the North. To the northwest, the Nagaur - Ganganagar Basin is interpreted to be
connected to the evaporite sequences of the Salt Range in Pakistan. .In the context of regional
geology, the halogenic deposits in Iran, Oman and Pakistan are considered to form a single
basin situated on the south western Asiatic continent at the time of formation, but moved to its
present geographic location due to plate tectonic movements. The Nagaur-Ganganagar basin is
considered the southmost continuation of this large Eocambrian (approx. 600 million years)
basin of which Salt Range of Pakistan forms the northern part. The northern extremity of the
Eocambrian basin is exposed in the Salt Rage (Pakistan) as Saline series. The sequence denotes
continuity in deposition of sediments from Late Proterozoic to Lower Cambrian age, the Saline
series has been assigned to the Eocambrain age.

22
The Nagaur-Ganganagar Basin is situated on the western part of the Indian Craton (Vindhyan
Platform). The shallow basin developed in late Proterozoic to Palaeozoic times in the western
foreland of the early to middle Proterozoic Aravalli-Delhi Fold Belt. Based on geophysical,
surface and subsurface geological data, it is known that the basin fill shows on lap onto the
Precambrian basement rocks of the Malani Igneous Suite and the Delhi Metamorphites to the
south. The eastern basin margin is overprinted by the NNE-SSW trending Sardasahar-Bidasar
Fault Zone as shown in Figure 2.7.

Fig. 2.7 General outline of the Nagaur-Ganganagar Basin

23
2.3 References

1 Baar, C. A., (1977) Applied Salt-Rock Mechanics-l The in-situ behaviour of salt rocks .
Elsevier Scientific Publishing Company Amsterdam - Oxford - New York.
2 Borchert, H., Muir, R.O., (1964) Salt Deposits: The Origin, Metamorphism, and
Deformation of Evaporites, 338 pp., D. Van Nostrand Co. Ltd., New York.
3 Dasgupta, S.P., Virendra, K., Ramchandra, Jha, M.S., (1988) A frame work of the
Nagaur-Bikaner evaporite basin, Rajasthan, Indain Mineral Vol. 42(1): 57-64.
4 Dubey, R.K., (1996) Rheological behaviour of rocks of lesser Himalaya around Mandi
and Mussoorie areas, Ph.D. thesis, Department of Mining Engineering, B.H.U., 100.
5 Geological Survey of India (2005): Special Publication No.62, Geology and evolution
of Nagaur Ganganagar basin with special reference to salt and potash mineralisation, S.
P. Rastogi, Dy. D.G. (Retd.).
6 Gevantman, L. H., (1981) Physical Properties Data for Rock Salt, U.S. Department of
Commerce; National Bureau of Standards monographs.
7 Hsu, K.J., (1972) Origin of Saline Giants: A critical Review after the Discovery of the
Mediterranean Evaporite: Earth Science Reviews 8, 371-396.
8 Lefond, S.J., (1969) Handbook of World Salt Resources. Plenum Press, New York,
404pp.
9 Mazumdar & Bhattacharya (2004) Stable isotopic study of late Neoproterozoic -early
Cambrian (?) sediments from Nagaur- Ganganagarbasin, western India: Possible
signatures of global and regional C-isotopicevents - Geochemical Journal Vol.38 (2).
10 Ode, H., (1968) Review of Mechanical Properties of Salt Relating to Salt Dome Genesis:
Geological Society of America Special Paper 88, 544-595.
11 Rastogi, S.P., (2005), Geology and evolution of Nagaur-Ganganagar basin with special
reference to salt and potash mineralisation, Geological Survey of Indai, Spec Pub no. 62
12 Scruton, P.C., (1953) Deposition of Evaporites: American Association of Petroleum
Geologists Bulletin 37, 2498-2512.

24
 3. Investigation

3.1 Stages of Investigation

From project conception through feasibility studies to the basic design stage and throughout
the construction phase, geotechnical investigations for the rock salt cavern are designed to
provide the level of information appropriate to the particular project development stage. The
investigation program is to prepare to collate, generate and analyse engineering geological,
hydrogeological and rock mechanics data for site characterization in different stages of
investigation. Broadly the investigation program for storage caverns are carried out in four
stages:

Stage I: Pre-Feasibility Stage

Stage II: Detailed Feasibility Report
Stage III: Supplementary Investigation during Detailed Engineering
Stage IV: Investigations during Construction

(a) Pre-Feasibility Stage or Desk Top Studies

Sufficient precise and detailed geological information and maps are the basic pre-requisite
during the pre-feasibility or desk top studies stage. In principle, salt caverns is constructed
wherever the geological prerequisite are adequate i.e. the first-hand information about the
geological setup, structural features and lithological characteristics of evaporite basin floor of
the rock salt deposits are available and suitable hydrogeological condition which affect the
availability of ground water for cavern leaching be present. Primarily, suitable salt reservoir
(layered or dome) of sufficient thickness and aerial extent are necessary to solution mine the
projected numbers and the size of caverns.

Following pre-requisite to be assessed and analysed before taking up the project:

(i) Salt selection of sufficient thickness and purity that can be rendered impermeable by
solution process
(ii) Satisfactory roof or cap rock condition
(iii) Sufficient depth to permit confinement and suitable selection between cavity and the
surface

25
(iv) Tectonic setting of the storage horizon, regarding major faults of regional extend, minor
faults as well as smaller scale joints with non-salt layers
(v) Adequate surface provisions and resources of water supply, brine disposal and storage
(vi) Distance to populated areas as the cavern site should be located at a safe distance from
the populated areas
(vii) Proximately to other crude pipelines and refineries
(viii) Proximity to the other industrial facilities
(ix) Current and future use of adjacent properties that may withdraw large amount of ground
water and potentially increase subsidence rate
(x) Proximity to environmentally sensitive wetlands, streams and drinking water acquirers
(xi) Proximity to salt boundary
(xii) Proximity to other active or abandoned subsurface activities
(xiii) Climate
(xiv) Hazardous Safety against earthequake
(xv) Scope for enlargement of facilities for future

Final selection of the site is done after the area is detailed with the subsurface geological map
of individual salt layers, seismic mapping and exploratory drilling as well as the suitable geo
mechanical properties of rock salt formations.

(b) Stage II: Detailed Feasibility Report

Investigations during feasibility studies is designed to provide information at a level such that
critical geotechnical features of sites may be compared in the basic engineering stage. The
investigations should be sufficiently complete to permit selection of the most favourable site
areas within the regional physical setting, determine the general layout of structures best suited
to the site conditions, evaluate the influence of hydrogeology on site design and construction,
assess the geotechnical aspects of environmental impact, and to ascertain the costs of
developing the various project plans in sufficient detail to allow comparative cost estimates to
be developed.

On the basis of available information and under consideration of the various criteria discussed
during the pre-feasibility report, detail investigation are performed. These investigations

26
generally have three aspects i.e. development of regional geology, geotechnical assessment and
hydro geological exploration and is obtained with the following information/ database:

(i) Inter-agency coordination to obtain geological condition & history, ground water
studies, hydrological and soil data, insight into the geological hazards, seismicity and
prior regional experience etc.
(ii) Subsurface exploration well drilled up to the depth of around 1000m from surface to
determine the detailed geological framework of evaporite deposits.
(iii) Wireline well logging programme for the information on the petro physical rock
properties along the well path and further geological interpretation.
(iv) Seismic refraction survey to examine the thickness variation and lateral extent of the
existing evaporate group
(v) Physical and geo mechanical properties investigation of rock salt through laboratory
testing to determine the cavern dimensions, surface subsidence, cavern abandonment
etc.
(vi) Leaching velocity or leaching rate tests for the detailed leaching programme.
(vii) Regional hydro geological studies to specify the hydrogeological parameters and to
determine the salinity of the ground water
(viii) Estimation water quality parameters of ground water, rainfall data, source of water
nearby & its quality.
(ix) Collection regional data of temperature, rainfall, relative humidity, wind velocity etc,
for the calculation of evaporation rates.

(c) Stage III: Supplementary Investigation during Detailed Engineering

Prior to detailed engineering and construction, supplementary investigations are performed to

obtain more accurate information about the natural conditions in the project area which will
help to optimise details of the project design, especially the sizing and number of the ground
water wells and the evaporation basins.

Investigations during basic engineering stage is obtained with the following information/
database:

27
(i) Aquifer tests for the aactual hydrogeological conditions to conclude the number of
required water wells and final water well layout
(ii) Evaporation test to obtain more details information about the potential evaporation in
the project area
(iii) Topography survey for optimization of earthwork during construction of the facilities
especially the evaporation basins.

(d) Stage IV: Investigations during construction

(i) An echo-metric measurement (sonar survey) for the three-dimensional image of the
actual cavern design
(ii) Well coring – logging, sampling and testing during well drilling

3.2 Geological Investigation

The geological studies such as surface geological mapping, core logging, sampling etc are the
most vital as they form the cornerstone for the geological exploration. These studies gives an
insight into not only the lithology but also intricate details regarding the sedimentological
gradations, structural attitude of the formation, dislocations suffered by them, the physical
characteristics for the rock salt formation etc.

3.2.1 Geological Framework of the project site

(a) Preliminary data

Information regarding geology and geometry of the rock salt deposits including the overburden
strata of the project site is the prime intention to realise a strategic petroleum storage facility in
salt caverns. Geological information on these deposits can be derived from the dataset both
published and unpublished and should essentially consist of:

(i) Lithological description of the existing borehole sections (from available literature) as
well as exploration wells drilled in the proposed project area illustrating geological
succession within a tabular unit, layer or stratum, characteristics of the evaporite
deposits, major structures and their attributes such as dip & strike of beds,
discontinuities etc.

28
(ii) Results of seismic survey explored for hydrocarbons in deeper strata near the project
site (if available) and the seismic reflection survey performed in the proposed project
site to provide information on the regional bedding characteristics of the evaporite
basins for their thickness variation and lateral extent.

(b) Types of salt formations and its characteristics

Salt formation are either bedded sedimentary formations or intrusive salt formations (including
domes and ridges). Bedded formation often underline large basins, with relatively thin salt beds
(less than about 300m thick) often separated by beds of porous or impervious layers of other
sediments while domal salt formation intrude from roots in deeply buried salt beds and tend to
be plugs a mile or more in diameter with great depth.

Knowledge of the physical location and dimensions of the salts bed and any nonsalt layers that
might become part of the cavern are critical to design work. In bedded salt, knowledge of the
depth and thickness of the overburden is often required. Domal salt geometry is rather more
difficult to define, as the flanks may be very irregular and even overhung and the caprock
poorly defined.

(c) Faulting structural, anomalies and overburden structure

Faults in the surrounding formations that intersects the salt formation are noted and recognized
as areas to avoid intersecting with a well bore. Where possible, anomalies in the salt formation,
insoluble layers or intrusions or any permeable layers are identified and essentially taken into
consideration in the design.

Particularly in the development of bedded salt caverns, information on the thickness and
strength of the overburden rack layers is required because these layers are part of the structural
support for the cavern. Also any water bearing strata must be identified and the drilling plan
developed to ensure against introducing cross communication between water strata and
between the water strata and the salt formation.

29
3.3 Core Drilling

A quality drilling partly non-coring and partly coring of a depth of about 1000m or even more
can provide an accurate and comprehensive record of the lithology and stratigraphy of
evaporite sequence and other relevant information obtained during drilling, sampling, and
testing. Also the sample collected from core drilling is used for laboratory testing. Therefore,
a major part of the geotechnical investigations is the compilation of accurate borehole logs on
which subsequent geologic and geotechnical information and decisions are based.

In view of deep drilling and varied nature of formations like unconsolidated sediments at the
top, compact and fissile formation in the middle and soluble and hard formations in the
evaporite sequence of rocks, a comprehensive specified programme of drilling and casing
lowering needs to be planned. Bentonite powder of sodium base of specific gravity of 1.04 to
1.05 is used for circulating fluid while drilling in the top unconsolidated sediments and hard &
compact fissile formation. Special circulation fluid such as saturated brine by using common
salt and other chemical additives like organic polymer, starch and caustic soda etc of specific
gravity of 1.20 to 1.25 are used for drilling in evaporite sequence. The borehole will be of
“slim hole” type and coring is done using the wireline coring technique. The hole is completed
with casing subsequent to wireline logging. Normally a PDC type coring bit is used that provide
best quality of cores when drilling halite (having mohr hardness of 2.0 to 2.5) and other
relatively soft sedimentary rock. These rocks are not effectively drilled by a surface-set
diamond bit. As the succession is consolidated and compacted throughout, coring is preferably
be done with a double tube core barrel. However, if it turns out upon retrieval of the first core(s)
that recovery is insufficient, coring is continued with a triple tube core barrel. For that purpose,
a triple tube set is to be available at the site form immediate use.

The major drilling accessories required are:

 Coring drills equipped with wireline systems of various capacity to drill around 1500m.
 Dill machine of different capacity is selected according to the borehole depth and the
strata conditions
 Hook load capacity of the drill may be of around 11500 KG for drilling at higher depth
 HQ/HCQ & NQ/NCQ Drilling rods
 HQ double tube / HQ triple tube core barrels

30
 HS diamonds bits and reamer shells
 SX and PX casings
 Diverter
 Fishing tools and Mud tanks

All the drilling drills are equipped with matching tubular and other necessary accessories. The
drills also needs to be equipped with compressor of up to 150 psi/ 600 cfm capacity. The deeper
capacity drills are equipped with wireline system and most of them are mounted on truck.
Multishot borehole survey cameras are required to conduct deviation survey. A typical drill
well log is illustrated with a probable drilling and casing specification is shown in Figure 3.1.

Hole Section Casing

6 ¾” (171.4mm) PX size (144 /

(Water base sodium bentonite drilling fluid) 125.33mm)

5 5/8’’ (142.8mm)
HX size ( 114/100mm)
(Water base sodium bentonite drilling fluid)

Halite H6

36/8” (95.25mm) cored yielding 2.5”

i.e. 63.5mm core sample -
(Saturated sodium chloride brine fluid)

Fig. 3.1 A typical forecast log for exploration well

3.4 Geological logging of cores

On retrieval of the core after drilling, they are kept in the core boxes as per the standard.
Representative samples is collected for the purpose of correlation to mapped formations and
for subsequent laboratory analysis such as petrographic analysis. Field notations include the
locations and brief descriptions of samples with a unique sample number. The recovered
lengths are measured to arrive at the recovery percentage and the details are recorded in the
form of:

31
(i) Lithology,
(ii) Grain size,
(iii) Colour,
(iv) Compositional bands,
(v) Macroscopic identification,
(vi) Structural details including joints, fractures, slicken sides, their intensity, attitude of
beds etc.
(vii) Intensity of mineralization, type of ore, distribution of the mineral constituents etc.

3.5 Geophysical Investigation

Geophysical investigation consists of making indirect measurements from the earth's surface
or in boreholes to obtain subsurface information. Geophysical investigations are carried out to
supplement borehole and outcrop information and to aid in geological interpretation /
inferences. The campaign includes 2D Seismic Reflection Survey as well as wireline well
logging program.

The seismic refraction survey along longitudinal and cross lines will help to delineate different
subsurface stratigraphic units and assess their thickness, nature and contacts. The geophysical
bore hole or well logging will help to correlate and calibrate geophysical data with geological
information.

3.5.1 2D Seismic Reflection Survey

2D seismic reflection survey is carried to examine the fundamental architecture of the evaporite
succession on the project area. In particular, the seismic survey will:

(i) Determine the depth, thickness and regional thickness of the evaporite sequence
(ii) The existence of faults (if any)
(iii) Provide detailed stratigraphical and structural information to locate cavern well sites

Layout of the survey is designed to provide optimum coverage of the area of project and can
be as large as the extent of around 70-80 ground line kilometres (GKL), covering an area of
around 250km2.

32
Prior to the main survey, a “short spread refraction survey” is carried out to identify the first
reflector i.e. the weathered layer. The seismic velocity of the first reflector needs to be known
for the later static correction. This survey is performed in a spread length of around 200m with
2 measurements per km along the planned line.

The acquisition parameter such as number of channels, group interval, geophones per group,
shot point interval, coverage, spread length, CDP interval, energy source etc for the reflection
survey are chosen to provide a depth of penetration of around 1500m and an adequate degree
of resolution of the subsurface geology in both vertical and horizontal dimensions. Special
attention is given to the sweep parameters i.e. its bandwidth and length, which have direct
influence on vertical resolution. A field test is run prior to the survey to identify the most
suitable sweep length and bandwidth. An interpreted section of a typical seismic line of 2D
seismic reflection survey along with probable identified fault are shown in Figure 3.2.

H7 top

H6 top Possible fault

H6 Base

Jodhpur top

Fig. 3.2 A typical interpreted section of 2D seismological survey

3.5.2 Wireline well logging programme

Wireline well surveying provides valuable direct or indirect information in the petrophysical
rock properties along the well path. The combination of different logging tools has proven to
provide the best data pool for further geological interpretation.

33
All electro logging tools are calibrated for good response in halite before starting the runs. Pre
and post run calibration checks and repeat section are also carried out. All logs are run in open
hole, filled with conductive drilling fluid. Sondes (logs) are run in the following order to make
sure that the most important data is gathered first in case the borehole collapses due to multiple
runs:
(i) ELTS sonde for short and long normal resistivity, single point resistance, self-
potential, natural gamma and temperature log
(ii) FDG sonde for long and short spaced neutron and porosity log
(iii) FWS sonde for compensated sonic log
(iv) DNN sonde for long spacing density/compensated density, high resolution density,
natural gamma and calliper logs

3.6 Geotechnical Investigation

Geotechnical investigation to have information about the in-situ stress state and mechanical
properties of the salt are indispensably required for the proper design of salt cavern. In the case
of high purity products, knowledge of the entrained containments is also required. While stress
state and mechanical properties of salt formation are assumed for design, core samples are
taken in a manner to preserve in-situ condition and is analysed and compared with the design
assumptions, where possible.

3.6.1 Laboratory Testing

(a) Test and Sample Selection

The physical and mechanical characteristics of rock salt are evidenced in its compact texture,
low porosity, low permeability and strong rheological properties (creep behavior). These
properties are essential for the stability analysis and the design of Salt cavern and therefore
particular attention to be paid to the execution of tests on rock specimens in the laboratory.

Within the project requirements, a suitable suite of index and geo-mechanical property tests is
planned both in the vertical as well as lateral direction. Sample for the laboratory test from the
core box is selected so as to be representative of the envisaged cavern depth interval with
special focus on the load bearing strata above and below. As the evaporite succession is the
focus on interest and cores of maximum quality are required to obtain representative lab tets

34
results, coring has to be performed with utmost care. Preparation of rock samples for laboratory
testing is done be as per BIS 9179.

(b) Index and Classification Test

Types of index and classification tests that are typically required for rock salt are listed below:

(i) Specific Gravity

(ii) Bulk Density(Wet/Dry)
(iii) Water Absorption
(iv) Water Content
(v) Porosity
(vi) Sonic Velocity test
(vii) XRD and SEM analysis

(c) Engineering Properties Test

The engineering geo-mechanical properties of the cavern host rock in combination with depth
and thickness of geological formation are very important for the design of storage cavern.
Typical rock mechanical material properties required to be investigated are:

(i) Stress-strain behaviour

(ii) Strength (short term and long term)
(iii) Bond between layers
(iv) Creep behaviour of rock salt.

Short term tests are performed to assess the stress-strain behaviour and the strength of rock salt
while the long term test are performed to assess the visco-plastic creep behaviour. Long term
deformation tests are particularly applied to rock salt which is prone to visco-plastic
deformation. Both long term and short term tests are performed by applying either uni-axial or
tri-axial load on the core samples.

Tests required to be performed in laboratory for to get geo-mechanical properties are:

(i) Point load test
(ii) Brazilian test (Tensile strength)

35
(iii) Direct shear test
(iv) Unconfined uniaxial compressive test
(v) Triaxial test
(vi) Creep test under uniaxial and triaxial compression

Testing condition for the short term and long term behaviour of rock salt shall be of the
following:
(i) Execution at in-site temperature
(ii) Constant axial load (correction of axial force depending on axial strain)
(iii) Online measurement of axial and confining pressure, axial deformation and temperature

(d) Leaching test

Leaching behaviour of the salt rock of the storage formation is crucial to be known as this
feature determines the shape development of the cavern and the timing of the solution mining
process. The leaching test results are important basic input data for the numerical leaching
simulation. As the leaching rate at the cavern wall is different to the rate at the roof, it is required
to perform separate vertical and horizontal leaching tests. Thus each sample taken from a
selected depth/lithology must comprise of two pieces, one for vertical and one for horizontal
leaching tests.

(e) Determination of bulk content of insolubles

The bulk content of insolubles is one of the most important properties of the rock salt mass that
has to be known when the feasibility of leaching a storage cavern is evaluated. The insolubles
content has a strong influence on the development of the net cavern volume as well as its shape
if it alternates significantly between different strata. Once determined, it is used as an important
input-parameter for leaching simulation and has a significant influence on the technical design
of the cavern operations.

(f) Sieve analysis

As a standard procedure in sedimentary petrography and for hydrogeological assessment, sieve

analyses have to be carried out on material from the interval of the aquifer with the overburden
sequence as determined from interpretation of geophysical logging and drilling data.

36
(g) Additional Analyses

Should the core description, contrary to the expectation, reveals the presence of larger amount
of impurities, supporting analyses which could basically comprise additional qualitative visual
inspection under the microscope or mineralogical analyses such as XRF could be performed.

Additionally, chemical determination of the bromine content in halite provides an important

tool to identify potash salt layers and may also help in stratigraphic correlation if unexpected
structures with the salt sequence aggravate geological interpretation and correlation between
wells.

3.7 Hydro-geological Investigation

The purpose of hydrogeological investigation is to collect basic hydrogeological parameters

for the natural saline groundwater from the phreatic aquifer that is available for the purpose of
the project as well as the quality and availability of ground water in the immediate vicinity of
the proposed project sites for cavern leaching. Approximately eight volume of water is
required to leach one volume of salt therefore resulting in the requirement of a huge quantity
of water for the leaching operation that has to be extracted from the ground water. For this
reason, a detailed knowledge about the quality, quantity and yield of groundwater in the project
area is required.

Data on the regional hydrogeological conditions such as source of available water near the
project area (dug well, tube wells, canal etc), pumping tests performed (if any), quality of water
etc, needs to be collected from the authorized agencies and accordingly hydrogeological tests
are planned. Several hydrogeological tests i.e. borehole logging, chemical analysis of ground
water samples, flow meter survey and pumping test (step drawdown and well performance
tests) are conducted in comparatively shallow test well of around 300m.

The hydrogeological interpretation will specify the following parameters:

(i) depth and thickness of single aquifers and confining beds,
(ii) groundwater levels within the area,
(iii) groundwater flow direction,
(iv) aquifer parameters (hydraulic conductivity, transmissivity, storage capacity, etc.),
(v) extent of the cone of depression developing at specific (maximum) discharge rates,
(vi) well production capabilities (capacity, specific yield, etc.), and
(vii) groundwater quality,

37
Based on these results, especially the design of the wells, the depths of the filter screens, the
withdrawal rate per well and the distance between the wells.

3.7.1 Long duration pumping tests

Pumping test are conducted to determine the performance characteristics of the well and the
hydraulic parameters of the aquifer.

For a well performance tests, yield and drawdown are recorded so that the specific capacity
can be calculated. These data, taken under control conditions, gives a measure of the productive
capacity of the completed well and also provide information needed for the selection of
pumping equipment. The second purpose of pumping tests is to provide data from which the
principal factors aquifer performance-transmissivity and storage co-efficient can be calculated.

There are two types of pump tests i.e. constant rate tests and step-drawdown tests. In the
constant rate tests, the well is pumped for a specific length of time at one rate, whereas in a
step drawdown test the well is pumped out successively greater discharges for relatively short
periods. Data from both types of tests are analysed to determine the important hydraulic
characteristics of an aquifer and the well.

Pumping test consists of pumping a well at a certain rate until a static water level is reached
and recording the drawdown in the pumping well and in nearby observation wells at specific
times. The duration of such a stationary pump test depends on the discharge rate and the
characteristics of the aquifer, but usually is a few days. After having stopped the pumps, the
increase of the water level is measured. Values like pumping rates, drawdown of water level,
pH- value, electric conductivity, water temperature, sand content, turbidity, colour, and
chemistry of groundwater etc is recorded. Also measurements required for pumping tests
include the static water level just before the test is started, time since the pump started, pumping
rate, pumping level or dynamics water levels at various intervals during pumping, time of any
change in discharge rate and time the pump stopped. Measurement of water levels after the
pump is stopped are extremely valuable in verifying the aquifer coefficient calculated during
the pumping phase of the test.

38
3.7.2 Water quality tests

The quality of the groundwater, which is not usable as drinking water and only partially usable
for irrigation, is proven to be suitable for the leaching process. The TDS and the electrical
conductivity (EC) are related indicators for the water quality. TDS values describe the amount
of all solids (usually mineral salts) that are dissolved in water (evaporation residue). While EC
measures the potential for an electrical current to be transported through water. In many cases,
conductivity is linked directly to the total dissolved solids (T.D.S.).

3.8 In-situ testing During Construction

Direct inspection of a cavern is impossible, but sonar surveys allow the shape of a cavern to be
measured. The cavern volume and shape during construction is regularly investigated by sonar
measurements using specialized sonic also operated by a wireline unit. The method is based on
travel-time measurements which can be converted into distance when the acoustic velocity in
the medium (usually brine) is known. The cavern shape is surveyed by a multitude of
horizontal sections over the entire depth range of the cavern as well as by sections with tilted
probe head to measure bottom, roof or any irregularities in cavern shape. The resulting digital
data set of 3-dimensional co-ordinates can be visualized applying special computer software as
shown in Figure 3.3 and is carried out by specialist agencies.

Fig. 3.3 Cavern shape visualized special computer software

39
3.9 References

1 ASTM Codes for Investigation: Annexure 1

2 BIS Codes for Investigation: Annexure 2

3 Engineers India Limited Standards : Annexure 3

4 ISRM Standard : Annexure 4

5 SMRI Publication : Annexure 5

DEEP Underground Engineering GmbH (2009): Feasibility and Basic Engineering

6
Report: Strategic Petroleum Storage in Salt Caverns in NW Rajasthan, India.

Engineers India Limited (2013): Detailed project Report for Phase II of Strategic
7
Storage Program for Crude oil, Volume IV - Bikaner, Rajasthan, India.

Geological Survey of India (2005): Special Publication No.62, Geology and evolution
8
of Nagaur Ganganagar basin with special reference to salt and potash mineralisation

Rath, R., Wippich, M., Nanda, A., Wilke, H.F. (2014). Underground salt cavern
9
storages for crude oil at Bikaner, India. SMRI Tech Conf Proc. Netherlands; 124-135.

Annexure 1: ASTM Standards

S. No. Code No. Description

1 D 420 Guide to site character for site and design construction

2 D4630 Determining Transmissivity and Storavity of Low Permeability Rock by In-Situ

Measurements Using Constant Head Injection Test

3 D4631 Determining Transmissivity and Storavity of Low Permeability Rock by In-Situ

Measurements Using the Pressure Pulse Technique

4 D 4050 Standard Test method for withdrawal and injection well tests for determining
hydraulic properties of aquifer systems.

40
 Annexure 2: BIS Codes

S.No. Code No. Part/ Section Description

1 2131 Method for Standard Penetration test for soils

2 2132 Code of practice for thin walled tube sampling of soil

3 4078 Code of Practice for indexing and storage of drill cores

4 5313 Guide for core drilling observation

5 5529 Part 1 & 2 Code of Practice for In-situ Permeability tests.

6 6926 Diamond core drilling-site investigation for river valley

projects-Code of Practice

7 7022 Guide for geophysical logging of bore holes for hydro

geological purposes

8 8764 Method of determination of point load strength index of rocks.

9 9143 Method for the determination of unconfined compressive

strength of rock material

10 9179 Method for the preparation of rock specimen for laboratory

testing

11 9221 Method for the determination of modulus of elasticity and

poisons ratio of rock materials in uniaxial compression

12 10082 Method of test for determination of tensile strength by indirect

tests on rock specimen

13 10208 Diamond core drilling equipment

14 10782 Method for laboratory determination of dynamic modulus of

rock core specimen

15 11315 Part 1 to 12 Method for the quantitative description of discontinuities in

rock mass

16 13030 Method of test for laboratory determination of water content,

porosity, density and related properties of rock material

17 13047 Method for determination of strength of rock materials in

triaxial compression

18 14436 Guidelines on determination of resistivity of rock specimen

41
 Annexure 3: Engineers India Limited

1 6-73-0009 Standard specification for Site Investigation for Underground Rock Cavern

2 6-60-0010 Standard Specification for Onshore Geophysical Investigations

3 OISD – STD – 181 Geophysical Operations

Annexure 4: ISRM Standard

1 Suggested Methods for Determining the Creep Characteristics of Rock, Rock Mech Rock Eng (2014)
47: 275–290.

2 R. Ulusay, The ISRM Suggested Methods for Rock Characterization, Testing and Monitoring:
2007–2014, Springer Publication, 292 pages

Annexure 5 : SMRI Publication

1 Joe L. Ratigan, (2005), Lithology, In Situ Stress and Temperature, and Structural Geology of Salt
Deposits Important for Rock Mechanics

2 Kirby Mellegard , (2005), Laboratory Testing of Salt and Non- Salt Units, Strength and
Deformation Behavior

3 Pierre Bérest, (2005), In situ testing of Salt Caverns

4 Gary D. Callahan, (2006) Salt Dilation Testing

5 Joe Ratigan,(2006) Creep Testing of Salt

6 Christina Beckmann and Max Wippich , (2012), Basic Requirements in Geology for Cavern
Construction

7 Jeff Langlinais, (2012), Measurements, Most Common Practices in Logging, MITs, Coring, and
Lab Tests for Cavern Development

8 J.W. Bowcom, Rotary Drilling of Large Diameter Vertical Holes

9 Klaus Buschbom, (2012), Basic Requirements in Drilling and Completions for a Cavern Well
Construction

42
 4 Geo-mechanical Properties of Rock Salt

4.1 Physical Properties

Rock salt is dominantly halite (NaCl), but other minerals are also present such as calcite,
dolomite, anhydrite, gypsum, and potash minerals. (Potash is an industrial name for the
complex potassium and magnesium evaporites). These minerals may be deposited
contemporaneously in adjacent parts of the same basin. The mineralogy is dependent on
variations in topography, water depth, salinity, and other factors. Because these factors can
change during geologically shorter periods of time, beds of different evaporite minerals overlie
and overlap each other in sequences reflecting the changes in equilibrium that occur locally
during the subsidence and sedimentary filling of basins. Some of the most common minerals
found in evaporite deposits are given below in Table 4.1.

Table 4.1 Common Evaporite Minerals Found Along With Rock Salt

Mineral Chemical Formula

Halite NaCl
Sylvite KCl
Carnallite KMgCl3. 6H2O
Bischofite MgCl2. 6H2O
Gypsum CaSO4. 2H2O
Anhydrite CaSO4
Polyhalite 2CaSO4. MgSO4. K2SO4. 2 H2O

Although carbonates, calcites and dolomites are found associated with rock salt deposits, they
are not considered as evaporites. The term “salt rocks” is used for all the suite of rocks found
in evaporites. However, the term “Rock salt” is used for pure halite (NaCl) - the monomineralic
rock. Sylvite and Carnallite are called potash minerals.

Various chemical studies have been performed on the Rock salt taken form Guma, District
Mandi, (Indian Mineral Yearbook: Salt, Hindustan Salt Limited 20011), as well as Khewra salt
mines, Pakistan. Results are shown in Table 4.2. The results shows that the Nacl content in Salt

43
found near Mandi are far less than the one found in Khewra Salt mine and also the water
insoluble materials in salt near Mandi are distinguishably high varying from 18 to 30% as
compared to only 0.5% in salt from Khewra mines. The salt core sample taken from the
exploration well near Bikaner (ONGC exploration report on Rock Salt, 2008), the insoluble
content show a wide range close to an average of 8.5% over the depth of 600-750 m depth.

Table 4.2 Chemical Composition of Rock Salt

S.No. Characteristic Mandi, Himachal Pradesh Khwera Mines,

Pakistan

1. Sodium Chloride (Nacl) 67.81 78.84 98.30 98.10

2. Water Insoluble 31.23 18.92 0.40 0.50
3. Calcium Bicarbonates 0.35 - 0.30 0.25
4. Magnesium Sulphate 0.04 - 0.12 0.12
5 Magnesium Chloride - 2.14
6. Calcium sulphate 0.26 0.81
7. Sodium Sulphate 0.31 - 0.28 0.58
8 Sodium bicarbonate 0.2
(NaHCO3)
9 Potassium 0.046 0.046

4.2 Rock salt behaviour

Rock Salt possess unique physical properties, and exhibit unique mechanical behavior. Due to
high viscosity rock salt in situ always occurs unjointed and hence can be considered as a
continuum, which makes the mathematical formulation of material behavior easier than that of
most other rocks. On the other hand rock salt shows a very complex rate and temperature
dependent behavior. Temperature has the effect on the plasticity of the rock salt. The plasticity
of the rock salt is understood by knowing the following two terms:

(i) Equivalent viscosity, and

(ii) Viscoelasticity

44
Equivalent viscosity: Plastic behavior as exhibited by salt rocks differs from viscous behavior
as shown by fluids and amorphous material, such as certain clays and sands which consist of
grains without crystalline bonding. Many researchers have been trying to apply laws of
viscosity to the plastic behavior of salt rocks which is usually termed creep. An extensive study
about salt rock may be quoted as unfortunately, the slow flow or creep of solid materials,
including rocks and salt, is exceedingly complex. In particular the behavior of crystalline salt
masses probably closely resembles that of metal, which have been much studied. Consequently,
values of the equivalent viscosity (certain parameters which indicate the viscosity of the
Newtonian fluid) cannot be simply assigned. They are functions of temperature, confining
pressure, differential stress, time, etc; in the case of salt, other variables – the amount of water,
impurities such as gypsum, etc.-must also be considered.”

Visco-elasicity: Viscoelastic behavior can be considered as a slow (or delayed) development

of stress and deformations in time. Viscoelastic phenomena belong to the fundamental
rheological effect describing the relationship between stresses and deformations. A general
theory of viscoelasticity is thus based on the three dimensional analysis and consideration of
concepts as tensor values. Quite adequate understanding of basic ideas of viscoelasticity is
available for uni-dimensional deformations. Especially it is true, for small deformations when
non- linear effects of any kind do not appear.

4.3 Mechanical Properties of Rock Salt

The most distinguish geo-mechanical properties of rock salt comparing to other rock materials
has four aspects of advantages.

(i) Low permeability: The permeability of rock salt is about 10-21– 10-24 m2, which can
ensure the excellent sealing of salt cavern.
(ii) Good mechanical properties: Damage self-recovery capability of rock salt can ensure
the safety of salt cavern with frequent changes of gas pressure.
(iii) Solution in water. Rock salt is easily dissolved into water, which facilitates the
construction and shape control of salt cavern.
(iv) Abundant resources. Rock salt resource is a very rich mineral resource with wide
distributions and large reserves

45
However, volume shrinkage, excessive displacement, ground subsidence, and even collapse of
salt cavern gas storage, etc., have become challengeable problems to the engineers for the
typical creep and rheology of salt rock.

The opening made in bedded or domal salt, which is initially under lithostatic stress, causes
deviatoric stress states to develop and the salt begins to creep. Once disturbed, salt rocks
continue to creep as long as deviatoric stresses exist. Creep around unsupported openings will
stop only after the opening is completely closed by creep and lithostatic stresses are
reestablished in the salt. Despite the substantial deformations that accumulate by creep, there
is little evidence to suggest the permeability is increased provided the deformation occurs
through its volumetric creep. However, deformation that results in an increase in the volume
of salt (dilation) does result in increased permeability. The volume increase, called dilatancy,
is attributed to microcracking occurring both during load application and during creep tests
while loads are held constant.

The mechanical behavior of salt has been investigated intensively during recent years. Early
engineering studies were driven largely by the need to design safe salt mines. In recent times
more of the study is concerned to understand the long term behavior of the rock salt to develop
and assess the safety of rock salt deposits for the storage of radioactive waste. The physical and
mechanical characteristics of salt are evidenced in its compact texture, low porosity, low
permeability and strong rheological properties (creep behavior). These properties are however
necessary for stability analysis and the design of engineering structures in salt mines. Long
term effect of the temperature on rock salt mechanical behavior has been documented by
various researchers.

From the work – in laboratory, in situ and theoretically- mechanical properties of rock salt has
been derived and are used for prediction and recalculation of the rock behaviour. Average
mechanical properties of rock salt obtained from the experimental study conducted on natural
rock salt from different locations in the United States, Canada, Germany, Poland, China,
Pakistan and Bikaner, India etc are compiled and listed in Table 4.3.

46
 Table 4.3 Average Mechanical Properties of Rock Salt

Modulus of Friction
UCS Elasticity Poisson’s Cohesion Angle Dry Density
(MPa) (GPa) Ratio (MPa) (°) (KN/m3)

Average
Mechanical
Properties
(Literature) 19.3-41.40 24.5-35.6 0.24-0.31 2.08-4.68 55-62.1 20.2-21.6

Bikenar 31.23 17.6 0.29 4.5 51 21.25

The various factors that affect the mechanical behaviour of salt are illustrated in Figure 4.1. It
has to be considered, that the effects are often dependent on each other and may not be treated
independently, e.g. dilatancy and permeability. Engineering behavior of rock salt is affected by
many factors, such as crystal size, bonding between crystal, time, temperature, inclusions and
humidity etc. Deformation and creep properties exhibit the effect on rock salt characteristics
by above mentioned factors.

Fig. 4.1 Mechanical effects & influences in rock salt

47
Mechanical behaviour of rock salt, can be divided into four categories:

1. Elastic deformation.
2. Inelastic deformation.
3. Failure
4. Temperature effect on failure

(a) Elastic deformation of rock salt

When a crystalline material such as rock salt is stressed, the crystal lattice distorts. This
distortion results in relative displacement of the atoms (or ions) in the lattice from their
equilibrium positions at zero stress. The distortion is resisted by interatomic forces so that when
the stress is removed the atoms return to their original positions. This recoverable
deformation is elastic deformation. The elastic deformation of rock salt appears linear and
isotropic and, therefore, characterized by the elastic constants and can be described by Hooke's
law. The elastic constants for rock salt are nearly independent of the source of the salt. It is
valid for pure rock salt (halite) from salt diapirs. However, impurities in salt rocks or bedded
rock salt with interlayers of anhydrite, clay, marl, mudstone, etc. behave geomechanically
different because of its different origins, mineralogical components, lithostratigraphic
disposition, texture and diagenetic history (e.g. recrystallization), tectonic history and so on.

(b) Inelastic deformation of rock salt

When rock salt is loaded and then unloaded, some permanent deformation is observed. This
non-recoverable, inelastic deformation results from motion of line imperfections in the
crystal (dislocations), mass diffusion, and, at low pressure, microcracking. There is no
generally accepted model for the inelastic behaviour of materials as for the elastic deformation.
One of the most distinctive characteristics of the inelastic deformation of salt is its very large
rate or time dependence. These effects are manifested by the continued deformation under
constant stress in creep test, the deformation rate dependence of stress in a constant strain
rate test, and the decay of stress at constant strain in a stress relaxation test. The rate
dependence of the inelastic deformation is a strong, non-linear function of stress, temperature
and deformation. Creep behaviour of rock salt is one of the important inelastic deformation of
rock salt which is a time dependent behaviour which is defined below.

48
Under uniaxial and triaxial loading conditions, typical rock salt displays behaviour with an
elastic component initially, at low confining stress or strain. Beyond a certain stress or strain,
however, the salt shows more of a plastic behaviour. In case of some stress strain relationship,
the elastic-plastic transition is not pronounced and has arbitrarily been taken to occur at about
0.2% axial strain.

Strength and deformation at failure increase considerably with increasing confining pressure
for rock salt. However, in normal brittle rocks also we observe increase of strength with
increasing confining pressure, however deformation at failure is only slightly dependent on
confining pressure. Whereas in rock salt, deformation at failure increase considerably from 7%
strain at 2 MPa conf. pressure to 35% strain under 20 MPa conf. pressure under loading at same
strain rate of 10-5S-1 as shown in Figure 4.2. This behaviour of rock salt is a major difference
from other hard brittle rocks. Rock salt has simply the ability or enough time to deform than
other rocks.

Deformation 𝜖 [%]
Fig. 4.2 Results of uniaxial and triaxial compression tests on rock salt from Asse Mine
(Germany) at same strain rate but different confining pressure.

49
(c) Failure

Material failure is usually associated with loss of load-bearing capacity. Total loss of load-
bearing capacity typically does not occur until after significant strain-softening, localizaton or
development of some major instability. Therefore it is very difficult to model deformation up
to the point of total loss of load-bearing capacity. A conservative approximate definition of
material failure is that failure occurs when the material begins to lose its load-bearing capacity,
i.e. it cannot support an increase in shear stress. In the laboratory, failure would correspond to
peak stress in a constant strain-rate test or the onset of tertiary (accelerating) deformation in a
constant stress test.

(d) Temperature effect on failure

The effect of temperature on the uniaxial compressive strength of rock salt is shown in Figure
4.3. The uniaxial compressive strength of rock salt declines dramatically as the temperature
increases. Peak strength of rock salt at 86 ℃ is only 71.6 percent of that at 20 ℃. At high
temperature, the strengthening of plastic deformation and activity cause the peak strength to
decease. The decrease of rock salt strength is possibly related to the different thermal expansion
of mineral of rock salt, fractures in crystal particles under the influence of temperature and the
possible pre-existing cracks.

20

19

18
Peak Strength (MPa)

17

16

15

14

13

12
0 20 40 60 80 100

Temperature (°C)

Fig. 4.3 Peak uniaxial compressive strength of rock salt at different temperatures.

50
4.4 Rheological Behaviour

(a) Definition of creep

In a single sentence “Creep is time dependent deformation of any material which undergoes
through a process of applied stress or pressure or temperature or chemical action of fluid or
any combination of them thereof”. For a rock specimen, creep refers to the time-dependent
deformation resulting from internal rearrangement of particles in response to the application of
a sustained stress difference. And this stress difference with time may occur when the
loads/pressures on the rock change or the geometry of the loaded/excavated region changes or
the deformability properties of the rock change. When a force is acting on a system it has both
instantaneous and time dependent part. The instantaneous part is “Elastic rebound” and time
dependent part is “Creep recovery”

In the other word, at constant stress, σ0 applied at initial time (t=0), slow (or delayed)
development of deformations, ε (t), is observed, and this phenomenon is called creep. The
function ε (t) can be considered as consisting of three components:

𝑡
𝜀 (𝑡) = 𝜀0 (𝑡, 𝜎0 ) + 𝜓 (𝑡, 𝜎0 ) + . 𝜎0 (4.1)
𝜂 (𝜎0 )

Where 𝜀0 is an instantaneous deformation, ψ (t,𝜎0 ) a function describing delayed development

of deformations, 𝜂 (𝜎0 ) viscosity, which can depend on stress, t current time.

(b) Mechanism of creep

Main mechanisms and factors that attributes to the cause of creep are micro-fracturing,
cataclasis, and frictional sliding, mechanical twinning and kinking, diffusion creep,
dissolution creep and dislocation Creep. However, creep in salt is predominantly carried by
movement of dislocations in the crystal lattice known as the normal and inverse transient creep
as well as steady state creep. Though, this definition is often somewhat weakened in the case
of dependent dilatancy, which is still called “creep”, and even “stationary creep”. All kinds of
hard rock also exhibit creep over long time intervals. If the volume of the material increases
during deformation due to microcracking, this is called dilatancy. It causes damage and
produces a number of consequences like increasing permeability, accelerating creep and creep
rupture and mechanical weakening.

51
Creep behaviour is that part of geomechanical behaviour, which has the greatest influence on
the time –dependent strain and stress evolution in a system. Therefore, it is very important to
determine the creep behaviour with sufficient accuracy. In general, there is an initial period
during which the strain increases rapidly but at a decreasing rate (i.e., the curve is concave
down). This is followed by a period of time in which the strain increases at more or less a
constant rate (linear curve). For the two highest confining stresses, 25 MPa and 20.5 MPa, after
a certain time the strain begins to increase at an ever-increasing rate (concave-upward curve),
eventually causing complete collapse of the sample.

If a stress is applied instantaneously to a sample, it will immediately give rise to an

Instantaneous deformation which is mainly elastic, indicated by A in the Figure 4.4. Thereafter,
the strain continues to increase, albeit at an ever-decreasing rate i.e. work-hardening,. This
regime, denoted as I, is referred to as primary or transient creep. This is followed by regime II
of steady-state creep, in which the strain increases linearly with time i.e. the balance of work-
hardening and recovery takes place. Finally, there may follow regime III of tertiary creep, in
which the strain increases at an increasing rate, until failure occurs i.e. formation of internal
cracks, voids, grain boundary separation, necking, etc.

I II III

Fig. 4.4 Idealized creep curve showing primary, secondary, and tertiary creep & two unloading
paths.

The three regimes have different characters that extend beyond their differences with regards
to the second derivatives of the strain vs. time curves. If the applied stress is suddenly reduced
to zero during primary creep, the strain will relax back to zero, along a path such as PQR.

52
However, if the stress is removed during steady-state creep, the strain will relax to some
nonzero value along a path such as TUV, leaving a residual, permanent strain.

(c) Temperature effect on rheological behaviour

Comparing the three curves at different temperatures in Figure 4.5, it is observed that
temperature plays a large role in controlling the rheological behaviour of rock salt. Under
otherwise identical conditions, the time to reach steady-state creep becomes shorter for samples
at a higher temperature. However, the primary creep strain εp at the higher temperature is much
larger than that at lower temperatures. At elevated temperature, the deformation mechanisms,
such as dislocation, gliding and diffusion become much more active. So, it is easier to reach
the stable state, and the time to reach the steady-state creep stage reduces.

Fig. 4.5 Multi Step triaxial creep test at three different temperatures and at first stage deviatoric
stress of 10Mpa and the second stage deviatoric stress of 15 Mpa

Under a certain stress state, there is an observable change on the creep curve of rock salt when
the temperature changes as shown in Figure 4.5. Moreover, this change is smooth as the
temperature changes, and only a new steady-state creep stage develops, without a new primary
creep stage. Activation energies for diffusion and dislocation creep of rock salt were obtained
by fitting an Arrhenius type equation to the creep data:

εṡ = B・exp (−Q/RT) (4.2)

53
 where B is a constant, Q is the activation energy for creep, R is the gas constant (R = 8.31J•mol-
1•K-1), T is the temperature in Kelvin. In general, the experimentally determined activation
energies for salt vary between 16 and 200 kJ•mol-1, increasing with temperature.

(d) Stress effect on rheological behaviour

Strain-time creep curves at various stress levels at 50, 75 and 86 ℃ are shown in Figure 4.6.
Obviously, every creep curve can be divided into two stages: primary creep and steady-state
creep. Under the same confining pressure and temperature, the time to steady-state creep
increases with increasing deviatoric stress. This is because a considerable time lapse is needed
to adjust to the new stress state when the stress changes, especially when the change is very
large.

Fig. 4.6 Triaxial creep test under different temperatures. confining pressure = 20 MPa,
deviatoric stress =17 MPa

4.5 Constitutive Laws for Rock Salt

The behaviour of rock salt is more ductile, and its increased deformability is accompanied by
a strong time dependency. However, modelling of non-linear rate-dependent is a challenging
task, especially when dealing with the different inelastic phases which typically include quasi-
instantaneous (elastic and/or plastic), transient and steady-state responses.

Different constitutive laws have been developed for the description of the viscous behaviour
of rock salt. Amongst the theoretical investigations, comprehensive laboratory test have also
been performed to understand the complex behaviour of rock salt.

54
To describe the behaviour of rock salt, it is customary to identify up to four different straining
phases on the strain time curve. During the loading stage, there is the pseudo instantaneous
strain phase which includes the elastic 𝜀 𝑒 and plastic 𝜀 𝑝 strains. Then follow three phases of
time dependant (creep) 𝜀 𝑐 strains: transient 𝜀𝑡 (or primary), steady-state 𝜀𝑠 (or stationary,) and
accelerating 𝜀𝑎 (or tertiary), with the last one being frequently omitted in usual applications.

The total strain rate can then be expressed with the following portioned equations:

𝜀 = 𝜀𝑒 + 𝜀𝑝 + 𝜀𝑐 (4.3)

Where 𝜀 𝑐 = 𝜀𝑡 + 𝜀𝑠 + 𝜀𝑎

Some of the constitutive laws are discussed here:

4.5.1 Conventional constitutive laws for rock salt

Mostly, the strain rates of a rate dependent material are divided into an elastic and viscous part:

𝑒𝑙 𝑣𝑖𝑠
𝜀𝑖𝑗̇ = 𝜖̇𝑖𝑗 + 𝜖̇𝑖𝑗 (4.4)

Usually, the elastic strain rates are assumed to follow Hookes law

𝑒𝑙 1 1
𝜖̇𝑖𝑗 = 𝑠̇ + 9𝐾 𝜎̇𝑘𝑘 𝛿𝑖𝑗 (4.5)
2𝐺 𝑖𝑗

Where G and K denote the sear and the bulk modulus and 𝑠̇ 𝑖𝑗 are the deviatoric stress rates

1
𝑠̇𝑖𝑗 = 𝜎̇𝑖𝑗 − 𝜎̇ 𝛿 (4.6)
3 𝑘𝑘 𝑖𝑗

𝑣𝑖𝑠
So far, research concentrated upon the viscous part 𝜖̇𝑖𝑗 describing steady state creep with
constant creep rate or transient creep with decreasing creep rate. Tertiary creep followed by
creep rupture is not considered here.

55
Steady –state creep:

If we assume that the material behaviour can be described by the functional link

𝑓 (𝜀𝑖𝑗̇ , 𝜎𝑖𝑗 , 𝑇, 𝑆𝑖 ) = 0 (4.7)

Where T denotes the temperature and the state variables 𝑆𝑖 characterize the actual microscopic
structure, a general constitutive law for the secondary creep of an isotropic incompressible
material with multiplicative terms of stress and temperature dependency is given by

𝑣𝑖𝑠 𝑠̇
𝜖̇𝑖𝑗 = 𝑓1 (𝜎𝑒 ).𝑓2 (𝑇) 𝜎𝑖𝑗 (4.8)
𝑒

With
1
𝜎𝑒 = √2 ((𝜎1 − 𝜎1 )2 + (𝜎2 − 𝜎3 )2 + (𝜎1 − 𝜎3 )2 ) (4.9)

The structure variables 𝑆𝑖 are mostly assumed to be constant or to be included in the remaining
measuring parameters and therefore do not appear explicitly.

Commonly, the temperature dependency 𝑓2 (𝑇) follows the Arrhenius equation

𝑄
𝑓2 (𝑇) = 𝐶 exp(− 𝑅𝑇) (4.10)

with the universal gas constant R ( R= 8.314 J/mol ⋅ K ), an activation energy Q and a positive
constant C. The Arrhenius dependency is confirmed by many laboratory tests as well as
theoretical investigations. Steady state creep laws mainly differ in regard to the stress
dependent function 𝑓1 (𝜎𝑒 ). A frequently applied expression is the power dependency

𝜎 𝑛
𝑓1 (𝜎𝑒 ) = 𝐶 (𝜎𝑒 ) , 𝑛 ≥ 1 (4.11)
0

Where 𝜎0 denotes a reference stress proposed a power law for salt rock.

56
 Both the power law and the hyperbolic sine law can be derived from microstructural
considerations.

𝜎
𝑓1 (𝜎𝑒 ) = 𝐶1 sinh(𝐶2 𝜎𝑒 ) (4.12)
0

4.5.2 The Norton creep law

This constitutive model is a classical power law used to describe stationary creep. It can be
written as:

(𝜀𝑒 )s = B 𝜎𝑒𝑛 (4.13)

where (𝜀𝑒 )s is the steady-state creep rate (expressed using the equivalent von Mises strain),
𝜎𝑒 is the deviatoric stress state, B and n are material parameters.

It is important to mention that the Norton Creep Law is only an approximation of the actual
creep behaviour of salt. It neglects the straining occurring in the transient phase, and it idealises
the stress-strain rate relationship (which has been shown to better obey hyperbolic sine law.
This model is nevertheless largely used because of its simplicity of application .However,
fundamental limitations may induce some significant deviation from the actual rocksalt
behaviour, especially under the complex loading conditions encountered in a natural
geomechanical setting.

4.5.3 BGRa model

Since only steady state creep is used in many model calculations, a widely used one is BGRa:

𝑄 𝜎
𝜖̇∞ = 𝐴. exp (− 𝑅𝑇) . (𝜎 )𝑛 (4.14)
∗

Where A = 0.18 d−1

n = 5 (stress exponent)
Q = 54 kJ mol−1 (activation energy), 𝜎∗ = 1 MPa
R = 8.31441 kJ mol−1 K−1 (universal gas constant)

57
4.5.4 Lubby 2 material model

The mathematical–mechanical plot of creep behaviour using the Lubby2 material model is
achieved using a bipartitie approach where the creep deformation observed in the lab tests is
divided into transient and stationary creep deformation. The total creep rate 𝜀̇1𝑣 and the sum of
the transient creep rate 𝜀̇1𝑡𝑟 and the stationary creep rate 𝜀̇1𝑠𝑡 are calculated in accordance with
the following equation

1 εν,𝑡 ̅
1 . 𝐺𝑘(𝜎) 1
𝜀̇1𝑣 = 𝜀̇1𝑡𝑟 + 𝜀̇1𝑠𝑡 = [ 𝜂̅ . {1 − }+ ̅𝑀(𝜎,𝑇)
] . 𝜎𝜐 (4.15)
𝑘(𝜎) 𝜎𝜐 𝜂

Where
∗
𝜂̅𝑀(𝜎,𝑇) = 𝜂̅𝑀 . exp( 𝑚. 𝜎𝜐 ). exp(𝑙. 𝑇)
∗
𝐺̅𝑘(𝜎) = 𝐺̅𝑘 . . exp(𝑘1 .𝜎𝜐 )
𝜂̅𝑘(𝜎) = 𝜂̅𝑘∗ . exp( 𝑘2 . 𝜎𝜐 )
𝜀̇1𝑣 Viscous creep rate (d-1)
𝜂̅𝑀(𝜎,𝑇) Maxwell viscosity coefficient (MPa . d)
𝐺̅𝑘(𝜎) Kelvin shear modulus (MPa)
𝜎𝜐 Equivalent stress (MPa)
T Temperature (K)
𝑘1 , 𝑘2 ,𝑚, l Material Parameters

(a) Transient Creep Parameters

The parameters 𝐺̅𝑘∗ , 𝑘1 , 𝜂̅𝑘∗ and 𝑘2 is determined to characterise the transient creep with the
Lubby2 material model

The Kelvin shear modulus 𝐺̅𝑘 is defined as the quotient of the stress 𝜎𝜐 and the transient creep
strain for t = ∞
𝜎𝜐
𝐺̅𝑘(𝜎) = 𝑡𝑟 (4.16)
𝜀1∞

Where

58
 𝑡𝑟
𝜀1∞ = Transient creep strain for t = ∞

The Kelvin viscosity modulus η𝑘 is dependent on the kelvin shear modulus, the equivalent
stress and the transient creep strain (stain hardening approach). The parameter is calculated by

𝐺̅𝑘(𝜎) .𝑡
𝑛̅𝑘(𝜎) = ̅
𝐺 𝑘(𝜎)
(4.17)
ln(1− 𝜀1𝑡𝑟 . )
𝜎𝜐

Where

𝜀1𝑡𝑟 = Transient creep strain at time t

Unlike the kelvin shear modulus and the equivalent stress which can be directly determined
from the creep test results, determining the transient creep strain can only be done indirectly
by back calculating in accordance with equation. This involve reading the total creep strain
measuring a test from the test curve at specific point in time and reducing it by the amount of
stationary creep strain calculated for this point in time.

ε1tr = ε1v - ε1st (4.18)

𝜎𝜐
ε1st = 𝜂̅ .𝑡 (4.19)
𝑀(𝜎,𝑇)

(b) Parameter for stationary creep

The Maxwell viscosity coefficient 𝜂̅𝑀 is defined as the quotient of the equivalent stress σv and
the stationary creep rate 𝜀̇1𝑠𝑡 ∶

𝜎𝜐
𝜂̅𝑀(𝜎,𝑇) = (4.20)
𝜀̇ 1𝑠𝑡

The smaller the Mawell viscosity coefficient, the higher the resulting stationart creep rate for
each defined equivalent stress. This parameter describes the “time-dependent rigidity” of
viscous material in an analogous way to the time-dependent rigidity defined by Hooke’s
material model. 𝜂̅𝑀 is determined for each single test (each load stage) on the basis of the creep
rate readings determined during the stationary creep phase.

59
 The stationary creep rate can be calculated as a function of the equivalent stress from the
previously determined material parameters for stationary creep, and compared with the
laboratory tests.

𝜎𝜐
𝜀̇1𝑠𝑡 = 𝜂̅∗ (4.21)
𝑀 .exp(𝑚.𝜎𝜐 )

4.6 Failure Strength of Rock Salt

As discussed due to high viscosity rock salt in situ always occurs unjointed and hence can be
considered as a continuum, which makes the mathematical formulation of material behaviour
easier than that of most other rocks. However, on the other hand rock salt shows a very complex
rate and temperature dependent behaviour.

Figure 4.7 shows a more complete picture of the failure behaviour of rock salt derived from a
great number of compression tests with carious values of constant 𝜖̇ and at various confining
pressure. In this diagram, strength is given as function of p and 𝜖̇ for room temperature. The
failure domain shown is bounded on the left hand side by the steady state creep curve. This is
often denominated as the brittle-ductile transition. It is obvious from the figure that rock salt
has a weak maximum in strength at about 𝜖̇ = 10−4 𝑠 −1 for all confining pressure. This is
different for hard rock, where strength steadily increases with 𝜖̇. The minimum of strength at
high strain rate is probably caused by dynamic processes in combination with the
inhomogeneities of the material. Also the strain is dependent not only on p but also on 𝜖̇ since
the specimen has more time to deform in the slow tests than in the fast ones. This means that
the deformation up to failure itself does not posses a deep meaning and that there is no criterion
of this kind of soft rock. However, a comprehensive creep law must be able to describe these
deformations. The maximum failure deformation gives only the limit for deformation at failure
for certain confining pressure and deformation rates. These strains are different for different
types of rock salt.

From quasistatic deformation tests one can draw conclusions about the creep behaviour and
the microscopic mechanism, since the considerable part of the deformation is due to (transient)
creep in salt.

60
 Important is also the tensile strength of salt, because tension is always present in the vicinity
of opening. Generally, the short term tensile strength failure strength of rock ranges between 5
and 10% of the UCS. For salt the values varies between 1.5 and 2.4 MPa (Average UCS around
25MPa).

Fig. 4.7 Strength of rock salt at a greater number of confining pressure P and strain rate 𝜖̇

Therefore, the final aim of the failure criteria cannot be to determine strength of rock in terms
of stress only. A comprehensive picture should also include the pre failure behaviour and
perhaps the post failure behaviour and this means not only deformation but also volume change
(dilatation) and its related phenomena, permeability, creep failure, damage, micro acoustics
emissions, and others as the determination of dilatation is important parameter for rock salt as
it is used extensively in repositories.

4.7 Dilatancy of rock salt

When rock salt is mechanically under the influence of different stress conditions, it can react
in two different ways represented by two different stress dependent domains as shown in Figure
4.8. The two domains consist of the compaction domain and the dilatancy domain, which are
separated by the dilatation boundary. As long as the state of stress remain in the non-dilatant
compaction domain, rock salt deforms plastically without any crack formation and without
dilatant crack propagation. The volume of the material decreases until the micro cracks are
closed and healed. Besides this healing effect, one only observes pure creep produced

61
predominantly by the movement of dislocations and, of course, elastic deformation. Above the
dilatancy boundary, one finds, besides the pure creep, the microcracking related effects. The
transition in the state of stress, from the ductile deformation behaviour (no volume increase
starts to be associated with dilatant deformation behaviour corresponds to the dilation
boundary. Figure 4.8 summarize the results containing the dilatancy boundary which have been
derived by recording the volume change in the specimens. The dilatancy boundary for rock salt
is practically only dependent on stress condition (σ = mean stress, τ =octahedral shear stress)
and not on salt type, stress geometry, or loading rate.

In practice, the dilatation boundary, the dilatation boundary is more a band that a distinct line.
The dilatation boundary (grey line) is clearly distinguishable from the graph representing the
mean failure strength (black line) where short term failure is expected to occur.

Fig. 4.8 Dilation boundary and short-term failure strength (compression) for rock salt:
Octahedral shear strength 𝜏, versus mean stress 𝜎.

Important property changes associated with dilatancy are also indicted in Figure 4.8.The
dilatant domain is characterized by micro cracking, causing accumulation of damage.
Permeability and probability of creep failure are accordingly increasing. In the non-dilatant
domain, the rock salt is compressive and micro crack are compacted, closed or even healed and
further micro-cracking is supressed. accordingly, permeability decreases and no failure occurs
even during long-term deformation.

62
4.8 Dilatancy criterion of rock salt

A elasto-viscoplastic constitutive equation for rock salt which is able to describe not only the
elastic and creep behaviour, but also the time and stress dependent change of the volume
(dilatancy and compaction) as well as short- and long-term failure. The basic equation of this
model is:

𝜎̇ 1 1 𝑊(𝑡) 𝜕𝐹 𝜕𝑆
𝜀 ̇ = 2𝐺 + (3𝐾 − 2𝐺) 𝜎.̇ 𝐼 + 𝐾𝑇 ⟨1 − 𝐻 (𝜎)⟩ + 𝐾𝑠𝑡 𝜕𝜎 (4.22)
𝜕𝜎

Where 𝜀 ̇ : tensor of creep rate; 𝜎̇ : tensor of stress rate; 𝐺, 𝐾 elastic moduli; 𝐾𝑇 , 𝐾𝑠𝑡 viscosity
coefficient for transient or steady state creep; 𝐹(𝜎, 𝜏) is the viscoplastic potential for transient
creep and its entangled formulation describes volume change, the dilatation boundary and
short-term failure; 𝑆(𝜎, 𝜏) is the potential for steady state creep and for steady state volume
change as well; 𝑊(𝑡) is the irreversible (non-elastic) stress work per unit volume;𝐻(𝜎, 𝜏) is
the yield function for transient creep, i.e. the hardening limit. In the actual formulation the
influence of the third invariant (or Lode parameter 𝑚) has been disregarded. Therefore, the
stress tensor 𝜎 has been replaced by the two octahedral stresses 𝜎 and 𝜏 alone. Temperature
dependency is not included.

The first two terms of Equation 22 describe the elastic deformation in the usual way. The
third term calculates the transient creep and the related volume change and failure. The last
term is related to steady state creep, including volume change.

The rate of the non-elastic volume change can calculated directly from Equation 4.23 quite
easily:

𝑊(𝑡) 𝜕𝐹 𝜕𝑆
𝜀̇𝑣1 = 𝐾𝑇 ⟨1 − 𝐻 (𝜎)⟩ + 𝐾𝑠𝑡 𝜕𝜎 (4.23)
𝜕𝜎

63
4.9 References

1 Baar, C. A., (1977) Applied Salt-Rock Mechanics-l The in-situ behaviour of salt rocks.
Elsevier Scientific Publishing Company Amsterdam - Oxford - New York.
2 Chao, D., Chunhe, Y.,Yuanfeng, Y., Zhe, L.,Jie C., (2012) Mechanical Behaviour of
Deep Rock Salt under the Operational Conditions of Gas Storage, International Journal
of Earth Sceince and Engineering, ,P.P.1670-1676.
3 Cristescu, N., Hunsche, U., (1998) Time Effects in Rock pp. 217–230. Mechanics —
Series: Materials, Modelling and Computa- Hunsche, U., Schulze, O., 1996. Effect of
humidity and confintion.John Wiley and Sons, Chichester, 342 pp.
4 Dubey, R.K., (1996) Rheological behaviour of rocks of lesser Himalaya around Mandi
and Mussoorie areas, Ph.D. thesis, Department of Mining Engineering, B.H.U., 100.
5 Gevantman, L. H., (1981) Physical Properties Data for Rock Salt, U.S. Department of
Commerce; National Bureau of Standards monographs.
6 Hunsche, U., (1998) Determination of the dilatancy boundary and damage up to failure
for four types of rock salt at different stress geometries. In: Aubertin, M., Hardy Jr., H.R.
(Eds.), The Mechanical Behavior of Salt IV; Proc. of the Fourth Conf.,(MECASALT
IV), Montreal 1996. TTP Trans Tech Publications, Clausthal, pp. 163–174.
7 Hunsche U., Hampel A., (1999) Rock salt - the mechanical properties of the host rock
material for a radioactive waste repository. Eng Geol 52:271–291.
8 ISRM (2013) Suggested Methods for Determining the Creep Characteristics of Rock,
Aydan o. , Takashi I. , Ugur O., Kwasniewski M., Shariar K., Okuno T., Ozgenoglu
A., Malan D. F., Okada T. Rock Mech Rock Eng (2014) 47:275–29.
9 Jaeger, J.C., Cook, N.G.W., (1979) Fundamentals of rock mechanics, 3rd edn. Chapman
& Hall, London, pp 79 and 311.
10 Langer, M., (1979) Rheological behaviour of rock masses, Proceedings of 4th Int. Cong.
on Rock Mechanics, Montreux, Balkema, Rotterdam, Vol. 3, pp. 29–62.
11 Lux, K.H., Heusermann, S., (1983) Creep tests on rock salt with changing load as a basis
for the verification of theoretical material laws. In: Proceedings of 6th Symposium on
Salt, Toronto, vol. I, p. 417–35.
12 Passaris, E.K.S. (1979) The rheological behaviour of rocksalt as determined in an in situ
pressurized test cavity. Fourth International Congress on Rock Mechanics, Balkema,
Rotterdam, pp 257–264.
13 Paul, L.C., (1965) Creep in Rock Salt. The Journal of Geology, 73(3): 469-484.
14 Senseny, P.E., Hansen, F.D., Russell, J.E., Carter, N.L.,Handin, J.W., (1992) Mechanical
Behavior of Rock Salt: Phenomenology and Micromechanisms. Int J Rock Mech Min
Sci & Geomech Abstr; 29(4): 363-378.
15 Van Sambeek, L.L., (1986) Creep of Rock Salt Under Inhomogeneous Stress
Conditions, Ph.D. Thesis, Colorado School of Mines, Golden, CO.
16 Yu, M.H., Zan, Y.W., Zhao, J., Yoshimine, M., (2002) A unified strength criterion for
rock material. J Rock Mech Min Sci; 39:975-89.

64
 5 Design and Geo-mechanical Analysis of Salt Cavern

The rock mechanical design concept become of considerable significance in the layout of
caverns when considered in connection with the basic understanding of the material behaviour
of salt rock. The numerical processes gather the nonlinear time-dependent material behaviour
of the salt rock and serve to lay down the geometrical and physical conditions necessary for
establishing the storage operation conditions. The results of the recommendations for the
cavern design, also provide the confirmation of the stability, and also, from the point of view
of the operator, the confirmation of the economic usability of the cavern over the operating
time period (Heusermann et al., 2003).

5.1 Rock Mechanics Cavern design

5.1.1 Design Fundamentals

The basic fundamental of rock-mechanical design of a storage cavern has to guarantee its
serviceability and the stability of the surrounding formation during every mode of operation
(leaching, oil injection, oil storage, and withdrawal). The factors influencing the stability of an
unlined underground cavity are basically divided into four groups:

(i) The physical effects (formation pressure, internal pressure, temperature),

(ii) The rock mass stresses, theoretically determined from the above with the help of a
calculation model,
(iii) The resistance (strength) of the rock that acts against the physical effects (material
characteristics),
(iv) The influence of geological anomalies

In cavern design, these variables influencing the safety factors have to be taken into
consideration. In particular, establishment of the partial safety coefficient, as for instance,
demanded by the processes underpinned by the probability theory, is, however, very
problematic because the probability of the combined effect of different influences acting
together must be investigated.

65
In each group there are a number of uncertainties to be taken into consideration that could
reduce the load bearing capacity of the rock surrounding the cavity (Staudtmeister and Rokahr,
1997). Significant influences which can result in erroneous evaluations are as follows:

(i) the false estimation of the formation pressure (lateral pressure coefficients, densities,
(ii) the deviation of the static system from reality (e.g. excessive idealisation with respect
to the geometry or the material behaviour), calculation inaccuracies,
(iii) the deviation of the assumed rock strength from reality, inappropriate adoption of rock
characteristics to the formation characteristics, laboratory tests on non-representative
core material, time influences,
(iv) the presence of faults or in-homogeneities that could not be determined from the
engineering geological evaluation, or were not taken into consideration in the
theoretical calculation model.

The following important aspects must be taken into account when establishing the safety
factors with respect to the failure conditions when constructing salt caverns for the storage of
gas:

(i) Quality of the geological investigations,

(ii) Quality and scope of the laboratory investigations to determine the material
characteristics of the formation in situ,
(iii) Experience from the operation of already existing caverns in the same project area,
(iv) Level of idealisation of the theoretical calculation model,
(v) Different failure behaviour (failure mechanisms, effect on the surface or neighbouring
caverns; max. or min. internal pressure),
(vi) Possible negative effects on the environment,
(vii) Further potential use of the cavern after a failure.

5.1.2 Design Parameters

The design variables to be established for the layout of a cavern, are predominantly based on
the final use to which the cavern is to be put, e.g. mineral extraction, storage or disposal site.
The most important parameters to be taken into consideration in the dimensional analysis of a
cavern are as follows:

66
(i) depth of the cavern,
(ii) cavern geometry (diameter d, height h, roof shape),
(iii) distance b of the caverns from one another,
(iv) distance a or s to the neighbouring formation,
(v) the internal pressure conditions (𝑝𝑚𝑖𝑛 ; 𝑝𝑚𝑎𝑥 ).

Often, individual parameters are predetermined, for instance, the borehole separations in
caverns to be converted, the depths associated with the actual geological conditions, or the
minimum or maximum internal cavern pressure reflecting the operational or engineering
conditions.

Figure 5.1 shows the geometrical boundary conditions for a hexagonal cavern field in salt rock.
The individual caverns have a cylindrical shape in order to create the largest possible storage
volume in the salt dome structure.

The individual caverns have a cylindrical shape in order to create the largest possible storage
volume in the salt dome structure. The most important internal pressure conditions for the
operation of the caverns for gas storage. In addition to the limitation of the maximum pressure
to prevent seal failures or the fracturing of the formation, the minimum allowable internal
pressure during operation is also laid down, in order to avoid spalling from the walls of the
cavern.

Fig. 5.1 Rock Mechanics design of caverns in salt

67
Because of the pronounced time dependent material behaviour of rock salt, it is also necessary
to establish the time period during which the cavern can be operated at the minimum pressure.
Furthermore, the permissible rate of pressure change within the cavern has to be determined.
In addition to the requirements for a stable cavern, it is also necessary to investigate the size of
the volume losses which may occur as a consequence of convergence at internal cavern
pressures

5.1.3 Determination of design variables

(a) Allowable minimum pressure

The minimum allowable internal cavern pressure is basically dependent on the formation
pressure, itself largely dependent on the depth of the cavern. In the case of caverns intended
for gas storage, the shape of the cavern marginal areas have to be maintained during life time
of the cavern - several decades - even though the minimum pressure will be reached on several
occasions. The fracture strength of the rock salt formation must never be exceeded in any part
of the cavity marginal area.

Because, by exceeding the long-term strength of the rock salt, all stress conditions are
associated with a certain finite lifetime, the establishment of the minimum internal cavern
pressure is associated with a corresponding operational time limit reflecting this internal
pressure condition. The criteria used are the limitation of the stress intensity index and the
effective strain in the rock salt mass enclosing the cavern.

(b) Maximum permitted internal cavern pressure

The maximum internal cavern pressure is set at a value so that within the whole area of the
cavern in particular in the cavern roof and casing shoe area - fracturing of the rock and an
increase in the permeability as a result of micro fracturing can be avoided. Because leaky areas
in the rock are difficult to confirm by measuring techniques, and fracturing of the rock is
associated with high financial losses for the operator, a comparatively high amount of safety
compared to the failure states should be used.

68
Determining the maximum permissible operating pressure depends on the permeability
behaviour of the rock. Rock salt analysis looking into the stress-dependent permeability
behaviour has shown that permeability can increase considerably when the gas injection
pressure is approximately equal to the formation pressure. This means that the numerical
calculations must be used to confirm an adequate safety margin between the maximum
operating pressure and the stress state existing in the surrounding rock (Rokahr and
Staudtmeister 1993).

(c) Usability confirmation

In addition to establishing safe cavern operations reflecting the associated operating

parameters, the rock mechanical evaluation should also provide an assessment of cavern
usability. This evaluation involves describing the future rock deformation or cavern closure
effects on the basis of the calculation results and comparing them with the operation
specifications or the requirements of the operator.

Volume losses associated with internal pressure reductions are determined by time-dependent
calculations of the cavern closure taking into consideration realistic internal pressure changes
over time. It is also possible with the help of these calculation results to extrapolate the possible
effects of the cavern closure to the ground surface, and thus assess any possible risk to buildings
on the surface or any impact on the hydrogeology.

5.2 Rock Mass Model

A rock mass model is necessary to carry out the rock mechanical calculations that encompasses
the idealisation with respect to the geological structure, initial rock stress conditions and initial
rock temperatures, as well as the material models, with the associated material parameters for
the individual layers.

(a) Regional Geological Framework

This idealisation usually involves dividing up the rock mass into overburden, rock salt and the
underlying rock formation. Each of these areas is idealised as a homogeneous isotropic

69
continuum with respect to their material properties. An example for idealisation of a geological
profile into a rock mass model is shown in Figure 5.2.

(b) Initial Rock stress

The vertical initial rock stress 𝜎𝑣,𝑝 reflects the theoretical overburden pressure. The horizontal
initial rock stress 𝜎ℎ𝑝 results both from the lateral stress coefficient k0 and the vertical stress
𝜎𝑣,𝑝 . In thicker salt deposits, and especially at greater depth, the very ductile material behaviour
of rock salt means that a lateral pressure coefficient of k0 = 1.0 is used. This takes into
consideration the fact that rock salt over long periods of time behaves more like a highly
viscous fluid with no shear strength than a solid body with shear strength. When considering
caverns in bedded salt, this value needs to be looked at in detail when the salt layers are thin
because in these cases often k0 ≠ 1.0.

Fig. 5.2 Geological Profile and rock mass model

(c) Rock mechanical properties

The rock mechanical properties of the cavern host rock in combination with depth and hickness
of geological formation are very important for the design of storage cavern. Typical rock
mechanical material properties to be investigated are:

70
(i) Stress-strain behaviour
(ii) Strength (short term and long term)
(iii) Bond between layers
(iv) Creep behaviour of rock salt.

Short term test are performed to assess the stress-strain behaviour and the strength of rock while
the long term test are performed to assess the visco-plastic creep behaviour.

The input rock mechanical properties required for the analysis are listed in the table 1.

Table 5.1 Mechanical Parameters of Rock Salt

Sl. No. Property

1 Density ( kg/m3 )
2 Elasticity Modulus (GPa)
3 Poisson’s ratio
4 Cohesion (MPa)
5 Friction angle °
6 Tensile Strength
7 Creep Parameter

(d) Initial rock temperature

Where no direct temperature measurements are carried out, the initial rock temperature of a
central cavern zone has to be realistically assumed. Because the influence of temperature
fluctuations associated with operations only affects the bounding cavern wall at a depth of ≈1
m into the surrounding rock, this phenomenon is usually be ignored.

(e) Constitutive model

The rock being investigated incorporates lithologies with very different material properties.
Idealisation of the rock mass structure does however allow differentiation between a
viscoplastic–ductile material behaviour of rock salt and the more elastic–brittle behaviour of

71
the overburden and the underlying rock formation. This means that the rock zones behaving in
an elastic–brittle way can be described by a linear-elastic material model.

To take the ductile behaviour of rock into consideration as realistically as possible, a sufficient
material constitutive model must be used.

(f) Determination of creep parameters

Determination of creep parameters from laboratory test are an important aspect of the geo-
mechanical analysis. Creep parameters of suitable constitutive model are determined from the
results of multistage creep tests on borehole samples with step-wise load increase. The
dependence of the creep rate on stress is taken into account by using stress-dependent creep
parameters.

5.3 Numerical Analysis

The finite-element method can be used for the analysis of the rock stresses arising from the
construction and operation of the cavern. This method allows the rock structure, the mechanical
behaviour of these structures, and almost any geometrical cavity shape to be realistically
represented by mathematical–mechanical models.

The numerical analysis of these models allows a realistic evaluation of the mechanical
behaviour and stability of the caverns. Caverns must basically be considered as cavities in a
three-dimensional continuum. However, in most cases, it is sufficient to idealise the rock as a
two dimensional axisymmetrical continuum. Each cavern is then also considered as an
axisymmetrical cavity (Figure 5.3).

72
 Fig. 5.3 Geological Profile and rock mass model

It is important when assessing the rock stress state that develops during cavern operations to
take into consideration each stress component as well as the effective stress 𝜎𝑒𝑓𝑓 and the
effective strain 𝜀𝑒𝑓𝑓 .

In general, the cavern design differentiate between two types of proof as shown in Figure 5.4:

(i) Proof of stability (permissible stresses and strains),

(ii) Proof of usability (tightness of a cavern system, limitation of cavern closure).

Fig. 5.4 Major Criteria for the stability and usability of salt caverns

73
5.3.1 Stability criteria

Determining the stability of a cavern requires interpretation of the results derived from the
numerical calculations including the state of stress and strain around the cavern. This in turn
requires various assessment criteria. Typical criteria are as follows:

(i) exceeding the tensile strength of the rock under tensile stress;
(ii) exceeding the shear strength during deviatoric stress;
(iii) exceeding a specific deformation or deformation rate.

If one of these criteria is not complied with, it can lead to a reduction, and in exceptional cases
to complete loss in the load bearing ability of a component or a load bearing system.

Concerning the criteria of exceeding the shear strength for salt typical characteristic has to be
mentioned. Empirically, the failure strength of rock salt decreases with the duration of the
stress. The time-dependent reduction in material strength must be taken into consideration
when determining acceptable stress levels. Calculation of the stress reduction is made with the
stress intensity index, which has the meaning of a reduction factor. The stress intensity index
g indicates the percentage actually used of the potential stress which the rock mass could absorb
over a short time (Rokahr and Staudtmeister, 1993).As an example, Figure 5.5 shows a
simplified plot of the period of cavern stability as a function of stress intensity index g over
time.

Fig. 5.5 Calculated stress intensity index at the cavern wall over time

74
5.3.2 Available numerical modelling tools

The available numerical tools for the analysis of salt cavern are

(a) ABAQUS
(b) FLAC 3D
(c) ANSYS
(d) ADINA
(e) UDEC

5.4 References

1 ISPRL, March (2013) , Detailed Project Report for Phase-II of Strategic Storage
Program for Crude Oil (Volume IV)
2 Hunsche, U.E., (1992) Failure Behaviour of Rock Salt Around Underground Cavities.
7th International Symposium on Salt, Kyoto, Japan.
3 Lux, K.H., (2009) Design of salt caverns for the storage of natural gas, crude oil and
compressed air: geomechanical aspects of construction, operation and abandonment.
Geol Soc Lond, Special Publications; 313:93–128.
4 Mirza, U.A., (1981) Prediction of creep deformations in rock salt pillars. In:
Proceedings of the First Conference on the Mechanical Behavior of Salt. Trans.Tech.
Publications, pp. 311–325.
5 Mraz, D., (1980) Plastic behavior of salt rock utilized in designing a mining
method.CIM Bull;3:115–23.
6 Rokahr, R.B., Staudtmeister, K., (1993) The assessment of the stability of a cavern
field in bedded salt with the help of the new Hannover dimensioning concept. In:
Proceedings of 3rd Conference on Mechanica Behaviour of Salt, Palaiseau, p. 519–30.
7 Staudtmeister, K., Rokahr, R.B., (1997) Rock mechanical design of storage caverns for
natural gas in rock salt mass. Int. J. Mech. Min. Sci. Geomech. Abstr. 34,646–660.
8 Swift, G.M., Reddish, D.J., (2005) Underground excavations in rock salt. Geotech.
Geol. Eng. 23, 17–42.

75
 6 Evaporation

6.1 Introduction

One of the salient feature of constructing salt cavern is the solution mining (or leaching)
technology. The leaching operation is planned to be carried out by sourcing water from sea,
groundwater or any other sources as available, thus leaving a large artificial cavity billed with
saturated brine. The brine disposal is designed through solar evaporation in shallow basins
resulting in harvesting of solid salt. Brine which is also used for compensation of withdrawn
oil during commercial oil operation can either be stored aboveground in a pond or can be
produced ‘on demand’ from additional caverns (‘donor caverns’). Option of storage of brine in
brine evaporation basin is discussed in the present chapter.

Brine evaporation basin are artificial ponds with very large surface areas that are designed to
efficiently dispose brine produced during leaching operation or during oil injection by solar
energy though exposure to the ambient temperatures. The aboveground brine storage pond
needs always to be filled with brine and measures against enhanced evaporation have to be
taken.

To handle the peak production of brine from parallel leaching (or oil fill), a large surface area
of to the extent of more than 10 km² may be required. Due to this, climate conditions like
evaporation and precipitation are the important values to define the needed evaporation basin.
The process of obtaining salt which are dissolved in natural brine is done entirely by gradual
evaporation.

6.1.1 Definition

Evaporation can be defined as the physical process in which a liquid is changed into a gas by
molecular transfer. Evaporation occur when a liquid is not in equilibrium with its vapour or gas
phase. Evaporation is thus the process of equilibrating the vapour pressure of the liquid to the
equilibrium partial vapour pressure of the liquid in the ambient gas or air. Evaporation is
generally expressed by the fundamental equation which is also known as “The Dalton
Equation”:

76
 𝐸 = 𝐾(𝑒𝑠 − 𝑒𝑎 ) 1

where: 𝐸: m/month is evaporation rate

𝐾: m/month.Pa The factor ‘K’ in the equation is a
correction factor, which depends on the
surface area covered with water and the
water depth.
𝑒𝑠 : Pa is saturation vapour pressure at the
temperature of the water/brine surface
𝑒𝑎 : Pa is vapour pressure in the air

6.1.2 Factor affecting evaporation

A number of factors are known which affect brine evaporation, both meteorological and
physical and are discussed as below:

(i) Area of the Brine surface: The amount of evaporation is directly proportion to the area
of evaporation. If the exposed area is large, the evaporation will be more and vice versa.

(ii) Depth of the brine in brine body: The depth of brine influences the evaporation
considerably. More depth reduces the summer evaporation and increases the summer
evaporation.

(iii) Humidity : If the humidity of the atmosphere is more, the evaporation will be less as
during the process of evaporation, water vapour move from the point of high moisture
content to the point of low moisture content and the rate of this movement is governed
by the difference of their moisture content or the moisture gradient existing in the air.

(iv) Wind Velocity: The process of evaporation also depends upon the prevailing
turbulence in the air. If the turbulence is more or in other words if the velocity of the
air in contact with water surface is more, the saturated film of air containing the water
vapour will move easily , and the diffusion and dispersion of vapour will become easier,
causing more evaporation.

77
(v) Temperature: The process of temperature also depends upon the temperature. If the
temperature is more the saturation vapour pressure increases and the evaporation
increases.

(vi) Atmospheric pressure: If the atmospheric pressure (is more, naturally there will be
lesser evaporation (Dalton’s law). At higher altitudes, the atmospheric pressure is less
and hence the evaporation should normally is higher, however, this is not exactly so,
because of the decrease in temperature at higher attitudes, which reduces the
evaporation.

(vii) Vapour pressure of Brine: The equilibrium vapour pressure of pure water is well
known. The difference in vapour pressure between water at 30 degrees and 40 degrees
Celsius is an increase of 73%. Temperature increases the amount of vapour suspended
over of body of liquid (water) at equilibrium.

(viii) The temperature of a brine or water in a field situation is dependent on the incident
solar radiation as the (only) source of energy input. To maximise the temperature of an
evaporating solution, it is important to expose the water/brine to the greatest solar
radiation and retain as much of the incident radiation as possible; minimising heat loss
in reflection and absorption through the ground, with the only energy loss through
evaporation.

(ix) The area of the evaporating brine exposed to incident radiation is therefore critical; this
is normally interpreted as surface area of the evaporation pond but may not always. As
the area covered by brine increases, the greater the potential incident radiation.

(x) Loss of incident radiation through reflection is an important consideration when the
water being evaporated is brine. This is because the salt precipitated from brine
increases the reflection and therefore heat loss. This is countered commercially by the
use of dyes to increase solar absorption or by increasing the depth of the brine being
evaporated. Increasing the depth of brine reduces the chance of the solar energy being
reflected before the energy has been absorbed. The optical density of brines increase
with salinity meaning more efficient light absorption and the conversion of light to heat
is higher for a smaller depth of brine than pure water. In practice, if the depth of brine

78
 is greater than 50cm then changing the absorption and reflective characteristics of the
brine is not considered commercially viable for salt production. The criteria, of course
would be different for the current application.In theory, it is possible to lose energy
through diffusion of heat to the atmosphere. In practice this is not a major heat loss
compared to the loss of heat through water vapour or evaporation

(xi) Dissolved salts decrease the vapour pressure of an aqueous solution. Brines have a
lower equilibrium vapour pressure at the same temperature as pure water. To minimise
the effect of the equilibrium vapour pressure decreasing with salinity in commercial
solar evaporation plants, brines are separated into an evaporating sequence over a
number of ponds, from lowest salinity to highest salinity, with the highest salinity ponds
windward of the lower salinity ponds. The mean vapour pressure of the brines in ponds
in a salinity sequence is higher than the vapour pressure of sole pond with homogenous
brine (this can be demonstrated using calculus).

6.2 Measurement of evaporation

As discussed evaporation is strongly dependent of factors like wind velocity, air temperature
or relative humidity. Generally, these factors are difficult to determine, and changes in one of
these factors have a great influence on the calculated evaporation rate. Potential evaporation
rates are assessed either by using empirical formulae based on mass-transfer principles, or
empirically by means of measured evaporation rates in the considered location. The empirical
method, which integrates the effects of several climate elements, is referred to as ‘pan
evaporation’.

(a) Empirical formulae

Various empirical formulae have been suggested for estimating the rate of evaporation per day
from a large surface of water. These formulae are based on the Dalton’s law of evaporation.
The two formulae which are most commonly used are the Meyer’s and Penman equations and
are discussed as below:

79
The Meyer Equation

The Meyer Equation enables the prediction of monthly potential evaporation with regard to the
wind velocity.

𝐸 = 82.5 × 10−6 (1 + 62.14 × 10−3 𝑢)(𝑒𝑠 − 𝑒𝑎 ) 2

where: 𝐸: m/month is evaporation rate

u: km/h is wind velocity
𝑒𝑠 : Pa is saturation vapour pressure at the
temperature of the water surface
𝑒𝑎 : Pa is vapour pressure of air

The Penman Equation

Using the Penman Equation, the daily potential evaporation rate can be predicted. Wind
velocity is considered in this equation.
𝐸 = 2.6 × 10−6 (1 + 149.13 × 10−3 𝑢)(𝑒𝑠 − 𝑒𝑎 ) 3

where: 𝐸: m/day is evaporation rate

𝑢: Km/h is wind velocity
𝑒𝑠 : Pa is saturation vapour pressure at the
temperature of the water surface

𝑒𝑎 : Pa is vapour pressure in the air

(b) Pan Measurement method

The ‘Pan A’ evaporimeter as shown in Figure 6.1 is a cylindrical basin with a diameter of 47.5”
(120.7 cm) and a depth of 10” (25 cm). The evaporation is measured daily. The measurement
day begins when the pan is filled to exactly two inches (5 cm) from the top of the pan. After
24 hours, the amount of water to refill the pan to exactly the same level is measured. The
amount of water needed for refill equals the daily measured evaporation. Pan evaporation data
provided by the Government agencies (Metrological department) for the project area needs to
be made available.

80
 Fig 6.1 ‘Pan A’ evaporimeter

Evaporation values calculated by using available mass-transfer based empirical equations such
as Dalton, Meyer, Penman etc are compared with each other and with the evaporation rates
reported by ‘pan evaporation’ and accordingly sizing and design of the evaporation basin are
made.

6.3 Evaporation Pond Design

(a) Sizing of the Evaporation Basin

To determine the required surface area and height of the evaporation basin, the production rate
of brine during leaching and oil fill, the additional input due to rainfall, and the evaporation
rate of brine have to be considered. Most methods for calculating the evaporation rate consider
evaporation directly proportional to vapor pressure. Salinity reduces evaporation primarily as
vapor pressure of saline water is lower than that of fresh water, and because dissolved salt
lowers the free energy of water molecules. Cohesive forces acting between the dissolved ions
and the water molecules may also be responsible for inhibiting evaporation, making it more
difficult for the water to escape as vapor. For water saturated with sodium chloride, the solar
evaporation rate is generally about 70 % of the rate for fresh water (Joshi and Bhatt, 1983).
This percentage is referred to as the ‘salinity factor’.

81
Another factor, which has to be considered, is the ‘lake factor’ or ‘pan coefficient’. A standard
evaporation pan is ‘Class A pan’ has a diameter of only 120.7 cm and a height of 25 cm.
Because of the small heat capacity of evaporation pans, these tend to heat up and cool more
rapidly than lakes in the same area and therefore tend to evaporate at a higher rate than an
adjacent natural pond of water. In general, experience has shown the evaporation rate from
large bodies of water to be approx. 70 % of that one measured in a Class A pan (Joshi and
Bhatt, 1983). This percentage must be applied to measured pan evaporation in order to achieve
the actual lake evaporation.

(b) Location and design

The location of evaporation basin is chosen to keep it as far as possible from the surrounding
inhabitations and as close as possible to the leaching facilities to reduce the costs of brine
transport.

The total area of the brine evaporation basin is further subdivided into individual, much smaller
ponds. Expert services in surface geological mapping are required for creating large brine pits
for evaporation of brine and recovery of salt. Information about the distribution, orientation
and inter – relationship of the various rock formation at or below the surface of the earth are
gathered for the 2 dimensional spatial depiction of these data. Exposures of rocks and earth
materials in stream banks, road cuts, mines and quarries need to be examined. The geological
map thus prepared with theirs accompanying geological records are the principal source of raw
and interpreted information for nearly all-basic and advanced investigations. The map is
updated from other sources of information such as drilling and geological data. Mapping using
satellite imagery may also be taken up as the covered area is large.

Brine evaporation basin is encompassed by an earth dam of around 1.5 m height and is
connected with around 6m wide roads. The road network has an inlet ramp to enable salt
harvesting machines to enter into the ponds as shown in Figure 6.2.

82
 Fig. 6.2 Salt harvesting machines in operation

The brine evaporation basin is lined with a clay layer of around 5 cm thickness to reduce the
hydraulic conductivity of the soil. After the first evaporation cycle, a thin layer of salt and
gypsum is formed, which further prevents brine from seeping into the ground.

Also each pond is provided with a pipe inlet equipped with a hand operated valve designed for
the maximum brine flow rate.

(c) Monitoring system

A systematic monitoring system is installed to control any brine flow into the ground. The
system consists of several monitoring wells and production wells. The location of the
production wells is made within the evaporation basin. Minimum one monitoring well is
located in the groundwater upstream to obtain reference values for salt content and groundwater
level. Minimum three monitoring wells is located in the groundwater downstream to observe
and compare the groundwater parameters there to the upstream values. All wells are planned
to reach the top of the first aquiclude below ground level. The groundwater flow direction,
depth and thickness of the aquiclude is determined in the aquifer test.

If brine seeps through the clay layer of the basin, it accumulate above the aquiclude. By taking
water samples, a change in water salinity and water level can be recognized. In case of high
seeping rates, the brine is extracted by the production wells and is pumped back into
evaporation basin or is also used for the leaching process. By this, a cone of groundwater
depression is develop beneath the evaporation basin as shown in Figure 6.3. The groundwater

83
flow direction change towards the production well. Due to this, no pollution or adverse effect
is made to the surrounding aquifer or groundwater.

Fig. 6.3 Ground water monitoring layout

6.4 References

1 DEEP Underground Engineering GmbH (2009): Feasibility and Basic Engineering

Report: Strategic Petroleum Storage in Salt Caverns in NW Rajasthan, India.
2 Engineers India Limited (2013): Detailed project Report for Phase II of Strategic
Storage Program for Crude oil, Volume IV - Bikaner, Rajasthan, India.
3 B. S. Joshi and R. B. Bhatt (1983), Design and Layout of Solar Salt Works, Sixth
international Symposium on Salt, 1983 Vol. II Central Salt & Marine Chemicals
Research Institute, Bhavnagar. India
4
Government of Rajasthan, Groundwater Department (2001): Groundwater Resources
of Sri Ganganagar District Rajasthan.
5
Mark Coleman (2000), Review and Discussion on the evaporation rate of brines

84
 7 Bikaner Rock Salt: A Case Study

7.1 Project Description

The proposed site for the storage is located near Udepur, in the Ganganagar District (EIL, DPR,
2013). The site is approx. 150 km NNE of Bikaner and is accessible by NH-15 connecting
Amritsar to Kandla via Bikaner as well as by the broad-gauge rail of North Western Railways
from Suratgarh to Bikaner. The project site is located on either side of NH-15 and the nearest
railway station is Suratgarh, located approx. 20 km from the project site.

Being located on the north-western region of the country, the storage facilities are planned to
be integrated to the three refineries namely GGSRL refinery at Bhatinda, and IOCL refineries
at Panipat and Mathura. The pipeline integration scheme envisages connectivity to the Mundra
Bhatinda Pipeline (MBPL), Mundra Panipat Pipeline (MPPL) and Salaya Mathura Panipat
Pipeline (SMPL) at Dungargarh, Sanganer and Chaksu respectively. The storage facility is
planned to have access to the offshore oil terminals (SPMs) located off Mundra on the northern
coast of the Gulf of Kutch for transshipment.

With the availability of a favorable geological setting, i.e. a bedded salt formation of
appropriate depth and thickness, and of a shallow aquifer to provide sufficient yield of (slightly
saline) groundwater, underground salt caverns is the selected storage alternative for the Bikaner
area. The total envisaged storage capacity is 3.75 MM T (or 4.4 MM m³) of crude oil. For the
purpose of early commissioning, the storage facility has been designed in terms of two
‘clusters’ (A and B) of four caverns each. Each cavern has a geometrical storage volume of
550,000 m³, totalling to a capacity of 1.875 MM T for each cluster. The caverns are designed
to be located at a depth of 630 to 750 m below ground with a maximum diameter of 110 m as
shown in Figure 7.1. With an objective of having an early and phase-wise commissioning, the
cluster A caverns are planned for the first oil fill after approximately 45 months after spud-in
of the first cavern well to be followed by the first oil fill of cluster B caverns approximately 30
months later.

While the leaching operation is planned to be carried out by sourcing slightly saline ground
water from the shallow aquifer, the brine disposal has been contemplated through solar
evaporation in shallow basins and removal of solid salt. However, for commercial oil operation,

85
sourcing of saturated brine (to be used for compensation of withdrawing oil) is being secured
through a combination of ‘donor caverns’ and above ground storage in a brine pond.

The project is envisaged to be completed in a period of 84 months, which includes 15 months

duration for pre-bid engineering, tendering and award. In view of the interface requirements of
the salt cavern project, the above ground facilities contract is planned to be in place by 9 months
providing necessary work front. Concurrently based on supplementary investigations, the detail
engineering for underground facilities will be carried out. Further the construction schedule for
salt caverns is planned to be in two stages; Cluster I involving four caverns to be completed in
66 months providing a storage capacity of 1.875 MMT; followed by commissioning of the
remaining four caverns under Cluster II in 84 months.

The present DPR envisaged storage scheme designed as a single purpose project with the
objective to store 3.75 3 MMT (4.4 MM m) of crude oil. Due to the characteristics of the project
it is deemed possible that other projects could hook up with the oil storage and could be sited
in the same project area. A possible gas storage project, which is envisaged in line with the
trans-national gas pipelines would be an ideal fit and could be taken up after the leaching of
the last oil storage caverns is finished. Additional investment would be limited to the
completion of gas storage cavern wells and construction of the aboveground gas facilities, as
the brine from the leaching of the gas cavern could be used for operation of the crude oil cavern.
Further some of these caverns could also be used to store crude oil, thus increasing the total
storage capacity.

Fig.7.1 Proposed salt cavern geometry

86
7.2 Geological Description of Project Site

The proposed salt caverns are planned to be located within the bedded salt formations of
Hanseran Evaporite Group in the Nagaur-Ganganagar Basin area of Bikaner, Rajasthan which
covers much of north-western Rajasthan state and extends from near the city of Jodhpur in the
South into Haryana and Punjab states in the North. The formation is interpreted to be connected
to the evaporite sequences of the Salt Range in Punjab platform of Pakistan in the northwest
(Rath et. al, 2014).

The Nagaur-Ganganagar basin with a surface area of approx. 50,000 km2 is considered the
south most continuation of the large Eocambrian (approx. 600 million years) basin of which
Salt Range of Pakistan forms the northern part as shown in Figure 7.2. The northern extremity
of the Eocambrian basin is exposed in the Salt Rage (Pakistan) as Saline series. The sequence
denotes continuity in deposition of sediments from Late Proterozoic to Lower Cambrian age,
the Saline series has been assigned to the Eocambrain age. The recent drillings for oil & gas
wells exploration in both India and Pakistan corroborates this further (Dasgupta et al., 1988).

The bedded salt formations in the Nagaur-Ganganagar Basin area are characteristically located
within a depth range of 500m to 750m below the surface but is not exposed to the surface. The
occurrence of these deposits, its nature and thickness are well known through the exploration
for the associated potash mining by Geological survey of India. Lithologically, these formation
known as “Hanseran Evaporite Group” represents a typical example of cyclic bedded salt
deposit characterized by the occurrence of carbonate, anhydrite, and halite (rock salt) and, to a
minor degree, polyhalite, sylvite and carnallite (GSI, 2005).

Internally, the Hanseran Evaporite Group is structured into seven evaporation cycles. The
thickness of each cycle varies slightly from one well location to the other. The rock salt (mainly
halite) portions of these cycles are serially numbered H1 to H7 from bottom to top. Out of the
seven layers, H6 which have the larger thickness approx. 160 to 220 m within the project area
as well as low insoluble material content in the rock salt was found to be the most appropriate
for construction of the storage site. As apparent from the seismic survey and the litho log
correlation, H6 is also laterally continuous and not affected by the syntectonic faults.

87
 Fig. 7.2 General outline of the Nagaur-Ganganagar Basin

7.3 Strength and Rheological Properties

(a) Investigation Program

Using data from adjacent exploration wells drilled for potash mineralization by GSI as
reference, three exploration wells SPR-1, SPR-2 and SPR-3 as shown in Figure 7.3 were drilled
to the envisaged storage horizon of contiguous salt beds within the Nagaur-Ganganagar Basin
in the northwest of Rajasthan (DEEP, 2009). Well SPR-2 which encountered the evaporite
sequence “H6” at a depth of 534.7 to 753.5m below ground level (total thickness 218.8m) as
shown in Figure 7.4 was considered as a reference well for the project. A total of 24 samples
were selected from the core recovered of this exploration well to conduct uniaxial and triaxial
short term as well as uniaxial and triaxial long term tests to evaluate the strength, stress-strain
behaviour and creep properties.
`

Fig. 7.3 Location Map for exploration well

88
 Halite H6

Fig. 7.4 Forecast log for exploration well SPR-2

All specimens were cylindrical with length of around 120 mm and diameter of around 60 mm
and were prepared according to ISRM, 2013. All the tests were performed at 25° C (DEEP,
2009).

(b) Physical Property

Halite in the layer “H6” occurs in form of fine crystalline to very coarse crystalline blocky
(hyp-) idiomorphic varieties. Crystal size ranges from 4 to 60mm. As observed in the cores,
halite is mainly colorless and greyish or reddish to orange in color. Most of the colored varieties
are transparent and vitreous.

Insoluble impurities with the rock salt mainly occurs as veins, thin bands or lenses. X-ray
diffraction analysis identified anhydrite as the dominant insoluble phase in the halite layer
besides the clay fraction. Significant insoluble layers with the otherwise clean H6 horizon
consists of claystone and anhydrite and minor dolomite.

Ten samples were selected to determine the leaching behaviour of the rock salt to determine
the shape development of the proposed storage caverns and timing of the solution mining
process. The leaching testes revealed that the H6 layers generally have very low insoluble
content of around 8.5%. The test results also indicate a heterogeneous multi component system
including more than one soluble phase.

89
(c) Short Term Behavior of Rock Salt

The average weight density of the studied rock salt is 21.195 KN/ m3 which is comparable with
the reported average density from literature in the range of 20.20 to 21.6 KN/m3 (Gevantman,
1981).

Three UC-tests (Uniaxial Compression) and six TC-tests (Triaxial Compression) tests have
been performed at quasistaitc condition with deformation control, where strain rate 𝜖̇ is held
constant (0.25%/min = 4.17.10-5 s-1) at confining pressures between 0 and 8 MPa to assess
the time independent stress-strain behaviour and the strength of rock salt. For rock salt, failure
strength as determined from the peak stress is only slightly dependent of strain rate 𝜖̇ which is
different from the brittle rock (Cristescu and Hunsche, 1998). The test results of UCS and
triaxial are presented in Table 1. The data obtained from these tests reveal that confining
pressure has e a significant effect on the deformation and mechanical parameters of rock salt.

The average uniaxial compression of the Bikaner rock salt is about 31MPa and the deformation
is about 4%. The average UCS of pure rock salt generally from diapirs and from bedded strata,
is usually reported to be in the range of 15 – 41MPa (Liang et. al., 2007). At confining pressure
of 8MPa, the failure stress increased up to around 75MPa and the salt underwent about 25%
shortening before it failed to hold increasing loads.

Table 7.1 UCS and Triaxial test results

Peak stress , Axial strain at Elastic modulus

Specimen σ3 (Mpa)
(Mpa) rupture (%) (Gpa)
1 0 30.61 3.33 17.27
2 0 31.2 4.54 16.67
3 0 31.87 4.54 17.00
4 2 30.86 5.06 13.93
5 2 51.72 12.5 15.67
6 4 54.93 11.94 18.57
7 4 61.2 15.57 16.78
8 6 68.34 22.35 21.15
9 8 74.52 25.79 21.53

A typical stress-strain curve of uniaxial and triaxial test results at confining pressure of 2MPa
are shown in Figure 7.5. Compared with the stress-strain curve of ordinary rock samples,

90
during the uniaxial compression tests rock salt does not have an obvious fracture closure
behavior, probably because the initial porosity in rock salt is very low.

As shown in Figure 7.5, after reaching the initial specified axial compression of around 1%,
the next phase of UCS and triaxial test involved a stress-controlled stress release and stress
buildup cycle to determine the deformation modulus. The axial stress is then increased under
constant strain rate until the failure load is reached. The average Young's modulus was
measured as 17.6 GPa. The elastic constants for salt are nearly independent of the source of
salt (Hansen et. al., 1984) and vary in the range from 19 GPa to 30 GPa. Under high pressure,
noticeable effect of pressure on the elastic modulus could be observed. At the constant
temperature, the elastic modulus increases with increasing pressure. But the decreasing rate of
the increase indicates that the effect of pressure on elastic modulus becomes relatively lower
as the effective pressure increases

No measurement were made for poisson’s ratio , however, reported observations for poisson’s
ratio indicates a range of 0.29 to 0.38 and average value of 0.35 (Hansen et. al., 1984).

60

50 3 = 2MPa
Stress (MPa)

40

30 3 = 0

20

10

0.0
2 4 6 8 10 12 14 16
Strain (%)

Fig. 7.5 A typical Stress-strain curve in UCS and triaxial condition at 𝜎3 = 2 MPa

Figure 7.6 shows test results for the influence of the confining pressure 𝜎3 on the strength and
ductility of rock salt. The peak strength and deformation at failure increases considerably with
increasing confinement. Whereas there is a significant strength increment between values of
𝜎3 = 0, 2, 4 and 6 MPa, and the variation decreased at 8 MPa.

91
 𝜎3 = 8 MPa
𝜎3 = 4 MPa 𝜎3 = 6 MPa

𝜎3 = 2 MPa

UCS

Fig. 7.6 Results of uniaxial and triaxial compression tests on rock salt

Sufficient data were available to construct the Mohr’s circle diagram for stresses at failure
presented in Figure 7.7. Typically, ultimate shear strength increased to approximately 30 MPa
at the highest confining pressure pressures of 8 MPa. Rock salt at lower confining stresses and
rapid strain rates (> 4 x 10-5 s-1) behaves as a Mohr- Coulomb material (Liang et. al., 2007).
Fully ductile materials such as rock salt under extremely high 𝜎3 (>20 MPa) behave as purely
plastic material. The values for the cohesion and angle of friction were determined around 3.24
MPa and 53° from the linear Mohr-Coulomb Failure envelope. The average cohesion and angle
of friction of rock salt varies in the range of approximately 2.0-5.0 MPa and 55-60° respectively
(Hansen et.al., 1984).

Fig. 7.7 Mohr-coulomb plot for tested rock samples

92
(d) Failure Criteria

For rock salt, the failure strength (ultimate stress) as determined by the peak stress from
experimental results and predicted from the available strength criteria with variable parameters
as well as the coefficient of determination R2 are shown in Table 7.2 and Figure 7.8. As can
be seen in Figure 7.8, the linear Mohr-columb criteria fit poor to the test data. The failure
criteria tend to overestimate the strength value in the high confining compression region and to
underestimate the failure strength in the low confining pressure region. The Hoek-Brown
criterion with the best fit parameters goes right through most of the data points and reproduces
the trend of the test data closely for the whole range of loading condition.

Table 7.2 Estimated and observed failure strength of rock salt

Mohr- Coulomb Criterion Hoek-Brown Criterion
1 sn 2𝑐. s 𝜎
𝜎1 = 𝜎3 + 𝜎1 = 𝜎3 + 𝜎𝑐 𝑚. +𝑠
Observed 1−s n 1−s n 𝜎
(Mpa) cohesion c Mpa 3.5 5 mi 14 mi 12 mi 15
Angle of friction ϕ ° 53 46 s 1 s 1 s 1
σ3 σ1 σ3 σ1 σ3 σ1 σ3 σ1 σ3 σ1 σ3 σ1
-1.872 0 -2.34 0.00 -4.04 0.00 -2.22 0 -2.58 0 -2.07 0
0 31.2 0 20.92 0 24.75 0 31.20 0 31.20 0 31.20
2 41.29 2 38.79 2 37.00 2 44.98 2 43.50 2 45.70
4 58.06 4 56.65 4 49.26 4 56.16 4 53.71 4 57.34
6 68.34 6 74.51 6 61.51 6 65.95 6 62.74 6 67.49
8 74.52 8 92.38 8 73.76 8 74.84 8 71.00 8 76.68
R2 0.6438 0.8603 0.9826 0.9489 0.9808
Least Square 9.7054 6.0789 2.1454 3.6753 2.2510

Fig. 7.8 Best fitting solutions of failure criteria for the bedded rock salt

93
(e) Rheological Behavior of Rock Salt

Nine UCc-tests (Uniaxial Compression creep) with a total of 18 load steps and 5 TCc-tests
(Triaxial Compression creep) with a total of 10 load steps were performed to determine the
creep deformation of rock salt on stress. Depth and stress dependent effects to the creep
behaviour of the rock salt are taken into account by executing uniaxial and triaxial creep tests
with different load levels.

A typical result from a long term creep test is shown in Figure 7.9, which is commonly
represented by elastic deformation followed by the sequence of specific time dependent
deformations i.e. the primary transient phase where the strain rate decreases with time and the
steady-state creep where the strain rate remains essentially constant. After the initial 30 days
of creep test, deviatoric stress was further increased which results in another elastic as well as
transient creep phase followed by another steady-state creep. Tertiary creep which is typically
characterized by a rapidly accelerating strain rate to ultimate fracture failure has not been taken
under the scope of the present paper.

transient phase
1.4

1.2

1.0  = 20MPa
Axial Strain (%)

0.8

0.6  = 15MPa

0.4
steady state
0.2 elastic phase phase

0.0
10 20 30 40 50 60 70
Time (d)

Fig. 7.9 A typical creep tests of rock salt

The steady state strain rate determined from the creep tests under uniaxial and triaxial
compression are presented in Table 7.3. The table shows the stationary creep rate 𝜖 𝑠̇ (which is
the main feature of the specific time-dependent deformation behaviour of each of the
specimens) estimated as the average of the readings and taken from the shape of the
measurement curves over specific time periods. The average steady state creep rate under the
deviatoric stress of 8MPa is around 1.10 E-05 while under 10MPa is 2.13 E-05 under the UCS

94
condition while the average steady state creep rate under triaxial compression with confining
pressure of 5MPa depending upon deviatoric stress varies from 1.25 E-05 to 2.95E-05. The
results shows that steady-state strain rate appears not to be sensitive to confining pressure and
the same has also been reported by many researchers including Cristescu and Hunsche, 2008.

Table 7.3 Test results from uniaxial and triaxial creep test

UCS crep tests Triaxial Creep Tests at 𝜎2,3 = 5MPa

𝜎1
Sample No. Sample No.
MPa (1/d) MPa (1/d)
1 08 / 10 1.00E-05 1.18E-05 15 2.00E-05
1
2 08 / 10 5.90E-06 6.36E-06 20 1.87E-05
3 08 / 10 5.00E-06 5.83E-06 15 2.50E-05
2
4 10 / 12 1.00E-05 1.00E-05 20 4.00E-05
15 1.50E-05
5 10 / 12 4.50E-06 8.18E-06 3
20 3.00E-05
6 10 / 12 5.00E-06 7.73E-06
12 1.30E-05
7 07 / 10 2.40E-05 5.83E-05 4
17 2.40E-05
8 07 / 10 1.00E-05 3.75E-05 12 1.20E-05
5
9 07/ 10 2.50E-05 4.58E-05 17 3.50E-05

Figure 7.10 lustrates a plot between steady state (SS) creep rate versus deviatoric stress. It
shows that the SS creep strain rate increases with increases deviatoric stress which have been
reported by many researcher. The steady state creep rate of the Bikaner salt are comparable to
strain rate reported in the literature (Cristescu and Hunsche, 1998).

Fig. 7.10 SS strain rate versus deviatoric stress q = (σ1 – σ3)

95
Figure 7.11 shows a typical multiple creep test results at different deviatoric stress and with
confining pressure of 0 and 5 MPa. Though the confining pressure have limited effect on creep
rate, however, the time to reach steady-state creep increases with increasing deviatoric stress
as well as increasing confining pressure. As illustrated, time to reach SS creep rate under UCS
condition is around 10 days which increases up to 20 days with a confining pressure of 5MPs.
This may be probably because a considerable time lapse is needed to adjust to the new stress
state when the stress changes, especially when the change is very large (Chao et. al, 2012).

Fig. 7.11 Multi-Step Triaxial Creep Tests at different confining pressure

(f) Constitutive modelling of Bikaner Rock Salt

The mathematical–mechanical plot of creep behaviour using the Lubby2 material model is
achieved using a bipartite approach where the creep deformation observed in the lab tests is
divided into transient and stationary creep deformation (Lux and Heusermann, 1983). The total
creep rate 𝜀̇1𝑣 and the additive superposition of the transient creep rate 𝜀̇1𝑡𝑟 with time or strain
hardening and a constant stationary creep rate 𝜀̇1𝑠𝑡 are calculated in accordance with the
following equation

1 εν,𝑡
1 . 3𝐺𝑘(𝜎𝑣 ) 1
𝜀̇1𝑣 = 𝜀̇1𝑡𝑟 + 𝜀̇1𝑠𝑡 = [ 3𝜂 . {1 − }+ ] . 𝜎𝜐 (7.1a)
𝑘(𝜎𝑣 ) 𝜎𝜐 3𝜂𝑀(𝜎𝑣,𝑇)

96
or

1 εν,𝑡 ̅
1 . 𝐺𝑘(𝜎𝑣 ) 1
𝜀̇1𝑣 = 𝜀̇1𝑡𝑟 + 𝜀̇1𝑠𝑡 = [ 𝜂̅ . {1 − }+ ̅𝑀(𝜎𝑣,𝑇)
] . 𝜎𝜐 (7.1 b)
𝑘(𝜎𝑣 ) 𝜎𝜐 𝜂

For abbreviation
3𝜂𝑀(𝜎𝑣, 𝑇) = ̅𝜂𝑀(𝜎𝑣 ,𝑇) , 3𝐺𝑘(𝜎𝑣) = ̅𝐺𝑘(𝜎𝑣) , 3𝜂𝑘(𝜎𝑣) = ̅𝜂𝑘(𝜎𝑣)
Where
∗
𝜂̅𝑀(𝜎,𝑇) = 𝑀𝑎𝑥𝑤𝑒𝑙𝑙 viscosity coefficient (MPa . d) = 𝜂̅ 𝑀 . 𝑒𝑥𝑝( 𝑚.𝜎𝜐) . 𝑒𝑥𝑝( 𝑙 .𝑇)
∗
𝐺̅𝑘(𝜎) = 𝐾𝑒𝑙𝑣𝑖𝑛 shear modulus (MPa) = 𝐺̅𝑘 . . 𝑒𝑥𝑝( 𝑘1 𝜎𝜐)
𝜂̅𝑘(𝜎) = 𝐾𝑒𝑙𝑣𝑖𝑛 viscosity modulus (MPa . d) = 𝜂̅𝑘∗ . 𝑒𝑥𝑝( 𝑘2 𝜎𝜐)
𝜀̇1𝑣 = viscous creep rate (d-1), 𝜀̇1𝑡𝑟 = transient creep rate (d-1), 𝜀̇1𝑠𝑡 = stationary creep rate (d-1)
1
𝜎𝜐 = Equivalent stress (MPa) = √2 [( 𝜎1 − 𝜎2 )2 + ( 𝜎1 − 𝜎3 )2 + ( 𝜎2 − 𝜎3 )2 ]

Special case : 𝜎𝜐 = |𝜎1 – 𝜎3 | 𝑇𝑟𝑖𝑎𝑥𝑖𝑎𝑙 𝑐𝑎𝑠𝑒 𝑤ℎ𝑒𝑛 𝜎2 = 𝜎3 and 𝜎𝜐 = |𝜎1 | Uniaxial state of
stress when 𝜎2 = 𝜎3 = 0
T = Temperature (K)
𝑘1 , 𝑘2 ,𝑚, l =Material Parameters

Parameter for stationary creep

The Maxwell viscosity coefficient 𝜂̅𝑀 is defined as the quotient of the equivalent stress 𝜎𝜐 and
the stationary creep rate 𝜀̇1𝑠𝑡 ∶

𝜎𝜐
𝜂̅𝑀(𝜎,𝑇) = (7.2)
𝜀̇ 1𝑠𝑡

𝜂̅𝑀 is determined for each single test (each load stage) on the basis of the creep rate readings
determined during the stationary creep phase as shown in Table 7.4 .

97
 Table 7.4 stationary creep rate and calculation of Maxwell viscosity coefficient 𝜂̅𝑀

Depth of core confing pressure Deviatoric stress equivalnet stress strain rate Maxwell viscosity
Sample no. (m) σ3 (Mpa) σ1 (Mpa) σ1 (Mpa) έ s (1/d) coefficient ῆM (Mpa d)
5 15 10 2.00E-05 5.00E+05
11 547
5 20 15 1.87E-05 8.02E+05
5 15 10 2.50E-05 4.00E+05
12 553
5 20 15 4.00E-05 3.75E+05
5 15 10 1.50E-05 6.67E+05
13 622
5 20 15 3.00E-05 5.00E+05
5 12 7 1.30E-05 5.38E+05
14 622
5 17 12 2.40E-05 5.00E+05
5 12 7 1.20E-05 5.83E+05
15 663
5 17 12 3.50E-05 3.43E+05
8.2 8.2 1.00E-05 8.20E+05
16 547
10.1 10.1 1.18E-05 8.56E+05
8 8 5.90E-06 1.36E+06
17 553
10 10 6.36E-06 1.57E+06
8 8 5.00E-06 1.60E+06
18 553
10 10 5.83E-06 1.72E+06
10.3 10.3 1.00E-05 1.03E+06
19 553
12.3 12.3 1.00E-05 1.23E+06
0 10.5 10.5 4.50E-06 2.33E+06
20 591
12.5 12.5 8.18E-06 1.53E+06
9.3 9.3 5.00E-06 1.86E+06
21 591
11.3 11.3 7.73E-06 1.46E+06
6.7 6.7 2.40E-05 2.79E+05
22 703
9.5 9.5 5.83E-05 1.63E+05
7 7 1.00E-05 7.00E+05
23 702
10 10 3.75E-05 2.67E+05
7.5 7.5 2.50E-05 3.00E+05
24 738
10.5 10.5 4.58E-05 2.29E+05

The material parameters for the stationary creep can be expressed as

∗
𝜂̅𝑀(𝜎,𝑇) = 𝜂̅ 𝑀 . 𝑒𝑥𝑝( 𝑚.𝜎𝜐) . 𝑒𝑥𝑝( 𝑙 .𝑇) (7.3 a)

For constant temperature

∗
𝜂̅𝑀(𝜎,𝑇) = 𝜂̅ 𝑀 . 𝑒𝑥𝑝( 𝑚.𝜎𝜐) (7.3 b)

Logarithmize equation (7.3 b)

∗ )
ln ( 𝜂̅𝑀(𝜎,𝑇) ) = ln( ̅𝜂𝑀 + 𝑚. 𝜎𝜐

y = b + m.x (linear equation)

Δ𝑦 ̅𝑀(𝜎,𝑇)
Δ ln 𝜂
Gradient 𝑚 = =
Δ𝑥 Δ 𝜎𝜐

98
 ∗
The parameters 𝜂̅𝑀 in equation is the intercept of the linear correlation between 𝜎𝑣 and 𝜂̅𝑀(𝜎,𝑇)
and can be directly determined from the exponential regression analysis as shown in Figure 12.

1.00E+07
maxwell -viscocity coefficient (Mpa d)

1.00E+06

1.00E+05

1.00E+04

1.00E+03

1.00E+02

1.00E+01

1.00E+00
0 5 10 15 20 25 30 35
equivalent stress (MPa)
Uniaxial
Figure 1: Comp. Creep of steady state
Determination Triaxial Comp. Creep
parameters
Fig. 7.12 determination of steady state parameter

The steady state parameters in accordance with the plot as shown in figure 1 are as follows:

∗
𝜂̅𝑀 = 2.00 𝑒 + 06 𝑀𝑃𝑎. 𝑑
𝑚 = −0.1 𝑀𝑃𝑎 −1

The spread of the determined stationary creep rate could be made by taken into account the
depth of the samples as shown in figure 7.13. The rock salt taken from the depth of 660m to
740m arises a higher creep ability than that taken from a depth between 550m to 620m.

99
 1.00E+00

1.00E-01

1.00E-02
stationary creep rate (1/d)

1.00E-03

1.00E-04

1.00E-05

1.00E-06

1.00E-07

1.00E-08
1 10 100
equivalnet stress (MPa)
550-620 650-750

Fig. 7.13 Steady creep rate as a function of depth

Parameters for Transient Creep

The parameters 𝐺̅𝑘∗ , 𝑘1 , 𝜂̅𝑘∗ and 𝑘2 is determined to characterise the transient creep with the
Lubby2 material model

(a) Kelvin shear modulus 𝐺̅𝑘 and material parameters 𝐺̅𝑘∗ , 𝑘1

The Kelvin shear modulus 𝐺̅𝑘 is defined as the quotient of the stress 𝜎𝜐 and the transient creep
strain for t = ∞

𝜎𝜐
𝐺̅𝑘(𝜎) = 𝑡𝑟 (7.4)
𝜀1∞

Where

𝑡𝑟
𝜀1∞ = transient creep strain for t = ∞ and is calculated for the first loading stage in accordance
with

100
 𝑡𝑟
𝜀1∞ = 𝜀1𝑣 - 𝜀𝑒𝑙 - 𝜀̇ 1𝑠𝑡 . Δ𝑡 (7.5)

𝜀1𝑣 = total creep strain in the first loading stage,

𝜀𝑒𝑙 = elastic strain as a result of deviatoric load,
𝜀̇ 1𝑠𝑡 = stationary creep rate within the first loading stage,
Δ𝑡 = duration of first loading stage
𝑡𝑟
𝜀1∞ is calculated in an analogous way for the subsequent load stage 2 from equation by also
taking into consideration the stationary creep strain during the first load stage. Table 7.5
compiles the transient creep strain and kelvin shear modulus for each single test

Table 7.5 transient creep strain and calculation of kelvin shear modulus 𝐺̅𝑘(𝜎)
Kelvin shear
Depth of core confing pressure Deviatoric stress equivalnet stress Transient strain modulus
Sample no. (m) σ3 (Mpa) σ1 (Mpa) σ1 (Mpa) ε1∞tr Ḡ k (Mpa)
5 15 10 8.11E-03 1.23E+03
11 547
5 20 15 1.22E-02 1.23E+03
5 15 10 5.95E-03 1.68E+03
12 553
5 20 15 1.00E-02 1.50E+03
5 15 10 5.90E-03 1.69E+03
13 622
5 20 15 9.85E-03 1.52E+03
5 12 7 6.49E-03 1.08E+03
14 622
5 17 12 9.31E-03 1.29E+03
5 12 7 1.01E-02 6.93E+02
15 663
5 17 12 1.34E-02 8.96E+02
8.2 8.2 2.95E-03 2.78E+03
16 547
10.1 10.1 3.40E-03 2.97E+03
8 8 1.30E-03 6.15E+03
17 553
10 10 1.50E-03 6.67E+03
8 8 1.00E-03 8.00E+03
18 553
10 10 1.38E-03 7.25E+03
10.3 10.3 2.24E-03 4.60E+03
19 553
12.3 12.3 3.04E-03 4.05E+03
0 10.5 10.5 1.37E-03 7.66E+03
20 591
12.5 12.5 1.67E-03 7.49E+03
9.3 9.3 1.11E-03 8.38E+03
21 591
11.3 11.3 1.37E-03 8.25E+03
6.7 6.7 8.29E-03 8.08E+02
22 703
9.5 9.5 1.26E-03 7.54E+03
7 7 7.20E-03 9.72E+02
23 702
10 10 1.03E-02 9.71E+02
7.5 7.5 6.38E-03 1.18E+03
24 738
10.5 10.5 9.45E-03 1.11E+03

The kelvin shear material parameters 𝐺̅𝑘∗ and 𝑘1 are derived from the exponential formulation

101
 ∗
𝐺̅𝑘(𝜎) = 𝐺̅𝑘 . 𝑒𝑥𝑝( 𝑘1 𝜎𝜐) (7.6)

Logarithmize equation (7.6)

∗
ln (𝐺̅𝑘(𝜎) ) = ln 𝐺̅𝑘 + 𝑘1 . 𝜎𝜐

y = b + m.x (linear equation)

Δ𝑦 Δ ln 𝐺̅𝑘(𝜎)
Gradient 𝑘1 = =
Δ𝑥 Δ 𝜎𝜐

The parameters 𝐺̅𝑘∗ in equation is the intercept of the linear correlation between 𝜎𝑣 and 𝐺̅𝑘(𝜎)
and can be directly determined from the exponential regression analysis as shown in Figure
7.14.

1.00E+05

1.00E+04
kelvin shear modulus (MPa)

1.00E+03

1.00E+02

1.00E+01

1.00E+00
0 5 10 15 20 25 30 35
equivalent stress MPa

∗
̅ 𝑘 , 𝑘1
Fig. 7.14 Kelvin shear material parameters 𝐺
The kelvin shear parameters in accordance with the plot as shown in figure 7.14 are as follows:

𝐺̅𝑘∗ = 4000 𝑀𝑃𝑎

𝑘1 = −0.04 𝑀𝑃𝑎 −1

102
Kelvin viscosity modulus η𝑘 and material parameters 𝜂̅ 𝑘∗ , 𝑘2

The Kelvin viscosity modulus η𝑘 is dependent on the kelvin shear modulus, the equivalent
stress and the transient creep strain (stain hardening approach). The viscosity modulus is
calculated by

𝐺̅𝑘(𝜎) . 𝑡 𝐺̅𝑘(𝜎) . 𝑡
𝑛̅𝑘(𝜎) = 𝜀𝑡𝑟
= ̅
𝐺 𝑘(𝜎)
(7.7)
1
ln(1− 𝑡𝑟 ) ln(1− 𝜀1𝑡𝑟 . )
𝜀1∞ 𝜎𝜐

Where,
𝜀1𝑡𝑟 = Transient creep strain at time t
𝜎𝜐 = Equivalent stress (MPa)
𝐺̅𝑘(𝜎) = 𝐾𝑒𝑙𝑣𝑖𝑛 shear modulus (MPa)
𝜂̅𝑘(𝜎) = 𝐾𝑒𝑙𝑣𝑖𝑛 viscosity modulus (MPa . d)

Unlike the kelvin shear modulus and the equivalent stress which can be directly determined
from the creep test results, determining the transient creep strain at time t can only be done
indirectly by back calculating in accordance with equation 8 . This involves reading the total
creep strain 𝜀1 v measured during a test from the test curve at a specific point in time and
reducing it by the amount of stationary creep strain 𝜀1 st calculated for this point in time.
Calculation of the stationary creep strain at time t involve rearranging equation 2 and could be
rewritten as equation 9.

𝜀1 tr = 𝜀1 v - 𝜀1 st (7.8)

𝜎𝜐 𝜎𝜐
𝜂̅𝑀(𝜎,𝑇) = 𝑖. 𝑒. ε1st = 𝜂̅ .𝑡 (7.9)
𝜀̇ 1𝑠𝑡 𝑀(𝜎,𝑇)

103
 Table 7.6 transient creep strain and calculation of kelvin viscosity modulus 𝑛̅𝑘(𝜎)
Kelvin viscosity
Depth of core confing pressure Deviatoric stress equivalnet stress Transient strain modulus
Sample no. (m) σ3 (Mpa) σ1 (Mpa) σ1 (Mpa) ε1 tr ῆk (Mpa d)
5 15 10 8.50E-03 2.98E+03
11 547
5 20 15 1.30E-02 2.90E+03
5 15 10 6.55E-03 4.57E+03
12 553
5 20 15 1.17E-02 3.83E+03
5 15 10 6.40E-03 4.62E+03
13 622
5 20 15 1.12E-02 3.64E+03
5 12 7 6.80E-03 2.96E+03
14 622
5 17 12 1.04E-02 2.71E+03
5 12 7 1.04E-02 1.64E+03
15 663
5 17 12 1.47E-02 1.60E+03
8.2 8.2 3.52E-03 -
16 547
10.1 10.1 4.39E-03 4.35E+03
8 8 1.76E-03 1.24E+04
17 553
10 10 2.25E-03 -
8 8 1.44E-03 1.74E+04
18 553
10 10 2.07E-03 1.57E+04
10.3 10.3 2.80E-03 1.21E+04
19 553
12.3 12.3 3.96E-03 1.15E+04
0 10.5 10.5 1.88E-03 2.45E+04
20 591
12.5 12.5 2.50E-03 1.61E+04
9.3 9.3 1.58E-03 2.54E+04
21 591
11.3 11.3 2.14E-03 2.06E+04
6.7 6.7 8.70E-03 2.58E+03
22 703
9.5 9.5 1.38E-02 2.40E+03
7 7 7.20E-03 3.51E+03
23 702
10 10 1.10E-02 3.20E+03
7.5 7.5 6.70E-03 3.85E+03
24 738
10.5 10.5 1.08E-02 3.55E+03

The kelvin viscosity moduli material parameters 𝜂̅𝑘∗ , and 𝑘2 are derived from the exponential
formulation

𝜂̅𝑘(𝜎) = 𝜂̅𝑘∗ . 𝑒𝑥𝑝( 𝑘2 𝜎𝜐) (7.10)

Logarithmize equation (7.10)

ln (𝜂̅𝑘(𝜎) ) = ln 𝜂̅𝑘∗ + 𝑘2 . 𝜎𝜐

y = b + mx (linear equation)

Δ𝑦 ̅𝑘(𝜎)
Δ ln 𝜂
Gradient 𝑘1 = =
Δ𝑥 Δ 𝜎𝜐

104
The parameters 𝜂̅𝑘∗ in equation is the intercept of the linear correlation between 𝜎𝑣 and
𝜂̅𝑘(𝜎) and can be directly determined from the exponential regression analysis as shown in
Figure 7.15.

The kelvin viscosity parameters derived in accordance with the plot as shown in figure 4 are as
follows:

𝜂̅𝑘∗ = 10000 𝑀𝑃𝑎 𝑑

𝑘2 = −0.05 𝑀𝑃𝑎 −1

1.00E+05

1.00E+04
kelvin- viscocity modulus (MPa-d)

1.00E+03

1.00E+02

1.00E+01

1.00E+00
0 5 10 15 20 25 30
equivalent stress (MPa)

Fig. 7.15 Kelvin viscosity material parameters ̅𝜂∗𝑘 , and 𝑘2

105
7.4 References

1 Bays, C.A., (1963) Use of Salt Solution Cavities for Underground Storage, Symp. on Salt, Northern Ohio
Geol Soc. 564.
2 Chao, D., Chunhe, Y.,Yuanfeng, Y., Zhe, and L.,Jie C., (2012) Mechanical Behavior of Deep Rock Salt
under the Operational Conditions of Gas Storage, International Journal of Earth Science and Engineering,
P.P.1670-1676.
3 Cristescu, N., and Hunsche, U., (1998) Time Effects in Rock Mechanics — Series: Materials, Modelling
and Computation. John Wiley and Sons, Chichester, 342 pp.

4 Dasgupta, S.P., Virendra, K., Ramchandra, and Jairam, M.S., (1988) A frame work of the Nagaur-
Ganganagar evaporite basin, Rajasthan, Indian Minerals Vol. 42(1): 57-64.

5 DEEP Underground Engineering GmbH (2009): Feasibility and Basic Engineering Report: Strategic
Petroleum Storage in Salt Caverns in NW Rajasthan, India.
6 Engineers India Limited (2013) Detailed project Report for Phase II of Strategic Storage Program for Crude
oil, Volume IV - Bikaner, Rajasthan, India.
7 Geological Survey of India (2005) Special Publication No.62, Geology and evolution of Nagaur Ganganagar
basin with special reference to salt and potash mineralisation.

8 Gevantman, L. H., (1981) Physical Properties Data for Rock Salt, U.S. Department of Commerce; National
Bureau of Standards monographs.
9 Hansen, F. D., Mellegard, K. D., and Scnseny, P. E., (1984) Elasticity and strength of ten natural rock salts,
First Conf. on the Mechanical Behavior of Salt. University Part, PA, 1981, pp. 71-83.

10 ISRM (2013) Suggested Methods for Determining the Creep Characteristics of Rock, Rock Mech Rock Eng
(2014) 47: 275–290.
11 Liang, W., Yang, C., Zhao, Y., Dusseault, M. B., and Liu, J. (2007) Experimental Investigation of
mechanical properties of bedded salt rock, International Journal of Rock Mechanics and Mining Science,
44, 400-411.
12
Rath, R., Wippich, M., Nanda, A., and Wilke, H.F. (2014) Underground salt cavern storages for crude oil at
Bikaner, India. SMRI Tech Conf Proc. Netherlands; 124-135.

13 Richner, D.R., Shock, D.A., Ahlness, J.K., Tweeton, D. R. Larson, W.C., Millenacker, D.J., and Schmidt,
R.D., (1992) Solution Mining: In situ Techniques, SME Mining Engineering Handbook, 2nd Ed., V2, 1493-
1528.

106
Acknowledgement

The author is greatly indebted to the management of EIL for their support and encouragement
in preparation of this capacity building exercise in salt cavern technology.

The author would also like to thank Dr. Atul Nanda, Dr. Ranjit Rath and Sh. N.S. Vasudev for
their invaluable suggestion, supervision and critical observation.

107