PROJECT REPORT

ON

NATURAL GAS HANDLING AND IT’S QUALITY

ENHANCEMENT
IN LIGHT OF
OIL AND NATURAL GAS CORPORATION
GAS PROCESSING PLANT
HAZIRA, SURAT(GUJRAT)

DURATION: 24/05/2018-25/06/2018

Submitted to: Submitted by:

Ms. Vishakha Singh Rajkishan Gupta
DGM (Production) B. Tech ( 2nd Year )
ONGC Corporate Office Chemical Engineering
New Delhi. RGIPT, Jais, Amethi.

1|Page
 ACKNOWLEDGEMENT
Classroom learning is good and of course is necessary to build foundation and
create a distorted image of what you have been taught but to more clarify the
mind-maps, concepts an industrial training is very much necessary. Industrial
training gives us the opportunity to apply the knowledge which we have learnt.
It gives us a practical feel and of course excitement but, very few get this
opportunity to have an industrial exposure. I was one of those lucky learners
who was given this opportunity of industrial training at ONGC, a Maharatna
company of Indian Govt.
Acknowledgement is the best column to express our gratitude for the learning
platform provided to us and to those who all have been a medium for my
learning.
The completion of this report would not have been possible without the
valuable support of the ONGC family. It exhilarates me to write a report on the
topic “Natural Gas Handling and Its Quality Enhancement”.
I would like to pay my deepest gratitude to my mentor ‘Ms. Vishakha Singh’,
Deputy General Manager (Production) who gave me such an interesting topic
to work upon, she in spite of being immensely busy at her work took my training
very seriously and gave me necessary guidance, technical manuals and arranged
all facilities to make my life easier. She had a very keen interest to know my
learning progress and used to regularly ask if I had any doubt or I wanted any
help. I am very grateful to her. She made my existence there to be learning
oriented rather than a training full of formalities.
I would like to pay my special thanks to my parents for their moral support and
who gave easy permission to carry out my project at ONGC.
Finally, I apologise to all other unnamed who helped me in various ways to have
a good training.

I perceive this opportunity as a big milestone in my career development. I will

strive to use gained skills and knowledge in the best possible way, and I will
continue to work on their improvement, in order to attain desired career
objectives. Hope to continue cooperation with all of you in the future too.

Rajkishan Gupta

2|Page
 PREFACE
This project report deals with the field operations and onshore natural gas handling systems
and various processes involved to make sour natural gas upgraded and up to the mark for
efficient selling and transportation. In the report the ONGC Hazira Gas Plant has been taken
for reference. The main objective of this report was to study on the various types of
Dehydration processes available in industry, their execution and applicability depending upon
the feed gas, capital and space availabilities. This report talks about the principles,
mathematics, physics, chemistry calculations and the technology involved in the various
processes based on the various textbooks and research papers available on the internet. This
report is the short compilation of insights of the process; with the help of a practical training
and this report, one could find himself to be the master in the Gas upgrading processes.
The project although was to calculate the pre-requisites of the processes like Hydrate
Inhibition, Separators, TEG Dehydration, Molecular Sieve Dehydration in that the objective
was to calculate amount of inhibitor needed in hydrate inhibition, Glycol circulation rate in
TEG dehydration, etc. based on the feed gas specifications and product specifications which
was available in the technical manual provided by the ONGC and in few reference sources.
The various formulas involved in the calculations have been sufficiently mentioned in the
report. In spite of giving a detailed and elaborated description of the Dehydration processes
the report also has short compilation of the gas handling process right from field operations
to its sweetening and dehydration for its quality enhancement and easy transportation. No
other processes after gas dehydration has been talked in the report for example, sulphur and
hydrocarbon recovery processes.
The reference book mostly used for this report was “Fundamentals of Natural Gas Processing”
by Taylor and Francis Group. The various other references used have been clearly mentioned
at the last. I pay my sincere thanks for the knowledge I obtained from those sources and the
images and contents as well that I put down directly in the report.

3|Page
 TABLE OF CONTENTS:
1. INTRODUCTION TO ONGC
2. INTRODUCTION TO HAZIRA GAS PROCESSING PLANT
3. FIELD OPERATIONS AND INLET RECEIVING
3.1 FIELD OPERATIONS
3.1.1 WELLHEAD OPERATIONS
3.1.2 PIPING
3.1.3 COMPRESSION STATIONS
3.1.4 PIGGING
3.3 FLOW ASSURANCE
3.3 INLET RECEIVING
3.3.1 SLUG CATCHER
3.3.2 SEPARATOR PRINCIPLES
3.3.3 GAS LIQUID SEPARATORS
3.3.4 SLUG CATCHER AT HAZIRA PLANT
3.3.5 PRIMARY SEPARATION
3.3.6 GRAVITY SEPARATION
4. GAS SWEETENING/TREATING UNIT
4.1 INTRODUCTION
4.2 HAZIRA GAS SWEETENING PLANT
5. GAS DEHYDRATION UNIT
5.1 INTRODUCTION
5.2 VARIOUS DEHYDRATION UNITS
5.2.1 ABSORPTION METHOD
5.2.2 MOLECULAR SIEVE DEHYDRATION METHOD
5.2.3 JOULE-THOMPSON EXPANSION METHOD
5.2.4 OTHER PROCESSES
5.2.4.1 DESSICANT PROCESS
5.2.4.2 REFERIGERATION PROCESS
6. REFERENCES

4|Page
 1. INTRODUCTION TO ONGC
Oil and Natural Gas Corporation Limited (ONGC) is an
Indian Multinational Oil and Gas company headquartered at Deendayal Urja Bhavan (also
called Green Building), Vasant Kunj, New Delhi 110070 India. It is a Public Sector
Undertaking (PSU) of the Government of India, under the administrative control of
the Ministry of Petroleum and Natural Gas. It is India's largest oil and gas exploration and
production company. It produces around 70% of India's crude oil (equivalent to around 30%
of the country's total demand) and around 62% of its natural gas.
Maharatna ONGC is the largest crude oil and natural gas Company in India, contributing
around 70 per cent to Indian domestic production. Crude oil is the raw material used by
downstream companies like IOC, BPCL, and HPCL to produce petroleum products like Petrol,
Diesel, Kerosene, Naphtha, and Cooking Gas-LPG.
This largest natural gas company ranks 11th among global energy majors (Platts). It is the
only public sector Indian company to feature in Fortune’s ‘Most Admired Energy Companies’
list. ONGC ranks 18th in ‘Oil and Gas operations’ and 183rd overall in Forbes Global 2000.
Acclaimed for its Corporate Governance practices, Transparency International has ranked
ONGC 26th among the biggest publicly traded global giants. It is most valued and largest E&P
Company in the world, and one of the highest profit-making and dividend-paying enterprise.

Figure-1- ONGC Corporate Office, New Delhi.

5|Page
 2. INTRODUCTION TO HAZIRA GAS
PROCESSING COMPLEX
Natural gas has gained increased importance in the recent past by virtue of its usage
and a good substitute for coal, petrol and diesel as fuel in industrial boilers and
furnaces and various other places. Natural Gas being clean gas has less
environmental issues compared to all other types of fuel. Natural gas being rich in
Propane and Butane gives straight run LPG which has huge demand in countries
like India. It has now become possibly to liquify and transport NG. It is available
for uses fuel in automobiles also. Some of the NG fields in India produce sour NG
containing highly poisonous gas H2S in different-different amount. Hydro-carbon
condensate associated with sour natural gas also becomes sour and gives rise to
production of LPG rich in H2S which requires additional treatment for making it
sweet, marketable and safe for use.
Hazira Gas processing complex is receiving sour NG from South Basin Gas fields
which is a subsea reservoir. The gas is transported from South Basin Gas Field to
Hazira Gas Processing Complex through subsea pipeline. The gas received at Gas
Terminal in a slug catcher where gas and slug containing HC condensate, moisture
and chemicals (like corrosion inhibitors) are separated. Gas and associated
condensate are sent further in separate system for processing.
The sour gas processing system at Hazira Project, ONGC consists of followings:
1. Gas Receipt Terminal
2. Gas Sweetening Unit
3. Gas Dehydration Unit
4. Dew Point Depression Unit
5. Sulphur Recovery Unit
6. Sour Condensate Processing Unit
7. Gas based LPG Recovery Unit
8. Kerosene Recovery Unit
The top consumers of natural gas from Hazira are: GAIL, KRIBHCO, ESSAR,
RPL, GGCL, HWP.

6|Page
 3. FIELD OPERATIONS AND INLET
RECEIVING:

Figure.2 Overall gas processing flow chart.

7|Page
3.1 FIELD OPERATIONS
It comprises following steps:
1.Wellhead operations:
The operations done to basically control the flow of the gas.

2. Piping:
Many wells are tied to a platform, and then from those multiple platforms, the gas is tied
together into large pipelines that go to the gas plant for further processing.

3. Compression Stations:
The compression stations also called booster stations which are there to pump the gas to
make it move into the gas plant. In addition to this these stations usually have inlet suction
separators (scrubbers) to remove condensed liquids. Some stations even have dehydration
facilities to reduce corrosion and mitigate hydrate formation. The booster horsepower
ranges from 10 to 1000.

Figure.3 A Booster station

Figure.4 Schematic of common operations at a compression station on a gathering System.

8|Page
4.Pigging
Pigging is the process of forcing a solid object through a pipeline. The process involves
inserting the pig, via a pig launcher, into the pipelines and removing it by use of a pig
receiver. Pigging is used to perform any of the following functions:
1. Provide a barrier between liquid products that use the same pipeline
2. Check wall thickness and find damaged sections of lines
3. Remove debris such as dirt and wax from lines
4. Provide a known volume for calibrating flow meters
5. Coat inner pipe walls with inhibitors
6. Remove condensed hydrocarbon liquids and water in two phase pipelines

FLOW ASSURANCE:
It is the technology used that ensures hydrocarbon flow is unimpeded by line blockage from
walls to point of processing.
The flow is disturbed by the presence of below things in oil and gas industry:
1. Wax & asphaltene solid deposition.
2. Scale deposition
3. Gas dehydrates solids formation.
Wax & asphaltenes are no serious problems for Gas gathering systems which if present can
be easily solved by pigging. Scale formation problem is only encountered at Gas wellhead but
not in pipelines so we are only concerned about gas hydrate formation in this flow assurance
of natural gas transportation for onshore processing.
Gas hydrate formation is a serious problem to be concerned is because of two reasons
associated with it. First, the hydrate plugs can form even in minutes and few hydrates could
time in days, months or even in years so there is fluctuation in time required for hydrate
formation. Secondly, although hydrate formation can be inhibited in a number of ways,
injection pump failure, separator failure, and process upsets can suddenly make pipeline
contents vulnerable to hydrate formation.
Gas hydrate properties:
Gas hydrates are a class of solids, non-stoichiometric compounds which are called
“clathorates”. Clathorates are chemical substances consisting of a lattice that traps or
contains different molecules. It is a kind of host-guest complexes in which guest molecules
are trapped by lattice of host molecules. In gas hydrates water is host and gases are guest
molecules.

9|Page
HYDRATE FORMATION PREDICTION:
Using thermodynamics, we can predict the temperature and pressure at which hydrate
formation will occur on the basis of gas composition. But, sometimes even when
thermodynamically they should form, they don’t get formed because hydrate crystal growth
is a random process and thus, sometimes can not be accurately predicted using
thermodynamics.
Then came the statistical thermodynamics which gave a range of temperature in few
Fahrenheits between which hydrate formation could occur (Using standard deviation
method and making the range lie in mean+/-3σ). This method uses computer programs to
calculate the temperature pressure conditions.
Before this there were two different methods which can give approximate results by use of
simple correlations, they are:
1. K-value method: [6][7]
In this method K value was obtained using equilibrium conditions. The aim of the first step of
this study is to develop a simple-to-use correlation for predicting hydrate-forming
conditions of sweet natural gases. This simple correlation estimates hydrate formation
pressure of sweet natural gases for pressures up to 40 000 kPa and temperatures between
260 K and 298 K as well as molecular weights in the range of 16 to 29. In the next step,
novel empirical correlations are developed to predict the required MEG weight percent in
the rich solution and the flow-rate for desired depression of the gas hydrate formation
temperature. These correlations are generated for a natural gas with relative density of 0.6
at pressures of 3, 5, 7, and 9 MPa, which are applicable to wet gas temperatures of 20, 30,
40, and 50 °C. In order to extend the application of these correlations to wide ranges of
natural gas mixtures with specific gravities of up to 0.8, two generalized correction factors
are also provided. The accuracy of this simple method is compared with the simulation
results obtained by commercial software which showed excellent agreement. In all cases
the error percent was approximately 2% and 5% for predicting hydrate formation
temperature depression and MEG injection rate, respectively.
This method is based on distribution coefficients (Ki values) for components on a water-free
basis. In the finalized method, they determined that hydrates were a solid solution that
might be treated similarly to an ideal liquid solution and defined the value as the
vapor/solid equilibrium ratio of a component in LW-H-V equilibrium by the following
equation:

where yi = mol fraction of component i in the vapor phase and si = mol fraction of
component i in the solid phase. Therefore, similar to the dewpoint calculation in
vapor/liquid equilibria, the K −Value charts are used to calculate the hydrate formation
10 | P a g e
temperature or pressure of three-phase (LW-H-V) solution in a manner that satisfies the
following equation:

The K-Value method was generated before determination of the hydrate-crystal structure
and was improved by Katz and co-workers. The K-Value charts were generated form
methane, ethane, propane, butane, carbon dioxide, hydrogen sulphide, and nitrogen.
Having the K-Value of every component in the mixture at three-phase (LW-H-V)
equilibrium, the hydrate-formation pressure at a given temperature or vice versa can be
determined by using correlation. This method is limited to the hydrate-formation pressures
up to 4,000 psia for methane, ethane, and propane; up to 2,000 psia for isobutane and
hydrogen-sulphide; and up to 1,000 psia for carbon dioxide.

2. Gas Gravity Method:

In this method for natural gas of various specific gravity the hydrate formation temperature
at a given pressure or vice versa (whichever is known) can be determined by using below
chart (figure-) which has hydrate forming line for natural gas of different specific gravities.

Figure-5 - Pressure−temperature curves for estimation of hydrate-formation

conditions as a function of gas gravity.

11 | P a g e
For pressure below 1000 psi above chart can be approximated in form of below
correlations:

Where specific gravity is defined as the ratio of the mass of a given volume of a gas to that
of an equal volume of air; both volumes are measured at 14.7 psia and 60°F (1.01 bar and
15.6°C). For an ideal gas, the specific gravity is molar mass of the gas divided b wife y the
molar mass of air (28.96).
Workout Problem 1:
Estimate the hydrate-formation temperature at 325 psia (22.4 bar) for the gas with the
composition given in the table below and Compare the results from Figure-5 and Equation
3.1.

Solution:
First, we need to calculate specific gravity of gas and using formula and chart we can get
hydrate forming temperature the problem has been solved in Excel sheet provided below.
Using equation 3.1a (since pressure 325 psi < 1000 psi) the temperature is 49.11 0F.
Using chart, the temperature was found out to be 50 0F.

12 | P a g e
 Excel sheet-1-Predicted temperature for hydrate formation.

HYDRATE INHIBITION:
There are three ways to avoid hydrate formation in natural gas streams:
1. Operate outside the hydrate formation region.
2. Dehydrate the gas.
3. Add hydrate inhibitors.
In first option we need to calculate the pressure and temperature required to form hydrates
using above methods. This option appears to be simple and convenient is impractical in its
way.
The second option is the most practical and economically feasible. Because of the high gas
volumes and high pressures, offshore operations frequently use dehydration but is lesser
used in onshore facilities.
Third option, adding inhibitors is the best one to prevent hydrate formation and is widely
used in industry.

13 | P a g e
Whenever operating conditions are prone to hydrate formation and liquid water may form
in the lines, field operations control hydrate formation by use of chemical inhibitors. They
are added at the wellhead and at booster stations through positive displacement pumps so
that the addition rate can be accurately controlled.
For Dehydration it is uncertain how much wet a gas must be to form gas hydrates.
Thermodynamics says it an unsaturated gas phase with water can not form hydrates but to
counter this we have fact that nominally 6 water molecules are required to enclose the guest
gas molecule and almost 20 or more water molecules are required to cage around the gas
molecule and many cages must combine to form a hydrate lattice. Therefore an hydrate
lattice to form requires sufficient amount of water in the gas. So, if we reduce the water
content from the gas we reduce hydrate forming tendency of the gas.
Using a chemical inhibitor seems least effective because of below reasons:
 The proper inhibitor dosage must be known to avoid plugging or needless chemical
costs, but oftentimes it is determined empirically although chemical cost is much
lesser to the operating costs.
 The reliability of inhibitor injection can be a problem because of malfunctioning
injection pumps and depleted inhibitor reservoirs, especially at remote sites.
 The possible interaction between hydrate inhibitors and other additives reduces the
effectiveness of some of additives, an effect that is usually determined empirically.
Despite of these many disadvantages still billions of dollars are spent each year on hydrate
inhibitor because of this method being cheapest in nature as require minimum capital
investment.
There are three types of chemical inhibitors found:
1. Anti-agglomerates (AA)
2. Kinetic (KHI)
3. Thermodynamic
Anti-agglomerates prevent small hydrate into bulking and forming bigger hydrate plugs
which could disturb the flow.
Kinetic inhibitors actually slow down the crystal formation by interfering with the formation
of the crystal. They are required in very small quantities say, 1 wt% range in aqueous phase
and they are non-volatile too because of these reasons they have been always selected but
their opposite side is that very precise amount is needed as if the amount exceeds than the
required it could even facilitate the hydrate crystal growth. These inhibitors are limited to a
subcooling (difference between desired operating temperature and hydrate formation
temperature at constant pressure) of 28°F (15.5°C). A maximum subcooling of 20°F (11°C) is
recommended for these kinds of inhibitors. Kinetic inhibitors are being used in many
offshore operations and will become more widely applied as experience with their use
increases.

14 | P a g e
Thermodynamic inhibitors mainly methanol and ethylene glycol, are widely used. These
inhibitors basically act as antifreeze which prevents hydrate formation. Industries largely use
methanol or glycol as chemical inhibitor so I will be continuing with those only. The required
amount of inhibitor is calculated using below formula they may be required in large
quantities say more than 50% of the water phase.
For 20-25% of methanol concentration in water phase and for 60-70% of ethylene glycol
Hammerschmidt equation can be applied:

Or after rearranging we have,

where t is the hydrate-depression temperature,°F, Xi is the mass fraction of inhibitor in the

free-water phase, and MWi is the molecular weight of the inhibitor.
Another formula is given by Nielsen and Bucklin which is given below (equation 3.3)

where xw is the mole fraction of water in the aqueous phase.

This formula is recommended to use for methanol concentrations up to 50% in water phase.
Methanol v/s Glycol:
When it comes to choose between methanol and glycol we will have to see what our
objective is and what are the conditions required for preventing hydrate formation. Both
glycol and methanol have certain advantages and certain limitations which comes from their
variation in chemical and physical properties like viscosity, volatility, boiling point etc.
Both inhibitors remain in condensed phase even if condensed hydrocarbon phase is also
present. However, methanol is volatile since the gas volume greatly exceeds the water
volume methanol loss is of great concern.
Ethylene glycol, vaporization losses are negligible, and a solubility loss of 0.07 kg/m3 of NGL
is a common estimate for design purposes. These inhibitor losses to the condensate assume
the condensate is low in aromatics. If aromatic concentrations are typically high, inhibitor
losses can be significantly higher so it is assumed that condensate hydrocarbon has very
less aromatics. Ethylene glycol is relatively easy to recover, as it remains with the aqueous

15 | P a g e
phase and can be concentrated by evaporation of the water. Separation of methanol from
water requires distillation. It is required that before sending the natural gas for offshore gas
processing these methanol content should be properly recovered from natural gas. If these
gases are sold to refineries directly a penalty fee is charged to the gas exploiter or is bought
at lower rates for not recovering methanol well. So, to earn handsome money for the
natural gas the exploiter company should have good recovery facilities. Below chart
provides a means for estimation of vaporization losses and losses in the hydrocarbon
phase.

Figure-6-Ratio of methanol-vapor composition to methanol-liquid composition.

Numerical Example 2:
A sweet gas with a specific gravity of 0.73 leaves a gas−liquid separator at 100°F and 600
psia saturated with water. The gas drops to 35°F before reaching the next booster station at
500 psia. Assume no hydrocarbon condensate has formed in the line. Calculate how much
methanol must be added to prevent hydrate formation between the separator and the
booster station per MMscf. Repeat the calculation for ethylene glycol, which is added in an
80 wt% mixture with water.
Hint: Proceed the question through determining each of below things:
1. Determination of the amount of liquid water formed
16 | P a g e
2. Calculation of the required amount of inhibitor in the water phase
3. Calculation of the required amount of inhibitor in the gas phase.
Solution:
First of all, we have to check whether there will be any hydrate formation with the cooling
from the temperature 100 to 35 °F so, we will use equation 3.1 to calculate hydrate forming
temperature at 600 psi of pressure. We as calculated in below excel sheet see that at
58.96°F the hydrate formation occurs so there will be hydrate formation below 58.96°F if
no inhibitor is added.
Secondly to calculate the water content in the hydrocarbon gas which is a function of
temperature and pressure so we go through the chart**** given in dehydration topic. So,
at temperature 100 °F and 600 psi the water content is 95 lb/MMscf and at temperature 35
°F and 600 psi the water content is 13 lb/MMscf taking the worst case 95 – 13 = 82
lb/MMscf of water content lies in the hydrocarbon natural gas.
To estimate the concentration of methanol, after rearrange Equation 3.3 to give
ln xw = −t(°F)/129.6 = −24/129.6 = −0.185
xW = 0.831 (mole fraction of water)
or xMeOH = 1-xW = 0.169 mole fraction.
This value is (0.169 × 32)/(0.831 × 18 + 0.169 × 32) = 26.6 wt% total water phase or 0.362 lb
methanol/lb water. Thus, the methanol needed in the water phase is 0.362 × 82 = 29.7
lb/MMscf of natural gas.
To estimate the methanol in the vapor phase we use figure for pressure 600 psia,and
temperature 35 0F , gives 1.22 lb methanol vaporized per wt% methanol in the aqueous
phase, or 1.22 × 26.6 = 32 lb methanol per MMscf. Thus, the vapor phase consumes more
methanol than does the aqueous phase. The total amount of methanol required is (26 + 32)
= 58 lb/MMscf. If a condensate phase was present as well, the losses estimated by use of
Figure 3.10 would have to be added into that phase.
Ethylene Glycol Requirement :
Use Equation 3.2b with the constant 3,225 and a molar mass of ethylene glycol of 62 to
obtain the mass fraction of pure glycol required:
However, the glycol is diluted to 80 wt%. To obtain the mass of inhibitor solution added per
unit mass of free water present to obtain the desired concentration, we use X0 /(Xi – X0),
where X0 is the weight fraction of inhibitor in the solution to be added. The amount of
inhibitor solution added per pound of free water initially present is then 0.8/(0.8 – 0.40) =
2.00, and the total amount of ethylene glycol solution added is 2.00 × 82 = 164 lb/MMscf.
At these conditions, glycol loss into the vapor phase is negligible, so the total amount of
solution required is 164 lb/MMscf.

17 | P a g e
To know methanol content in hydrocarbon phase:
For methanol content to be present in condensate phase as well with paraffinic
hydrocarbons to know the quantity we have below chart:

FIGURE-7-Methanol content in hydrocarbon in condensate phase as a function of

temperature.

18 | P a g e
3.2 INLET RECEIVING:
Gas and liquids that enter the gas plant pass emergency shutdown valves, which isolate the
plant from incoming streams and pig receivers, and then go to inlet receiving, where
condensed phases drop out.
The very first liquid-gas separations occurs in slug-catcher.

3.2.1 Slug Catcher:

A slug catcher is gas liquid separator cum liquid reservoir separator, where separation
typically occurs between the heavy liquid hydrocarbons and the gaseous lighter ends i.e. the
gas or vapor is separated from the liquids. The slug catcher has been sized so as to allow the
separation of the maximum condensate flow from natural gas.
Slug catchers are critical because downstream gas processing units rely on a continuous gas
stream free of liquids, even when surges of liquid enter the plant problems occur. A slug
catcher is a gas−liquid separator sized to hold the biggest slug a plant will experience.
Depending upon slug catcher design, inlet receiving handles just slugs or combines slug
catching with liquid storage.

3.2.2 Separator principles:

Separators also called scrubbers, knock out pots, inlet receivers. Effective phas separators
protect downstream equipment designed to process a single phase. It is the critical first step
in most processes in gas plants and typically is a simple vessel with internal components to
enhance separation.

3.2.3 Gas-liquid separators:

Separator vessel orientation can be vertical or horizontal. Vertical separators are most
commonly used when the liquid-to-gas ratio is low or gas flow rates are low. They are
preferred offshore because they occupy less platform area. However, gas flow is upwards
and opposes the flow of liquid droplets. Therefore, vertical separators can be bigger and
more complex thus, more costly than horizontal separators. Hazira has horizontal
separators.
Vapor-liquid separator is one of the most common types of process equipment in natural gas
processing plants. A vapor-liquid separator is a vessel into which a liquid and vapor mixture is
fed and wherein the liquid is separated by momentum difference and gravity, falls to the
bottom of the vessel, and is withdrawn. The vapor travels upward at a design velocity which
minimizes the entrainment of any liquid droplets in the vapor as it exits the top of the vessel.
A vapor-liquid separator might consist simply of an empty vessel, which causes the fluid
velocities in the entering pipe to be reduced by enlarging the cross-sectional area of flow.

19 | P a g e
Usually, however the separator includes internal parts, to promote separation of the
process, such as:

1.Primary separation section (entrance): This is for separating the bulk of

the liquid from the gas. It is desirable to remove the liquid slugs and large droplets of liquid
quickly from the gas stream, and to remove gas from the liquid.

2.Secondary separation section: This is for removing smaller particles of liquid

by gravity settling depends to a large extent on the decreased gas velocity and reducing the
turbulence of gas.

3. Liquid separation section (or the liquid accumulation section):

This is for removing gas bubbles which may be occluded with the liquid, and for sufficient
storage of the liquid to handle the slugs of liquid anticipated in routine operation.

4. Mist extractor or eliminator section (mist pad): This is for removing

entrained drops of liquid from the gas, which did not separate in the secondary separation
section. Mist extractor might be used to decrease the amount of entrained liquid in the gas
and to reduce diameter of the vessel. Thickness of mist eliminator is typically 6 inches. Mist
extractor is basically a kind of clarifying filter which is made of wire mesh pads are made of
finely woven mats of stainless steel wrapped into a tightly packed cylinder

5. Vortex breaker (in the bottom of the vessel): It prevents potential

pump suction problems if a pump is used to remove collected liquids. Provide vortex
breakers on outlet nozzles that are piped to a pump and where flow continuity at minimum
liquid level is critical. Hydrocarbon liquid outlets should project 4 to 6 inches above the drum
bottom if water is likely to be present and the water will interfere with downstream
processing.

Figure-8- The common two-phase separator designing. Figure-10-The finger type slug catcher.

20 | P a g e
3.2.4 Slug Catcher at Hazira Plant:
Slug catcher is made of 48 fingers [ 24 fingers each] having diameter of 48” for each finger. It
is split into two 24 pipe sectors fully self-sufficient for 50% running in the event of maintenance
of one of them.
Slug catcher length 490 metres
Area of separation 5% slope
Area of storage 0.5% slope
Length of storage area 459 metres

With this two different sloping greater efficiency of separation is ensured. This geometrical
configuration has helped as follows:
1. To keep the height of slug catcher within the limit.
2. To reduce storage area.
3. Uniform sharing of gas in different fingers.
4. Progressive decrease in the flow speed facilitating separation of liquid from gas.
There are 3 risers at beginning, centre and end of the section.

1. PRIMARY RISER: Close to the inlet side of the slug catcher, allows separated gas to come
out during normal flow.

2. SECONDARY RISER: Positioned approximately in the centre, diverts small quantities of

gas coming with liquid. Also, it acts as escape route for gas left in this area during arrival
of condensate during pigging.

3. EQUALISING HEADER: Positioned at the end of the storage area, to prevent liquid
agitation and entrainment produced by current flow of gas and liquid towards primary
header.
If we look at the ratio of total length of slug catcher to the length used for storage we see that
only 10% of the total length is utilised for the primary separation or simple two phase
separation rest 90% of length is used for secondary or gravity separation and storage.

3.2.5 Primary Separation:

Primary separation is done by making the use of the fact that there exist difference in the
momentum of the gas and liquid. Larger liquid droplets fail to make the sharp turn and thus
impinge on the inlet wall. This action coalesces finer droplets so that they drop out quickly.
Although inlet geometries vary as shown in the figure 2 below, most separators use this
approach to knock out a major portion of the incoming liquid.

21 | P a g e
Figure 6-Different types of inlet devices.

3.2.6 Gravity Separation or secondary separation:

About 90% of the separation is done by gravity or secondary separation which happens simply
because of difference in gravity difference. The main separators are gravity type. This type
tends to use gravity as the dominant force. Gravity settling requires low gas velocities with
minimal turbulence to permit droplet fallout. Gravity separations also depend essentially on
the density differences of the gas, solid, or liquids present in the mixture. The particle (droplet)
size of the dispersed phase and the properties of the continuous phase are also factors with
the separation motivated by the acceleration of gravity. Liquid droplet dropout when
gravitational pull is greater than the drag force of the flowing fluids around the droplet. These
forces can be described mathematically using the terminal or finite-settling velocity
calculation,

Where,
VT = terminal velocity, ft/s
Dp = droplet diameter, ft
g = gravitional constant, 32,17 ft/s2
ρL = liquid density, lb/ft3
ρV = vapor density, lb/ft3
C’ = drag coefficient, dimensionless
The drag coefficient (C’) is a dimensionless quantity that is used to quantify the drag or
resistance of an object in a fluid. It is always associated with a particular surface area.
For the purpose of this equation, particle shape (droplet) is considered to be a solid,
rigid sphere. This value also depends on the Reynolds number of the flowing fluid.

22 | P a g e
Where,
Re = reynold number, dimensionless
μ = viscosity, cP
Iteration is needed to solve this problem since the terminal velocity (VT) is involved in
both equations. Thus, the simplified equation is presented to avoid trial and error
solution.

Values of the drag coefficient as a function of the product of drag coefficient (C′) times
the Reynolds number squared are presented in Figure 6.

Figure-9-Drag coefficient for rigid spheres

Depending on Reynolds numbers, terminal velocity could be determined by several
laws:
1. Stokes’ Law
Stokes’ Law is commonly applied at low Reynolds number (<2).

23 | P a g e
2. Intermediate Law
This law is applied for Reynolds number between 2 and 500.

3. Newton’s Law
Newton’s Law is applicable for a Reynolds number range of approximately 500-
200,000. Use value of 0.44 for drag coefficient (C’), thus the terminal velocity is

For the Newton’s Law region, the upper limit to the Reynolds number is 200,000 and
KCR = 18.13.

24 | P a g e
SEPARATOR SIZING CALCULATIONS
An EXCEL sheet was prepared which has calculations required in designing of two
phase horizontal separator both with mesh pads and without mesh pads.
All the solution has been done in the excel sheet itself using the industry operating manual
and input data by using above formulae and principles. The excel sheet with name separator
sizing is the required excel sheet which has been pasted below too.

Screenshot -Excel sheet no. 4 separator sizing.

25 | P a g e
 4. GAS SWEETENING/TREATING UNIT
4.1 INTRODUCTION:
Gas treating involves reduction of the “acid gases” carbon dioxide (CO2) and hydrogen
sulphide (H2S), along with other sulphur species, to sufficiently low levels to meet contractual
specifications or permit additional processing in the plant without corrosion and plugging
problems.
Before going into the concept of GSU following questions comes in mind for sure which needs
to be answered:
1. Why are the acid gases a problem?
Answer: Hydrogen sulphide is highly toxic, and in the presence of water it forms a weak,
corrosive acid. The threshold limit value (TLV) for prolonged exposure is 10 ppmv
and at concentrations greater than 1,000 ppmv, death occurs in minutes. When H2S
concentrations are well above the ppmv level, other sulphur species can be present. These
compounds include carbon disulphide (CS2), mercaptans (RSH), and sulphides (RSR), in
addition to elemental sulphur. If CO2 is present as well, the gas may contain trace amounts of
carbonyl sulphide (COS). The major source of COS typically is formation during regeneration
of molecular-sieve beds. Carbon dioxide is non-flammable and, consequently, large quantities
are undesirable in a fuel. Like H2S, it forms a weak, corrosive acid in the presence of water.
Thus, the acid gases present in the natural gas are required to be removed.
2. What are the acid gas concentrations in natural gas?
Answer: A sub-quality gas is that contains CO2 ≥ 2%, N2 ≥ 4%, or H2S ≥ 4 ppmv.
3. How much purification is needed?
Answer: The inlet conditions at a gas processing plant are generally temperatures near
ambient and pressures in the range of 300 to 1,000 psi (20 to 70 bar), so the partial pressures
of the entering acid gases can be quite high. If the gas is to be purified to a level suitable for
transportation in a pipeline and used as a residential or industrial fuel, then the H2S
concentration must be reduced to 0.25 gr/100 scf (6 mg/m3) and the CO2 concentration must
be reduced to a maximum of 3 to 4 mol%.
4. What is done with the acid gases after separation from the natural gas?
Answer: For CO2, if the quantities are large, it is sometimes used as an injection
fluid in EOR (enhanced oil recovery) projects. Several gas plants exist to support CO2 flooding
projects; the natural gas and NGL are valuable by-products. If this option is unavailable, then
the gas can be vented, provided it satisfies environmental regulations for impurities. Moritis
(2001) gives a general discussion of the state of EOR in the United States. Although the United

26 | P a g e
States has not ratified the Kyoto Treaty, which limits CO2 emissions to the atmosphere, caps
may be placed on these emissions because of the preponderance of scientific data that shows
CO2 is a significant contributor to global warming.
In the case of H2S, four disposal options are available:
1. Incineration and venting, if environmental regulations regarding
sulphur dioxide emissions can be satisfied.
2. Reaction with H2S scavengers, such as iron sponge.
3. Conversion to elemental sulphur by use of the Claus or similar process.
4. Disposal by injection into a suitable underground formation.
5. What processes are available for acid gas removal?
Answer: Four scenarios are possible for acid gas removal from natural gas:
1. CO2 removal from a gas that contains no H2S
2. H2S removal from a gas that contains no CO2
3. Simultaneous removal of both CO2 and H2S
4. Selective removal of H2S from a gas that contains both CO2 and H2S
Because the concentrations of CO2 and H2S in the raw gas to be processed and the allowable
acid gas levels in the final product vary substantially, no single process is markedly superior in
all circumstances, and, consequently, many processes are presently in use.
The chart in the next page suggests what process to select for various percentage of CO2 and
H2S present in the sour natural gas and to make it meet product specifications.

27 | P a g e
28 | P a g e
4.2 HAZIRA GAS SWEETENING UNIT:
This is India’s first GSU based on the state of the art MDEA process for selective removal of
H2S from sour gas. Gas is treated with aqueous solution of MDEA in 7 identical and
independent trains each of 5 MMNM3/D capacity. Methyl diethanol amine(MDEA). A tertiary
amine, reacts with H2S selectively.
Amines are bases, and the important reaction in gas processing is the ability of the amine to
form salts with the weak acids formed by H2S and CO2 in an aqueous solution. When a gas
stream that contains the H2S, CO2, or both, is contacted by a primary or secondary amine
solution, the acid gases react to form a soluble acid−base complex, a salt, in the treating
solution. The reaction between the amine and both H2S and CO2 is highly exothermic.
Regardless of the structure of the amine, H2S reacts rapidly with the primary, secondary, or
tertiary amine via a direct proton transfer reaction, as shown in Equation 5.1, to form the
amine hydrosulphide:

The reaction is shown for a tertiary amine but applies to primary and secondary amines as
well. The reaction between the amine and the CO2 is more complex because CO2 reacts via
two different mechanisms.
First reaction mechanism:
When dissolved in water, CO2 hydrolyzes to form carbonic acid, which, in turn, slowly
dissociates to bicarbonate. The bicarbonate then undertakes an acid−base reaction with the
amine to yield the overall reaction shown by Equation 5.2:

Second CO2 reaction mechanism:

A second CO2 reaction mechanism, shown by Equation 5.3, requires the presence of a labile
(reactive) hydrogen in the molecular structure of the amine.

29 | P a g e
Sour gas from slug catcher enters the GSU and is heated in a steam heater to prevent hydrate
formation downstream of pressure let down valve, as a result of drop in pressure 64.5 Kg/cm2.
Sour gas is then mixed with the sour condensate gas from the Condensate Fractionation Unit
and passes through a knock out drum and flows to the Absorber at a pressure of 62.5Kg/cm2.
ABSORPTION SECTION:
Raw natural gas from slug catcher is mixed with a sour condensate from the condensate
fractionation unit. This sour condensate off gas enters the MDEA unit B.L. at 65.0Kg/cm2
pressure and a temperature of 450 C.
The raw gas mixture passes via a flow control valve to the knock out drum, where liquid carry
overs (water, hydrocarbons) are requested. Upon leaving separator, raw gas then enters the
amine absorber where it comes into counter contact with the aqueous solution of MDEA. The
absorber column contains 14 valve trays. The lean MDEA solution enters the column upper
section at 450 C. Different liquid feed nozzles on the column (at 1,3,5,7,9) give the flexibility of
allowing the correct number of trays to be selected to obtain the required results. The
required inlet point will depend upon the feed gas quality and operating conditions.
The treated gas leaves the top of the absorber containing less than 4 ppmv H2S at a
temperature of 400 C by heat exchanger with cooling water in the cooler before entering the
treated gas knock out drum. In this drum the liquid phase resulting because of cooling
(condensed water, amine carry over) is separated from the Rich Amine Flash Drum. The
treated gas leaves from the top of 31V302 (refer below figure) and is piped to the Dehydration
Units.

Figure-10-Absortion based Gas Sweetening Unit

30 | P a g e
Figure-11- Gas Sweeting Unit, Hazira Gas processing Plant.

31 | P a g e
 5. GAS DEHYDRATION UNIT
5.1 INTRODUCTION
The very idea of storing natural gas is to straighten demand and become less dependent on
supply. Natural Gas is stored in summer periods when the needs are lower and withdrawn in
the winter periods. For large volumes of gas to store the underground gas storages (UGS) are
the most advantageous. Types of UGS are:
1) Aquifer
2) Depleted oil/gas field
3) Salt cavern reservoir
Every type of UGS has different physical characteristics. Allowable gas pressure within the
UGS is usually around 20 MPa. The pressure within increases as the gas is being injected and
decreases during withdrawing. Output pressure depends on further pipeline distribution,
usually should exceed 7 MPa. Temperature of gas ranges from 20–35°C. Disadvantage of UGS
is that during the storage the gas is saturated by water vapours. In case of depleted oil field
UGS additionally vapours of higher hydrocarbons are contaminating stored gas. The
distribution specification set the allowable water concentration in Natural Gas by specifying
the dew point temperature (Tdew) of Natural Gas. Tdew is -7°C for Natural Gas at 4 MPa, this
value is equivalent to roughly 5 gH2O/m3 NG at 4 MPa. Water content which NG at saturation
can contain is dependent on the temperature and pressure within the UGS. Good
presentment is in the GPSA Data Book 12th Edition, Chapter 20 as Figure No. 20 [3]. The
average value of H2O in NG is five times higher than requested. Therefore, dehydration step
of Natural Gas from UGS is essential before its distribution.
Typically, dehydration has importance in below three areas:
Gas gathering: Water needs to be removed to reduce pipeline corrosion and eliminate line
blockage caused by hydrate formation. The water dew point should be below the lowest
pipeline temperature to prevent free water formation.
Product dehydration: Both gas and liquid products have specifications on water content.
Sales gas that leaves a plant is usually dry if cryogenic hydrocarbon liquid recovery is used.
Liquid and gas streams may be water saturated after amine treatment or coming from
underground storage. Most product specifications, except for propane, require that no free
water be present (Engineering Data Book, 2004a). This requirement puts the maximum water
in sales gas at 4 to 7 lb/MMscf (60 to 110 mg/Sm3).
Hydrocarbon recovery: Most plants use cryogenic processes to recover the C2+ fraction
from inlet gas. If acid gases are removed by use of amine processes, the exit gas leaves water
saturated.

32 | P a g e
Water content is stated in a number of ways:
• Mass of water per volume of gas, lb/MMscf (mg/Sm3).
• Dew point temperature, °F (°C), which is the point that liquid water, real or hypothetically
subcooled, will condense out of the hydrocarbon phase.
• Concentration, parts per million by volume (ppmv).
• Concentration, parts per million by mass (ppmw).

WATER CONTENT OF HYDROCARBONS

We assume ideal behaviour of the natural gas first then, we have a mixture where both gas and
liquid phase are in equilibrium, each component i, in the mixture obeys the below relationship

Where,
xi is the mole fraction of i, in liquid phase
ϒi is the activity coefficient,
PSat is the saturation or vapor pressure,
yi is the vapor phase concentration,
φi the vapor phase fugacity coefficient, and
P the total pressure.
Here the vapor pressure is temperature dependent, whereas the activity and fugacity
coefficients are temperature, pressure, and composition dependent. So, we have four
unknown variables temperature, pressure, gas composition and vapour composition. If
three of these four variables are known, the fourth variable can be calculated.
For pressure below 500 psi, we assume no hydrocarbon in water phase then both xi and ϒi
=1 and gas phase is as assumed earlier is ideal so we have φi=1.
Simplifying above equation 6.1 we have,

yH2O is the required quantity. This calculation is good for sweet gases.
A more accurate way to determine the water content is to use Figure-12 which is the
function of temperature and pressure at certain temperature and pressure water content
of a natural gas can be known.

33 | P a g e
Figure-12-Water content in sweet natural gas as a function of pressure and temperature.

For a natural gas of various specific gravity the below chart should be considered by which
water content obtained from above chart should be multiplied.

Figure-13- Gas gravity correction chart.

34 | P a g e
5.2 VARIOUS DEHYDRATION METHODS:
In this report I have gone through the main 3 types of dehydration methods, we will go
through each of the methods one by one and then we will see which method is the most
optimum. The 3 different methods are:
1. Absorption method
2. Adsorption method
3. Joule-Thompson expansion method

5.2.1 ABSORPTION METHOD:

In this method water gets absorbed by Tri-Ethyl Glycol (TEG). Absorption is done in glycol
contactor, that contactor be a tray column or a packet bed, by counter current flow of wet
gas (20-35°C) and TEG. TEG as gets enriched by H2O comes down and flows out in the bottom
of contactor, then runs through flash and heat exchanger and then into reboiler. In the
reboiler the water gets vaporised and is stripped out off the system. The temperature should
be of course more than 1000 C (to make water vaporise) but less than 208°C as above that
TEG gets decomposed. Lean (regenerated) TEG is then recycled back through heat exchanger
(it loses temperature and cools) and additional cooling unit back into the system from the
top of contactor column.

35 | P a g e
 Figure-14- Absorption Dehydration Scheme and the typical process conditions
required for dehydration.

36 | P a g e
NUMERICAL EXAMPLE 3:
30 MMscfd of a 0.65 gravity natural gas enters a TEG contactor at 600 psia and 100 0F. The
outlet water content specification is 7lb H2O/MMscf and the TEG circulation rate is 28 lb
TEG/lb H2O absorbed (3 gal TEG/lb H2O). How much water is to be absorbed?
SOLUTION:
1. Water content in inlet condition
For 600 psi and 100 0F, from figure 10, we have water content = 70lb/MMscf.
2. How much water is removed
Given specific gravity of natural gas to be 0.65 so from gas gravity chart we have Cg =
0.98.
So, net water content = 0.98*70 = 68.6 lb/MMscf.
Water content removed = inlet – outlet = 68.6-7 = 61.6 lb/MMscf.

Screenshot-4-Excel sheet showing calculations involved in numerical example-3. Refer

NUMERICAL-3.

37 | P a g e
ADSORPTION METHOD:

ADSORPTION PROCESS PRINCIPLES:

The adsorption process used in the dehydration process is a physical adsorption process.
Physical adsorption is an equilibrium process like vapor−liquid equilibria and equations
analogous to Equation 6.1 (given above) apply. Thus, for a given vapor-phase concentration
(partial pressure) and temperature, an equilibrium concentration exists on the adsorbent
surface that is the maximum concentration of the condensed component (adsorbate) on the
surface.
Because adsorbate concentrations are usually low, generally only a few layers of molecules
will build up on the surface. Thus, adsorption processes use solids with extremely high
surface-to-volume ratios. Commercially used synthetic zeolites (i.e., molecular sieves) have
surface-to-volume ratios in the range of 750 cm2/cm3, with most of the surface for
adsorption inside of the adsorbent. In the case of molecular sieves, the adsorbent consists of
extremely fine zeolite particles held together by a binder. Therefore, adsorbing species travel
through the macro-pores of the binder into the micropores of the zeolite. Adsorbents such
as silica gel and alumina are formed in larger particles and require no binder. Pore openings
that lead to the inside of commercial adsorbents are of molecular size; they normally range
from approximately 4 Å (1 Å = 10−10 m) to 100 Å. Molecular sieves have an extremely narrow
pore distribution, whereas silica gel and alumina have wide distributions. However, a
molecular sieve binder, which is usually about 20% of the weight of the total adsorbent, has
large pores capable of adsorbing heavier components. Two steps are involved in adsorbing a
trace gas component. The first step is to have the component contact the surface and the
second step is to have it travel through the pathways inside the adsorbent. Because this
process is a two-step process and the second step is relatively slow, solid adsorbents take
longer to come to equilibrium with the gas phase than in absorption processes. In addition
to concentration (i.e., partial pressure for gases), two properties of the adsorbate dictate its
concentration on the absorbent surface: polarity and size. Unless the adsorbent is nonpolar,
which is not the case for those used in gas plants, polar molecules, like water, will be more
strongly adsorbed than weakly polar or nonpolar compounds. Thus, methane is displaced by
the weakly polar acid gases that are displaced by the strongly polar water. How size affects
38 | P a g e
adsorption depends upon the pore size of the adsorbent. An adsorbate too large to fit into
the pores adsorbs only on the outer surface of adsorbent, which is a trivial amount of surface
area compared with the pore area. If the pores are sufficiently large to hold different
adsorbates, the less volatile, which usually correlates with size, adsorbates will displace the
more volatile ones. Therefore, ethane is displaced by propane. In commercial practice,
adsorption is carried out in a vertical, fixed bed of adsorbent, with the feed gas flowing down
through the bed. As noted above, the process is not instantaneous, which leads to the
formation of a mass transfer zone (MTZ) in the bed. Figure 12 shows the three zones in an
adsorbent bed:
1. The equilibrium zone, where the adsorbate on the adsorbent is in equilibrium with the
adsorbate in the inlet gas phase and no additional adsorption occurs
2. The mass transfer zone (MTZ), the volume where mass transfer and adsorption take place
3. The active zone, where no adsorption has yet taken place.
In the mass transfer zone (MTZ), the concentration drops from the inlet value, yin, to the
outlet value, yout, in a smooth S-shaped curve. If the mass transfer rate were infinite, the MTZ
would have zero thickness. The MTZ is usually assumed to form quickly in the adsorption bed
and to have a constant length as it moves through the bed unless, particle size or shape is
changed. The value of yin is dictated by upstream processes; the yout value is determined by
the regeneration gas adsorbate content. The length of the MTZ is usually 0.5 to 6 ft (0.2 to
1.8 m), and the gas is in the zone for 0.5 to 2 seconds. To maximize bed capacity, the MTZ
needs to be as small as possible because the zone nominally holds only 50% of the adsorbate
held by a comparable length of adsorbent at equilibrium. Both tall, slender beds, which
reduce the percentage of the bed in the MTZ, and smaller particles make more of the bed
effective. However, smaller particle size, deeper beds, and increased gas velocity will increase
pressure drop. For a point in the MTZ, the gas phase adsorbate content increases in time
from yin to yout. In principle, beds can be run until the first sign of breakthrough. This practice
maximizes cycle time (which extends bed life because temperature cycling is a major source
of bed degeneration) and minimizes regeneration costs. However, most plants operate on a
set time cycle to ensure no adsorbate breakthrough. Trent (2004) presents data that show a
change in the L/D from 0.8 to 2.7 in the bed increases the useful adsorption capacity from
8.7 to 10.0 wt% in useful water capacity for an equal amount of gas dried. However, the
pressure- drop increases from 0.4 to 4.3 psi (0.020 to 0.20 kPa). When used as a purification
process, adsorption has two major disadvantages:
1. It is a fixed-bed process that requires at least two or more adsorption beds for
continuous operation.
2. It has limited capacity and is usually impractical for removing large amounts of
impurity.
However, adsorption is very effective in the dehydration of natural gas because water is
much more strongly adsorbed than any of the alkanes, carbon dioxide, or hydrogen sulphide.

39 | P a g e
Generally, a higher degree of dehydration can be achieved with adsorbents than with
absorption processes.

40 | P a g e
ADSORPTION BED DESINGING:

Fig. 12- Vapor-phase concentration profile of an adsorbate in the three zones of an adsorption bed.

The combination of feed rate, pressure drop, and adsorbent crush strength dictates the
adsorption bed geometry. On minimizing MTZ thickness, the bed diameter should be kept
small. This feature also reduces the wall thickness of the high-pressure vessels and increases
the superficial velocity, which improves mass transfer in the gas phase. However, it does not
affect intraparticle mass transfer, which is the slower of the two processes. Thus, higher
velocities increase the MTZ thickness. Accurate calculation of the MTZ thickness is complex.
Trent (2004) suggests the following dimensioned equation for estimating the thickness of the
MTZ, LMTZ, in feet of 4 × 8 (1/8-inch [3 mm] diameter) mesh beads:
LMTZ (ft) = 2.5 + 0.025 VS (ft/min) ……..…………Equation 6.3
where VS is the superficial gas velocity in ft/min.
For 8 × 12-mesh (1/16-inch) beads, the length is about 70% of the values calculated by
Equation 6.3. The Engineering Data Book (2004b) suggests the following equation:
LMTZ (ft) = F [VS (ft/min)/35]0.3, …….……………..Equation (6.4)
where the factor F = 1.70 ft for a 1/8-inch (3-mm) sieve and 0.85 for 1/16-inch (1.5-mm)
material. Over the typical gas flow ranges, Equation 6.3 gives an LMTZ double that of Equation
6.4.

41 | P a g e
As noted above, higher velocities increase pressure drop through the bed. This pressure drop
has two adverse effects:
1. Higher inlet compression discharge pressures to maintain the same refrigeration
requirements and outlet pressure
2. Increased mechanical load on the adsorbent, which leads to particle breakdown and
causes further increases in pressure drop.
The adsorbent beds typically have a 6-inch (15-cm) deep layer of inert 1/2 - to 1-inch (13- to
25-mm) diameter alumina or ceramic balls (density of about 80 lb/ft3 [1,200 kg/m3]) resting
on a floating screen at the top of the bed. This layer is on top of another floating screen-
supported layer of 1/8-inch (3 mm) diameter beads. These layers help distribute the incoming
gas flow but, more importantly, help keep the bed from shifting. The screens float to account
for thermal expansion during regeneration. The bottom of the bed has a similar layering, with
the smaller beads on top of the larger ones, which are supported by a fixed screen. Some
manufacturers offer molecular sieves that are more resistant to the attrition caused by
refluxing at the top of the bed. The most common reasons for replacing a bed are loss of
adsorbent capacity and unacceptable pressure drop, which usually occur simultaneously.
Values for the loss of capacity with time vary considerably, but common values used for
molecular sieves in dehydration service are a 35% capacity loss over a 3 to 5 years period or
a 50% loss in approximately 1,600 cycles. Typically, a rapid loss occurs in the beginning and a
gradual loss thereafter. The adsorbent decays primarily because of carbon and sulphur
fouling and caking caused by instability in the clay binder. These effects occur during bed
regeneration. Increased pressure drop is usually caused by breakdown of adsorbent into finer
particles and by caking at the top of the bed because of refluxing. Attrition can occur when
the pressure is increased or decreased after or before regeneration. Monitoring the pressure
drop is important, as it provides a good diagnostic to bed health. A modified form of the
Ergun equation to compute pressure drop is given below:

with viscosity μ, in centipoises, density ρ , in lb/ft3, and superficial velocity, VS, in ft/min.
The coefficient values for typical adsorbents are given in Table below:

Table- -typical values of B and C for different adsorbent particle type.

42 | P a g e
DESIGNING STEPS IN A NUTSHELL:

43 | P a g e
44 | P a g e
45 | P a g e
5.2.2 MOLECULAR SIEVE DEHYDRATION METHOD:
Molecular sieve dehydration is done when natural gas is to be used for NGL recovery or in LNG
plants or when very low water dew point in the gas is required.

Molecular sieve adsorption method does below jobs:

1. It removes water from the natural gas streams to water content < 0.1 ppmv.
2. Dehydrates natural gas to be used for cryogenic applications.
3. Prevents hydrate formation and corrosion complications, the reasons have been
discussed above.
Molecular sieves dehydration units provide an effective means to remove water vapour from
a natural gas stream. Achieving cryogenic temperature (-500 F and less) in natural gas requires
the natural gas to be free from water vapours. High water vapour present can cause hydrate
formation which can be problematic in cryogenic process.
Basic design of the process unit consists of two or more identical molecular sieve (molsiv) unit
in a simple two unit molsiv unit design one operates in dehydration mode and the other in
regeneration mode. To switch from dehydration to regeneration mode there are automatic
switching valves, so as the dehydration unit becomes saturated with water vapour, using
sensors it automatically gets switched to regeneration mode where regeneration is done at
2080 C such that the water vapours are boiled out of the system.

FIGURE—Molecular sieve dehydration unit.

46 | P a g e
ADSORPTION PROCESS DESCRIPTION:
Molecular sieve is not only used for removing water vapours out of wet natural gas but can be
used for purification as well where gases like H2S, CO2 can be removed, not only to this even
a liquid impurity out of a liquid substance can be removed well using molecular sieves the only
difference that will come is the regeneration temperatures and the molecular size of the
molsiv.

Figure 12 – Schematic of a two bed adsorber gas dehydrating unit.

The figure 12 above shows a schematic of a typical two bed adsorber system in
which one adsorber bed, adsorber no.1 dries the gas while other bed, adsorber no.
2 goes through a regeneration cycle. At any point of time one bed is continuously
adsorbing while the other is regenerating. The wet gas to be dehydrated enters first
into an inlet separator where any entrained liquid goes out before the gas enters
the top the active bed. The flow of the gas is from top to bottom to avoid any bed-
fluidisation. Over a period of time, the solid bed surface will become saturated with
water and needs to be regenerated. The dried gas then goes through a dust filter
that will catch fine particles before the gas exists the unit. The filter should work
properly and be checked at regular intervals, especially if the gas goes on to a
cryogenic section with plate-fin heat exchangers, as dust can collect in the

47 | P a g e
exchangers and reduce the heat transfer and dramatically increase the pressure
drop.
The dried gas from the bottom of the adsorber after regeneration, a slip stream of
that dehydrated/dried gas is sent back for the purpose of regeneration of the same
bed, from where it came, and is heated up to 285 – 315 degrees C before being used
as regeneration gas. Sometimes a Sales gas is used instead of such slip stream. The
sales gas stream has the advantage of being free of heavier hydrocarbons as those
heavier hydrocarbons can cause coking. This gas is usually about 5 to 10% of gas
throughput.
Regeneration involves:
1. heating the bed,
2. removing the water, and
3. cooling.
For the first two steps, the regeneration gas is heated to about 600°F (315°C) to both
heat the bed and remove adsorbed water from the adsorbent. If COS formation is a
problem, it can be mitigated by lowering regeneration temperatures to 400 to 450°F
(200 to 230°C) or lower, provided sufficient time for regeneration is available. For
first two steps the regeneration gas flows upward through the tower (counter-
current to flow during adsorption) to ensure that the lower part of the bed is the
driest and that any contaminants trapped in the upper section of the bed stay out
of the lower section and in the bottom if contaminants are present come to the top
where they can be removed. The high temperature required makes this step energy
intensive and in addition to furnaces, other heat sources (e.g., waste heat from gas
turbines that drive compressors) are used when possible. Any residual water is left
at the top of the desiccant bed so as to not affect the dew point of the gas during
the adsorption process. The hot, wet regeneration gas (dried gas became wet
because of regeneration) then goes through a cooler and inlet separator to remove
the water before being recompressed and mixed with incoming wet feed. For the
cooling period, the regeneration gas flow through the tower can be downward or
upward depending on the use of wet or dry gas. When using dry gas, the flow will
be upward, and when using wet gas, the flow will be downward to ensure that any
water is adsorbed and collected at the top of the tower. As the adsober no.1
becomes saturates and undergoes regeneration, the wet gas is allowed to flow
through bed adsorber no.2 where similar steps are involved for saturation and then
regeneration.

48 | P a g e
MANUFACTURING OF ADVANCED MOLECULAR SIEVE:
Sieves are used for dehydrating and purifying natural gas because of their microporous
structure. These molecular sieves can sieve molecules as well as selectively absorb or
separate certain substances. Synthetic zeolites or molecular sieves are produced in the form
of a granules extrudates which have different microcrystalline structure.

The input window of the molecule lattice can vary in size from 2 to 15 Å depending on size of
the window of zeolite different size of the molecules can be adsorbed onto the surface. For
example, H2O molecules (2.8 Å) are smaller than CO2 (3.6 Å) molecules so are easily trapped
inside a 3 Å molsiv while CO2 remains only on the surface. Artificial zeolites have a secondary
pore/channel through which gases can be percolated.

Fig. 13- Molecular sieve windows.

Those pores are measured by special tools like Surface Analysers are used to measure
micropores and mesopores in the range of 0.8 to 100 nm. This method is based on zeolite
absorption of nitrogen at liquid nitrogen temperature. A mercury porosimeter is used for
measuring pores from 5 to 10,000 nm.

49 | P a g e
Zeolites as microscopic crystals in powder form before being moulded into extrudates and
granules zeolite crystals are grown in special solutions at high temperature. The solution is
prepared by loading mixture of different components in a specific ratio into a crystallizer.
Seeding (seeding is adding a small crystal to promote the crystallisation) the solution
guarantees the crystal growth in right direction. Crystal growth is checked hourly with X-ray
diffractors; the crystal growth is controlled. The suspension thus obtained is then filtered
when crystal growth has finished. Then washed using mother solution on pannevis belt filters
and is then dehydrated in drum driers which is the final stage in zeolite production.
The zeolite thus obtained should have 25% static water capacity (i.e. absorb water 25% of it’s
volume) as per theory at 100% crystallinity. Zeolites are insufficient in gas drying so there
must be a binder mixed with the zeolite granules and subjected to mix at high pressure binder
constitute 20% of total sieves thus produced. The extrudate now has lower water capacity
because of binder added which caused doesn’t absorb water. In gas industry we don’t look
at static water capacity but dynamic water capacity as how fast the water can be adsorbed
as these are subjected to rapid streams of gas so quickly it should adsorb. The zeolite with
binder gave dynamic water capacity only 13% as Zeolite channels are blocked by the binder
so the water molecules now can not gain access from all directions which can thus be
improved by using a special kind of binder that disintegrates in a specially prepared solution
allowing new crystals to activate this method gave a dynamic water capacity as 22%.

Table No. 2

50 | P a g e
5.2.3 JOULE THOMPSON EXPANSION DEHYDRATION
PROCESS:
JT dehydration process basically involves the adiabatic cooling of the gas using the principle
of Joule Thompson or Throttling effect which says if a gas at higher pressure is allowed to a
free sudden expansion the drop in its pressure and thus cooling of the gas occurs. The energy
required to overcome the attractive force will be consumed from the internal energy.
Derivation has been given below:
Consider an insulated vessel in which gas flows the vessel is divided by a porous membrane
into right and left at left side the gas is at higher pressure say P1 and at right side the gas is
at pressure P2 such that P1 > P2, since it is an insulated vessel no heat exchange inside or
outside occurs and thus is an adiabatic process thus ΔQ = 0.
Initial Analysis,
The process starts with volume V1 = V1 and volume V2 = 0. It ends with volume V1 = 0 and
volume V2 = V2.
The work done on the left side is: w1 = − p1(0 − V1).
The work done on the right side is: W2 = − p2(V2 − 0).
The total work done is then:
w = p1V1 − p2V2…………………………. (1)
Since the process is adiabatic, the total change in U is just the work,
or ∆U = U2 − U1 = p1V1 = p2V2……………….. (2)
U2 + p2V2 = U1 + p1V1 ……………………………(3)
so that: H2 = H1…………………………. (4)
and thus, the process is iso-enthalpic.
So, in a throttling process volume is abruptly increases and pressure decreases, but the net
effect is such that the pv term increases and to keep enthalpy h constant, term u or internal
energy decreases and hence there is a decrease in temperature.
The temperature change for an ideal gas undergoing throttling process is zero as for ideal
gases an iso-enthalpic process is isothermal in nature.
So, using above principle a JT plant is designed and the cooling effect thus caused
condensates the water. Created droplets are removed from the gas stream by a demister
inside the flash. Essential part of the system is injection of hydrate inhibitors (methanol or
Mono-Ethylene-Glycol – MEG). This prevents hydrate formation and thus plugging. In cases
where is insufficient pressure difference between the UGS and distribution network
available, additional external cooler is required. The fi g. 3 below shows the method.

51 | P a g e
ADVANTAGES OF JOULE THOMPSON EXPANSION DEHYDRATION PROCESS:
1. Control hydrocarbon dew point to comply with pipeline specifications
2. Extract valuable natural gas liquids (NGLs), significantly increasing revenue
3. Simple and less expensive alternative to standard refrigeration plant
4. No electricity required and contains no moving parts
5. Includes methanol injection pump, multiple injection ports and spray distributors
6. Standard and custom packaged JT plants available
7. Skid-mounted for simple installation at any remote location

5.2.3 OTHER PROCESSES OF DEHYDRATION:

5.2.3.1 DESICCANT PROCESSES

In some situations, such as remote gas wells, use of a consumable salt desiccant, such as
CaCl2, may be economically feasible. The system can reduce the water content down to 20
ppmv. Typical salt capacities are 0.3 lb CaCl2 per lb H2O.

5.2.3.2 REFERIGERATION PROCESS:

In this process methanol is first mixed with the gas and is cooled to very low temperatures.
The water−methanol mixture drops out and the methanol is recovered in a stripper column.
The several major advantages of this process are:
• Dew points in the range of −70 to –100°C can be obtained.
• No heat input other than to the methanol regenerator in the stripper is required.
• No venting of hydrocarbon-containing vapors is required. But, it does require external
refrigeration to cool the gas, and minimal methanol losses occur in the stripper.

52 | P a g e
5.2.4 COMPARISON OF DEHYDRATION
PROCESSES:
Each of presented method has its advantages and disadvantages and there are n number of
factors to be evaluated to choose the best process convenient to our feed, area, capital and
product specifications.
Absorption by TEG is nowadays the most widely used method and usually reaches the outlet
Tdew around -10°C. Furthermore, with improved reboiler design the outlet Tdew is even 2–
3 times lower. However, TEG has a problem with sulphur or higher hydrocarbons
contaminated gas. TEG in reboiler is foaming and with time is changing into ”black mud”.
Another disadvantage is BTEX emissions in reboiler vent.
Adsorption dehydration can achieve very low outlet water concentration Tdew < -50°C and
contaminated gases are not a problem. On the other hand, adsorption requires the highest
capital costs and space requirements. Expansion dehydration is the most suitable in cases
where a high-pressure difference between UGS and distribution connection is available.
However, the difference during withdrawal period decreases and when it is insufficient, an
external cooling cycle is needed. Also, a cycle for hydrate inhibitor regeneration (out of
liquids separated inside the flashes) is required. If the gas must be dried for cryogenic liquids
recovery, molecular sieve is the only long-term, proven technology available. It has the added
advantage that it can remove CO2 at the same time. If CO2 is being simultaneously removed,
because water displaces CO2, the bed must be switched before the CO2 breaks through,
which is before any water breakthrough.
COST INVOLVED DIFFERENCES:
Enhanced TEG regeneration systems may begin to compete with molecular sieve it has been
claimed TEG have obtained less than 0.1 ppmv water by use of TEG with a modified
regeneration system that uses about 70% of the energy required for molecular sieves. High
inlet water-vapor concentrations make molecular sieve dehydration expensive because of
the energy consumption in regeneration.
Two approaches are used to reduce the amount of water going to the molecular sieve bed.
First, another dehydration process, (e.g., glycol dehydration) is put in front of the molecular
sieve bed.
The second option is to have combined beds with silica gel or activated alumina in front of
the molecular sieve.
The bulk of the water is removed with the first adsorbent, and the molecular sieve removes
the remaining water. This configuration reduces the overall energy required for regeneration.
If dehydration is required only to avoid free-water formation or hydrate formation or to meet
the pipeline specification of 4 to 7 lb/MMscf (60 to 110 mg/Sm3), any of the above-

53 | P a g e
mentioned processes may be viable. Traditionally, glycol dehydration has been the process
of choice. System constraints dictate which technology is the best to use. Smith (2004)
provides an overview of natural gas dehydration technology, with an emphasis on glycol
dehydration. When considering susceptibility to inlet feed contamination, one should keep
in mind that replacing a solvent is much easier and cheaper than changing out an adsorbent
bed. However, prevention of contamination by use of properly designed inlet scrubbers and
coalescing filters, if required, is the best solution. In a conventional gas plant, where inlet
fluctuations are handled in inlet receiving, feed contamination is generally limited to possible
carryover from the sweetening unit. However, in field dehydration the possibility exists of
produced water, solids, oil, and well-treating chemicals entering the dehydrator.

COMPARISON OF OVERALL ENERGY DEMAND:

Content Credits: Michal NETUŠIL, Pavel DITL, Department of Process Engineering, Faculty of
Mechanical Engineering, Czech Technical University, Prague, Czech Republic.{ This section
has been fully copied and pasted as it is because it is data overloaded and itself concise and
complete.}
“As a referential situation is taken dehydration of 105 Nm3 /h water saturated NG at 30°C
and 10 MPa (also 15 MPa for expansion). Required Tdew is set to -10°C for NG at 4 MPa. That
means a water removal performance of 37 kgH2O/h.
First of all, the absorption method was calculated based on publication by GPSA (2004) . The
most economical designs employ circulation rates of about 25 to 40 litre TEG/kgH2O.
Therefore, circulation rate of 30 litre TEG/kgH2O is assumed. Appropriate concentration of
lean TEG is 99%. The temperature within the reboiler is 204°C. The losses of TEG are
estimated to 2 l/h. The particular heat requirements of reboiler and performances of
equipment used are listed in tab. 1 below.

The calculated performance of reboiler is in perfect conformity with practical data of

industrial application provided by ATEKO a.s. Secondly the energy demand for adsorption
method is presented. Due to lack of practical data it was used several calculation sources.
Unfortunately they did not show the same results. Calculations were based on publication

54 | P a g e
by GPSA (2004) [3], Gandhidasan (2001) [6] and Kumar (1987) [7]. For economical reasons it
is recommend to design long cycle operating. Therefore a 12 h cycle with 2 adsorbers was
assumed. As a suitable adsorbent the 4 Å molecular sieve in the form of spherical pellet
with diameter 3–5 mm is assumed. 100 g of the sieve can adsorb up to 12 g of H2O. Weight
of the adsorbent in the bed is around 3800 kg. Lifetime of an adsorbent depends on
operating conditions and usually varies around 3–5 years. Temperature of regeneration gas
is 300°C. Usually 10% of inlet gas is used as regeneration gas. The main difference between
calculations is the ”losing factor”. GPSA assumes that 40% of the heat in the regeneration
gas transfers to the bed, vessel steel, and heat loss to atmosphere, and the balance leaves
with the hot gas. E contra Gandhidasan calculates heat transfer of 55% for the same
situation. The particular heat requirements and performance of regeneration gas heater
(the sole energy consumer in adsorption method) are listed in tab. 2 below

The heat to the vessel could be saved by installment of proper internal isolation. Applying
this leads to reduction of total performance to 158 kW hence it is highly profitable. Finally
the energy demand for expansion method is presented. The basic step in this method is to
adequately calculate and employ Joule– Thompson effect (the temperature decrease of NG
as a consequence of its pressure drop). Hence the main role plays the available pressure
difference of NG between its inlet and outlet. For illustration 2 situations are compared: (1)
inlet pressure 15 MPa (2) 10 MPa. Operation design was taken from industrial application
of TEBODIN s.r.o. The 15 MPa wet NG is throttled in two stages and no external cooling of
NG is required. A heating inside each flash is recommended to easily carry away condensed
liquid. The 10 MPa wet NG needs to be firstly precooled by an air cooler and then is
throttled just in one fl ash. The rest of H2O in the gas has to be removed by an external
cooling cycle. In both cases MEG is sprayed into NG before each cooling to avoid hydrate
formation. In 15 MPa case 350 lMEG/h is added to wet NG stream and in 10 MPa case 500
lMEG/h. The hydrate inhibitor has to be regenerated (the MEG regeneration is quite similar
to TEG regeneration mentioned in absorption method) and recycled back to the process.
The particular performances of equipment used are listed in tab. 3 below.

55 | P a g e
CONCLUSIONS:
From the comparison of dehydration methods it can be concluded that if there is sufficient
pressure difference available in the UGS the most efficient method is expansion. However,
as the pressure difference decreases energy demand rises markedly as it is seen on fi g. 4

The steep rise on the left of graph is thanks to high energy demand of external cooling. This
is needed due to weak Joule–Thompson effect. External-cooling helps to reach specified
Tdew. Comparison of absorption and adsorption method speaks in favour to absorption. The
calculated energy demand for reference situation is 1:1.7 (Or 1:1.9 adsorbers without
internal insulation). This is in good conformity with comparison made by Kumar (1987). He
states that operating cost for these two methods are 1:1.53 for 1.2·105 Nm3 /h and 1:1.33
for 5.9·105 Nm3 /h in favour to absorption. Therefore, it is recommended to employ
absorption method in cases where it is possible. Although the heat requirements are nearly
similar for both methods, it is the “losing factor” (low heat transfer from regeneration gas)
which hinders the adsorption method. Adsorption dehydration method finds its use in cases
of treatment of sour gases because do not suffer on sulphur contamination. Advantages of

56 | P a g e
adsorption method are extreme low Tdew of NG obtained. The gas could be nearly totally
dehydrated. This is used as pre-treatment of NG before liquefying it.

57 | P a g e
 6. REFERENCES

1. http://behineh-sazan.ir/wp-content/uploads/2017/01/Fundamentals-
of-Natural-Gas-Processing.pdf [accessed on: 28/05/2018].
2. http://petrowiki.org/Predicting_hydrate_formation#Estimating_the
_total_amount_of_MeOH_or_MEG_to_inject_to_inhibit_hydrates
[accessed on: 06/06/2018].
3. http://inside.mines.edu/~jjechura/GasProcessing/08_Dehydration.p
df [accessed on: 06/06/2018].
4. file:///C:/Users/user/Downloads/bahadori_eng.pdf [accessed
on:7/06/2018].
5. https://oaktrust.library.tamu.edu/bitstream/handle/1969.1/4220/etd-
tamu-2005B-PETE-Ameripo.pdf?sequence=4&isAllowed=y
[accessed on:08/06/2018].
6. http://www.jouleprocessing.com/products/processing/jt-
plants[accessed on:09/06/2018].
7. http://inzynieria-aparatura-chemiczna.pl/pdf/2010/2010-
2/InzApChem_2010_2_87-88.pdf [accessed on: 10/06/2018].
8. file:///C:/Users/user/Desktop/ongc%20summer%20intern/08_Dehy
dration-glycol%20dehydration%20calculations.pdf [accessed on:
10/06/2018].
9. http://www.ijera.com/papers/Vol4_issue4/Version%201/Q0441141
19.pdf [accessed on: 10/06/2018].
10. http://www.kolmetz.com/pdf/EDG/ENGINEERING_DESIGN_G
UIDELINE__separator%20vessel_REV01.pdf [accessed
on:12/06/2018].
11. http://petrowiki.org/index.php?title=Special%3ASearch&search=m
olecular+sieve+dehydration&fulltext=Search [accessed
on:12/06/2018].
12. http://dynopt.cheme.cmu.edu/content/thesis/anshul_thesis.pdf
[accessed on:14/06/2018].
13. https://www.ajol.info/index.php/jasem/article/download/140385/13
0136 [accessed on:15/06/2018].
58 | P a g e