Desalter PDF

Modeling and Optimization
of
Crude Oil Desalting
By
Shahrokh Ilkhaani
A thesis
presented to the University of Waterloo
in fulfillment of the
thesis requirement for the degree of
Master of Applied Science
in
Chemical Engineering
Waterloo, Ontario, Canada, 2009
© Shahrokh Ilkhaani 2009

I hereby declare that I am the sole author of this thesis. This is a true copy of the
thesis, including any required final revisions, as accepted by my examiners. I
understand that my thesis may be made electronically available to the public.
ii
Abstract
When first received by a refinery, the crude oil usually contains some water, mineral
salts, and sediments. The salt appears in different forms, most often times it is
dissolved in the formation water that comes with the crude i.e. in brine form, but it
could also be present as solid crystals, water-insoluble particles of corrosion products
or scale and metal-organic compounds such as prophyrins and naphthenates. The
amount of salt in the crude can vary typically between 5 to 200 PTB depending on the
crude source, API, viscosity and other properties of the crude.
For the following reasons, it is of utmost importance to reduce the amount of salt in
the crude before processing the crude in the Crude Distillation Unit and consequently
downstream processing units of a refinery.
1. Salt causes corrosion in the equipment.

2. Salt fouls inside the equipment. The fouling problem not only negatively
impacts the heat transfer rates in the exchangers and furnace tubes but also
affects the hydraulics of the system by increasing the pressure drops and hence
requiring more pumping power to the system. Salt also plugs the fractionator
trays and causes reduced mass transfer i.e. reduced separation efficiency and
therefore need for increased re-boiler/condenser duties.
3. The salt in the crude usually has a source of metallic compounds, which could
cause poisoning of catalyst in hydrotreating and other refinery units.
Until a few years ago, salt concentrations as high as 10 PTB (1 PTB = 1 lb salt per
1000 bbl crude) was acceptable for desalted crude; However, most of the refineries
have adopted more stringent measures for salt content and recent specs only allow 1
PTB in the desalted crude. This would require many existing refineries to improve
their desalting units to achieve the tighter salt spec.
This study will focus on optimizing the salt removal efficiency of a desalting unit
which currently has an existing single-stage desalter. By adding a second stage
desalter, the required salt spec in the desalted crude will be met. Also, focus will be
on improving the heat integration of the desalting process, and optimization of the
desalting temperature to achieve the best operating conditions in the plant after
revamp.
iii
Acknowledgments
First and foremost, I would like to thank my advisors and supervisors, Prof. Ali
Elkamel, Prof. Mazda Biglari, Prof. Ting Tsui, and Prof. Ali Lohi, who have assisted
me patiently and generously to achieve another milestone in my life. They have been
exceptionally understanding and helpful through the course of preparation of this
thesis workbook. It has been an honour, an enriching experience and such superb
personal development for me to work with Dr. Elkamel, and other world-class
professors and students at the University of Waterloo.
I am also grateful to fellow researchers and potentially life-long friends and

collaborators in the Process Systems Engineering group and in the Department of
Chemical Engineering in the University of Waterloo.
Last but definitely not least, I would like to extend my most heartfelt gratitude to my
parents, my beloved mother, Noor Afagh Arabi and father, Shahpour Ilkhaani, for
sacrificing a great part of their lives through unconditional love to ensure that I will
receive the best of care, attention and education. I would also like to thank and
recognize truly my best friend, my brother Shaahin, who has played a significant role
in my achievements and personal development.
I wish I could hereby name each and everyone who has touched my life in so many
meaningful ways. I shall not forget your kind deeds and presence in my mind and my
heart. Indeed, I salute and thank you all with the utmost sincerity and appreciation.
iv
To my beloved mother, father and brother, and all my
respected teachers, past and present
v
Table of Contents
List of Figures .......................................................................................................................................viii
List of Tables...........................................................................................................................................xi
Chapter 1: Introduction to Crude Oil Desalting........................................................................................1

1.1. Introduction ...........................................................................................................................2
1.2. History of Desalting and Dehydration...................................................................................3
1.3. Global Trends in Crude Oil Quality ......................................................................................4
1.4. Sources of Wet Oil ................................................................................................................7
1.4.1 Primary Causes .................................................................................................................7
1.4.2 Secondary Causes .............................................................................................................9
1.4.3 Tertiary Causes ............................................................................................................... 10
1.5. Importance of Desalting in Refineries................................................................................. 11
1.5.1 Corrosion ........................................................................................................................ 11
1.5.2 Scale Accumulation ........................................................................................................ 11
1.5.3 Catalyst Activity ............................................................................................................. 11
1.6. Research Objectives ............................................................................................................ 12
Chapter 2: Process Design Parameters ................................................................................................... 13

2.1. Introduction and Background .............................................................................................. 14
2.2. Nature of Petroleum Emulsions........................................................................................... 15
2.2.1 Role of Emulsifying Agents ........................................................................................... 16
2.2.2 Stability of Emulsions..................................................................................................... 17
2.2.3 Emulsion Breaking or Demulsification .......................................................................... 18
2.3. Factors Affecting Desalting Performance ........................................................................... 19
2.3.1 Settling Time .................................................................................................................. 19
2.3.2 Chemical or Demulsifier Injection ................................................................................. 20
2.3.3 Heating ........................................................................................................................... 20
2.3.4 Dilution with Fresh Water .............................................................................................. 21
2.3.5 Mixing ............................................................................................................................ 21
2.3.6 Electrostatic Field ........................................................................................................... 22
2.3.7 pH ................................................................................................................................... 23
2.4. Comparison between Desalting Technologies..................................................................... 25
2.4.1 Cameron’s Bilectric Technology .................................................................................... 25
2.4.2 NATCO’s Dual Polarity Technology ............................................................................. 25
2.5. Electrical System for Desalters............................................................................................ 28
2.5.1 Cameron’s Bilectric System ........................................................................................... 28
2.5.2 NATCO’s Dual Polarity System..................................................................................... 29
2.6. Interface Level Control........................................................................................................ 30
Chapter 3: Determination of Optimum Temperature for Desalting Operation....................................... 32

3.1. Introduction ......................................................................................................................... 33
3.2. Analysis of Effect of Temperature on Desalting Process .................................................... 34
3.2.1 Density as a Function of Temperature ............................................................................ 35
3.2.2 Viscosity as a Function of Temperature ......................................................................... 36
3.2.3 Electrical Conductivity as a Function of Temperature ................................................... 37
3.3. Mathematical Modeling of Optimum Temperature............................................................. 38
3.3.1 Benefit Due to Flow Increase (BFI) ............................................................................... 40
3.3.2 Costs Due to Power Requirements (CP)......................................................................... 41
vi
3.3.3 Pumping Costs (CB) ....................................................................................................... 41
3.3.4 Preheating Costs (CC) .................................................................................................... 41
3.4. Results and Conclusions...................................................................................................... 42
Chapter 4: Process Design, Simulation, and Integration of the Desalter in the Crude Distillation Unit of
a Refinery ............................................................................................................................................... 44
4.1. Introduction to Modeling the Process in HYSYS................................................................ 45
4.2. Overview of Crude Distillation Unit (CDU) ....................................................................... 47
4.3. Overall Project Scope .......................................................................................................... 48
4.3.1 Process Design Criteria for Desalting Operation............................................................ 48
4.3.2 Feedstock ........................................................................................................................ 49
4.4. Crude Characterization........................................................................................................ 50
4.4.1 Brent crude ..................................................................................................................... 50
4.4.2 Conclusions for Brent Crude .......................................................................................... 67
4.4.3 Maya Crude .................................................................................................................... 68
4.4.4 Conclusions for Maya Crude .......................................................................................... 87
4.5. Thermodynamic Package .................................................................................................... 88
4.6. Process Description ............................................................................................................. 89
4.7. Process Flow Diagrams (PFDs)........................................................................................... 90
4.8. Heat and Material Balance (H&MB)................................................................................... 90
4.9. Equipment Design Consideration...................................................................................... 108
4.9.1 Parallel Wash Water Injection to Both Desalters.......................................................... 108
4.9.2 Counter-Current (Recycle) Injection of Wash Water ................................................... 108
4.9.3 Heat Exchange for Increased Desalter Temperature..................................................... 108
4.9.4 Heat Integration ............................................................................................................ 109
4.10. Environmental Considerations .......................................................................................... 110
4.10.1 Loss of Phenols into Brine ....................................................................................... 110
4.10.2 Loss of Oil into Brine .............................................................................................. 110
Chapter 5: Conclusions......................................................................................................................... 111
Appendix A: Process Flow Diagrams (PFDs) ...................................................................................... 114
References ............................................................................................................................................ 118
vii
List of Figures
Figure 1.3.1 - Average API Gravity of U.S. Refinery Input Crude Oil ....................................................4
Figure 1.3.2 - Price Differential between Brent and Maya Crudes...........................................................5
Figure 1.3.3 - Past and Predicted Trends for World Oil Production.........................................................5
Figure 1.4.1a - Early Life of a Field; Wells B and C Produce Dry Oil ....................................................7
Figure 1.4.1b - Aquifer Level Moving up With Time; Well B Produces Wet Crude...............................8
Figure 1.4.1c - Water Coning Phenomenon .............................................................................................8
Figure 1.4.1d - Water Encroachment/ Early Water Breakthrough ...........................................................9
Figure 1.4.1e - Water Fingering Phenomenon..........................................................................................9
Figure 1.4.2 - An Example of a Casing Failure ...................................................................................... 10
Figure 2.3.6 - Microscopic Representation of Attraction and Coalescence of Water Droplets .............. 22
Figure 2.3.7a - Effect of pH and Demulsifier Concentration on Emulsion Stability.............................. 23
Figure 2.3.7b - Effect of Brine and pH on Emulsion Stability ............................................................... 24
Figure 2.4.1 - Cameron Bilectric® Dehydrator/Desalter........................................................................ 25
Figure 2.4.2a - Temperature Requirement vs. API Gravity.................................................................... 26
Figure 2.4.2b - Throughput vs. API Gravity .......................................................................................... 26
Figure 2.5.1 - AC Electrostatic Coalescer .............................................................................................. 28
Figure 2.5.2 - Dual Polarity AC/DC Field.............................................................................................. 29
Figure 2.6a - Level Control in the Desalter Using Capacitance Probe ................................................... 30
Figure 2.6b - Level Control in the Desalter Using AGAR System......................................................... 30
Figure 3.2.1 - Maya Density vs. Temperature ........................................................................................ 35
Figure 3.2.2 - Maya Viscosity vs. Temperature ..................................................................................... 36
Figure 3.2.3 - Maya Electrical Conductivity vs. Temperature................................................................ 37
Figure 3.4a - Costs and benefit trends .................................................................................................... 42
Figure 3.4b - Profit trend vs. Temperature ............................................................................................. 43
Figure 4.2 - Block Flow Diagram for Crude Distillation Unit ............................................................... 47
Figure A1.0 – Brent Characterization – Crude Assay – TBP EP vs. Cumulative LV%......................... 50
Figure A1.1 – Brent Characterization – Crude Assay – TBP vs. Log Cumulative LV% ....................... 51
Figure A1.2 – Brent Characterization – Crude Assay – TBP vs. Log Residual LV%............................ 51
Figure A1.3 – Brent Characterization – Crude Assay – TBP EP vs. Cumulative LV%......................... 52
viii
Figure A2.0 – Brent Characterization – Crude Assay – Raw Density vs. TBP ...................................... 53
Figure A2.1 – Brent Characterization – Crude Assay – Vol Ave TBP vs. Ray Density ........................ 53
Figure A2.2 – Brent Characterization – Crude Assay – Log Vol Ave TBP vs. Raw Density ................ 54
Figure A2.3 – Brent Characterization – Crude Assay – TBP vs. Raw Density...................................... 54
Figure A2.4 – Brent Characterization – Crude Assay – Log Vol Ave TBP vs. Raw Density ................ 55
Figure A2.5 – Brent Characterization – Crude Assay – Raw Density vs. Log Vol Ave TBP ................ 55
Figure A3.0 – Brent Characterization – Crude Assay – Raw Density vs. Mid Cum LV% .................... 56
Figure A3.1 – Brent Characterization – Crude Assay – Calculated Density vs. Cum LV% .................. 56
Figure A4.0 – Brent Characterization – Crude Assay – Raw Viscosity vs. Mid Cum LV%.................. 57
Figure A4.1 – Brent Characterization – Crude Assay – Log Viscosity vs. API Density........................ 58
Figure A4.4 – Brent Characterization – Crude Assay – Log Viscosity vs. Cumulative LV% ............... 59
Figure B1.0 – Brent Characterization – Comparative Plot – TBP vs. Cum LV% .................................. 60
Figure B2.0 – Brent Characterization – Comparative Plot – Density vs. Cum LV% ............................. 61
Figure B3.0 – Brent Characterization – Comparative Plot – Log Viscosity vs. Cum LV%................... 61
Figure B3.1 – Brent Characterization – Crude Assay – Log Viscosity vs. Cum LV% .......................... 62
Figure C1.0 – Brent Characterization – Crude Assay – Calculated Kw vs. Log Cum LV%................... 64
Figure C2.0 – Brent Characterization – Crude Assay – Cetane Index vs. Log Mid Cum LV%............. 64
Figure C3.0 – Brent Characterization – Product Assays – Cloud Point vs. Mid Cum LV% .................. 65
Figure C4.0 – Brent Characterization – Product Assays – Pour Point vs. Mid Cum LV% .................... 65
Figure C5.0 – Brent Characterization – Product Assays – Freeze Point vs. Mid Cum LV% ................. 66
Figure C6.0 – Brent Characterization – Crude Assay – Sulfur Content wt% vs. Mid Cum LV% ......... 66
Figure D1.0 – Maya Characterization – CALII Cuts – TBP vs. Cumulative LV%................................ 69
Figure D1.1 – Maya Characterization – CALII Cuts – TBP vs. Log Cumulative LV% ........................ 69
Figure D1.2 – Maya Characterization – CALII Cuts – TBP vs. Log Residual LV% ............................. 70
Figure D1.5 – Maya Characterization – Comparative Plot – TBP vs. Cumulative LV%....................... 71
Figure D1.6 – Maya Characterization – Comparative Plot – TBP vs. Cumulative LV%....................... 72
ix
Figure D2.0 – Maya Characterization – CALII Cuts – Volume Ave TBP vs. Density .......................... 72
Figure D2.1 – Maya Characterization – CALII Cuts – Log Vol Ave TBP vs. Density.......................... 73
Figure D2.2 – Maya Characterization – CALII Cuts – Log Vol Ave TBP vs. Density.......................... 74
Figure D2.2a – Maya Characterization – CALII Cuts Linear Segment – Log Vol Ave TBP vs. Density
................................................................................................................................................................ 74
Figure D2.2b – Maya Characterization – CALII Cuts Curved Segment – Log Vol Ave TBP vs. Density
................................................................................................................................................................ 75
Figure D2.3 – Maya Characterization – CALII Cuts – Vol Ave TBP vs. Density................................. 75
Figure D3.0 – Maya Characterization – CALII Cuts – Density vs. Cumulative LV%........................... 76
Figure D4.0 – Maya Characterization – CALII Cuts – Kw vs. Cumulative LV% ................................. 77
Figure D5.0 – Maya Characterization – CALII Cuts – Cetane Index vs. Cumulative LV% .................. 77
Figure D6.0 – Maya Characterization – CALII Cuts – Cloud Point vs. Cumulative LV% .................... 78
Figure D7.0 – Maya Characterization – CALII Cuts – Pour Point vs. Cumulative LV% ...................... 78
Figure D8.0 – Maya Characterization – CALII Cuts – Freeze Point vs. Cumulative LV% ................... 79
Figure D9.0 – Maya Characterization – CALII Cuts – Viscosity vs. Cumulative LV% ........................ 81
Figure D9.1 – Maya Characterization – CALII Cuts – Kinematic Viscosity vs. CAL II Density.......... 81
Figure D9.1B – Maya Characterization – CALII Cuts – Kinematic Viscosity vs. CAL II Density ....... 82
Figure D9.2 – Maya Characterization – CALII Cuts – Viscosity vs. Density........................................ 82
Figure D9.3 – Maya Characterization – CALII Cuts – Viscosity vs. Cumulative LV% ........................ 83
Figure D10.0 – Maya Characterization – CALII Cuts – Sulfur Content wt% vs. Cumulative LV% ..... 83
Figure D11.0 – Maya Characterization – CALII Cuts – Flash Point vs. Cumulative LV% ................... 84
Figure D12.0 – Maya Characterization – CALII Cuts – Molecular Weight vs. Cumulative LV% ........ 84
Figure E1.0 – Maya Characterization – Product Assays – Molecular Weight vs. Cumulative LV%..... 85
Figure E2.0 – Maya Characterization – Product Assays – Log Viscosity vs. Cumulative LV% ........... 85
Figure E3.0 – Maya Characterization – Product Assays – Log Viscosity vs. Cumulative LV% ........... 86
Figure E4.0 – Maya Characterization – Product Assays – Density vs. Cumulative LV% ..................... 86
Figure E5.0 – Maya Characterization – Product Assays – Kw vs. Cumulative LV% ............................ 87
PFD1 – The Cold Preheat Train for Crude Distillation Unit ................................................................ 115
PFD2 – 1st and 2nd Stage Desalters ....................................................................................................... 116
PFD3 – The Hot Preheat Train for Crude Distillation Unit.................................................................. 117
x
List of Tables
Table 3.3.1 - December 2003 Price of Crude Products .......................................................................... 40
Table 4.3.2: Feedstock properties for Crude Distillation Unit................................................................ 49
Table 4.4.1 - Calculated Weight for Brent Crude................................................................................... 63
Table 4.4.3a - Calculated Weight for Maya Crude................................................................................. 80
Table 4.4.3b - Actual Cumulative Weight of the Whole Maya Crude ................................................... 80
Table 4.4.3b – Summary of Results for Maya Crude Actual Weight ..................................................... 80
Table 4.8.1a: H&MB for CDU – Streams 2, 3, 4, 5, 6 and 7.................................................................. 91
Table 4.8.1b: H&MB for CDU – Streams 8, 8A, 8B, 9A, 9B and 10A ................................................. 92
Table 4.8.1c: H&MB for CDU – Streams 10B, 10, 111, 112, 113 and 118 ........................................... 93
Table 4.8.1d: H&MB for CDU – Streams 119, 128, 129, 208, 209 and 220.......................................... 94
Table 4.8.1e: H&MB for CDU – Streams 232 and 240 ......................................................................... 95
Table 4.8.1f: H&MB for CDU – Streams 129, 208, 209, 220, 232 and 240 .......................................... 96
Table 4.8.1g: H&MB for CDU – Streams 11, 12, 14, 70, 70A and 70B ................................................ 97
Table 4.8.1h: H&MB for CDU – Streams 71, 72, 73, 74, 75 and 78 ..................................................... 98
Table 4.8.1i: H&MB for CDU – Streams 79, 82, 83, 84, 85 and 86 ...................................................... 99
Table 4.8.1j: H&MB for CDU – Streams 87, 88, 90, 91, 91A and 92 ................................................. 100
Table 4.8.1k: H&MB for CDU – Streams 93, 94, 98 and 99 ............................................................... 101
Table 4.8.1l: H&MB for CDU – Streams 14A, 14B, 14C, 15A, 15B and 15C .................................... 102
Table 4.8.1m: H&MB for CDU – Streams 16A, 16B, 16C, 17, 17A, 17B .......................................... 103
Table 4.8.1n: H&MB for CDU – Streams 18A, 18B, 18, 20, 77 and 123 ............................................ 104
Table 4.8.1o: H&MB for CDU – Streams 124, 132, 133, 218, 219 and 231........................................ 105
Table 4.8.1p: H&MB for CDU – Streams 236, 237 and 239 ............................................................... 106
Table 4.8.2a: System Salt Balance - Parallel Wash Water Injection (Normal Operation) ................... 107
Table 4.8.2b: System Salt Balance - Recycle Wash Water Injection (Counter-current Mode) ............ 107
xi
Chapter 1: Introduction to Crude Oil Desalting
1
1.1. Introduction
As oil production is often accompanied by significant amounts of water, it is
necessary to provide desalting and dehydration systems to separate the oil and water
before the oil can be sold. Oil desalting and dehydration process is the process of
removing the water-soluble salts from the crude oil.
In view of the expected oil shortage worldwide and the fact that most crude oil is
produced with some entrained water, the ability to describe the relationship of crude
to water percentage with all the various factors that affect the desalting process has
become increasingly important. Therefore all oil industries like petroleum technology,
production operations and oil refining will greatly benefit from such correlations, in a
direct approach for the study of water-in-oil emulsion formation in petroleum fluids as
well as for understanding the behavior of interfacial tension.
With the increasing regulations on effluent water purity and the ever-increasing cost
of producing a barrel of oil, the use of emulsion-treatment plants have become an
important aspect in crude oil processing. Treating of emulsions has always ranged
from the simple ways of gravity settlement to the highly sophisticated ways of
electrostatic desalting and dehydration systems. The development of desalting
systems has always been evaluated in terms of quantities of salt and water being
removed. When crude oil is heated in various refining processes, the water could be
driven off as steam. The salt in the water, however, wouldn’t leave with the steam and
could crystallize and either remains suspended in oil or could form scale within heat-
exchangers and other equipments. Entrained salt crystals could deactivate catalyst
beds and plug processing equipment. Therefore, desalting and dehydration facilities
are often installed in crude oil production units in order to minimize the occurrence of
water-in-oil emulsions.
Because of these potential problems, refineries usually reduce crude oil salt contents
to very low levels prior to processing. To reduce the amount of desalting required at
the refinery some oil purchasing contracts specify a maximum salt content as well as
maximum water content.
Due to the fact that processes are becoming more complex, more dependent on
catalyst, less tolerant for downtime of equipment, and more intense operating
conditions are deployed, the level of salt in the crude for refineries is a lot more
stringent than before, specs of 1 PTB or less are defined by refiners at present. To
satisfy such tight specifications producers are usually required to perform extensive
crude oil desalting.
The desalting process involves six major steps including separation by gravity
settling, chemical injection, heating, addition of fresh (less salty) water, mixing, and
electrical coalescing. These steps are further explained in Chapter 2.
2
1.2. History of Desalting and Dehydration
In the mid 1800’s, there was increasing demand on salt production industries in the
United States, based on evaporation of underground brines to recover salt. At that
time, crude oil was a contaminant that would often accompany the produced brine. It
was skimmed off and then discarded. The first analysis of crude oil at Yale University
revealed the origin and organic nature of oil and its valuable properties and
enterprising petroleum producers were intrigued by this new product, the rock oil. The
search technique for salt was slowed down and the race for oil production started.
Thus, the roles of contaminant and product have been reversed in the case of brine
and oil, which since the beginning have been associated in the underground and
offshore reservoirs. Since then, all phases of petroleum technology have kept pace
with the ever-lasting industrial thirst for more oil production and the never-ending
search for better and more efficient methods. Oil production techniques have
advanced from the very crude wooden troughs and pipes used in the early
development of the industry to the modern complex gathering systems, staged
separation, and treating plants.
In the early days water-in-oil emulsions were treated by allowing time for water to
settle out and later be drained off. Settling time and draining are accomplished in
various mechanical devices such as wash tanks. However, this mechanism was time
taking and resulted in a crude oil with a high salt content because of the inefficient
separation process. Therefore, to speed up settling time, and in order to increase the
efficiency of the process, other factors were to be found and applied.
Heating was later found to be an efficient means of reducing oil viscosity, allowing
water droplets to settle out faster. At best, however, the heating factor was also
unreliable because crude oil, in which the water remains emulsified, would not
separate with moderate temperatures or time. The demand for efficient methods of
desalting and dehydration continued. The advent of two techniques in 1910 changed
our perception of emulsion treatment. One of these techniques was the introduction of
a proper chemical that causes water droplets to fall out more easily and faster by
breaking up the emulsion film around the water droplets in oil and hence speeding up
the separation process. The other technique was introduction of a high voltage field to
water-in-oil emulsions through which the small droplets are forced to coalesce.
Coalescing would increase the separation efficiency by increasing the gravity.
Many commercial installations nowadays are employing chemically aided electrical

dehydration, which is a complex employing chemical demulsifiers, heat, dilution
water, mixing and electrostatic field to dehydrate and desalt the crude.
3
1.3. Global Trends in Crude Oil Quality
Conventional crude oil composition and properties could range broadly from heavy
and sour to light and sweet crude. Heavy (low API) and sour (high sulphur content)
crude oil is more difficult and more expensive to refine compared to light and sweet
crude. Global production of light sweet crude peaked in the year 2000, and has been
declining since. The diminishing supply of light sweet crude oil will also contribute to
its price volatility.2 As the world supplies of light sweet crude dry up, increasing
attention is being turned to the heavier sour crudes. More than half of the global oil
production is currently heavy and sour, and is expected to increase in the future.3 This
includes oil produced by OPEC member nations, Venezuela and Saudi Arabia in
particular, as well as non-OPEC members such as Russia. Figure 1.3.1 shows a plot of
the average API gravity of crude oils entering U.S. refineries.4
34
33
API Gravity
32
31
30
29
1985 1987 1990 1993 1995 1998 2001 2004 2006
Year
Figure 1.3.1 - Average API Gravity of U.S. Refinery Input Crude Oil
The higher demand for light sweet crude reduces supplies and drives up the selling
cost.5 This is illustrated in Figure 1.3.2, which charts the price differential between
Brent Crude, a light and sweet crude, and Maya Crude, a heavy and sour crude.6
4
20
18
16
14
Premium ($/Barrel)
12
10
8
6
4
2
0
1989 1991 1994 1997 1999 2002 2005 2008
Year
Figure 1.3.2 - Price Differential between Brent and Maya Crudes
This increasing reliance on cheaper, lower quality crudes underlies the impact of
increasingly stringent legislation on sulphur-content in gasoline,5 which may increase
reliance on low-sulphur crudes.2 Figure 1.3.3 displays the history of world oil
production, and the predicted trends for the future.7
50
45
40
35
Million Barrels Per Day
30 Light sw eet
Light Sour
25
Heavy Sour
20 High TAN
15
10
0
1990 1995 2000 2005 2010 2015 2020
Year
Figure 1.3.3 - Past and Predicted Trends for World Oil Production
*TAN = Total Acid Number
5
Much of the world-wide refining infrastructure is not equipped to refine the lower-
quality crudes. As it stands, the upgrading process is a multi-billion dollar, multi-year
process. However, due to the scarcity of light crude and the fact that as a well starts to
deplete the remaining crude oil in that well will be heavier in composition compared
with its early days of production, the trend of the crude oil production is towards
heavier and more difficult crudes.
6
1.4. Sources of Wet Oil
Water-contaminated oil reservoirs are subject to water influx. Water is often present
at the bottom of reservoirs and exerts pressure on the oil accumulations. As the oil is
produced and withdrawn up to the surface, the water advances into the void spaces
replacing the oil. Emulsions generally occur as a result of flowing crude oil streams
and shaking (agitation) of water along the flowing streams. However, when
discussing the main sources of wet oil production, there are three main causes
encountered in both theory and practice i.e. the so called primary, secondary and
tertiary causes.
1.4.1 Primary Causes
At some time in the production history of almost every oil well, more water is
withdrawn with oil than is acceptable to the buyer. Some wells produce water from
the beginning of production and others come much later in the life of the field. Figure
1.4.1a shows a very simplified form of three wells, A, B and C drilled at a distance
from one another, on the same reservoir.
Figure 1.4.1a - Early Life of a Field; Wells B and C Produce Dry Oil
The reservoir contains oil and water. In this case, a large quantity of water lies under
the oil and acts as the driving force from the bottom. Early in the life of the field, well
A, drilled deep near the point of oil-water contact interface or at the edge of the
reservoir, produces too much water. The other wells B and C drilled higher up on the
reservoir structure produce dry oil at the beginning.
7
Figure 1.4.1b - Aquifer Level Moving up With Time; Well B Produces Wet Crude
Figure 1.4.1b shows the same reservoir later in the life of the field. At this later phase,
well A is completely watered out. Well B produces some percentage of water
associated with oil and well C continues to produce dry oil.
Other primary causes could be one or a combination of the incidents such as water
coning, water fingering or an early water breakthrough shown in Figures 1.4.1c,
1.4.1.d and 1.4.1.e.
Figure 1.4.1c - Water Coning Phenomenon
8
Figure 1.4.1d - Water Encroachment/ Early Water Breakthrough
Figure 1.4.1e - Water Fingering Phenomenon
1.4.2 Secondary Causes
Other possible causes of oil wells producing salty water are those of sudden irregular
water intrusion such as following.
• Inter-communication between tubing and casing strings.

• A hole in the casing near water formation.
• Fracture or crack between oil and water formations.
• Casing failure due to corrosion or,
• Channeling caused by a poor cementing job.
Figure 1.4.2 shows one of those possible causes, casing failure. The casing failure
caused by either corrosion or poor cementing job at a point above the producing zone,
which allows water from an upper zone to enter the well and contaminate the oil
production. However, the above secondary causes can possibly be rectified in practice
and therefore prevent water intrusion.
9
Figure 1.4.2 - An Example of a Casing Failure
1.4.3 Tertiary Causes
There are still other causes of water intrusion that are induced as a result of later
technology in stimulating or enhancing the production of oil. Among these
technologies are steam or water injections into the oil reservoir. These injection
methods are used to help or increase the amount of oil recovered from depleted
pressure reservoirs. The injection of water or steam, of course, causes water to be
mixed and produced with oil. These causes usually come into the picture at later steps
in oil recovery. Sea water or steam injection plants are implemented mainly to boost
oil recoveries.
The aforementioned causes are the main producers of wet crude. Nevertheless, water-
in-oil emulsions reaching desalting and dehydration plants are also caused by mixing-
intensifiers like moving and agitation of formation brine with crude oil. The agitation
normally takes place when producing a well via subsurface pumps or gas lift methods.
The agitation influence is also intensified when flowing through casing perforations,
production tubing, subsurface safety valves, bottom and well head chokes, or in the
flow lines and pipeline restrictions.
10
1.5. Importance of Desalting in Refineries
The removal of formation water from wet oil streams has long been an essential part
in the crude oil processing. Amongst many reasons why desalting and dehydration
units are installed is avoiding transportation of high viscosity liquid, as well as water-
in-oil emulsions, which require more pumping energy. Nevertheless, crude oil
desalting and dehydration has become a necessity because of the salts carried to
refineries and the problems caused as a result.
In most oil refineries, salts and water are removed in day to day operation because of
three major reasons: corrosion, scale accumulation and catalyst poisoning.
1.5.1 Corrosion
The most frequent problem that salts and water cause is corrosion in pipelines,
vessels, valves and instrument parts in the processing plants. Chloride salts melt in
heaters, where the temperature could reach as high as 300°C. As a result, and in the
presence of water, HCl forms, which could cause serious corrosion problems with
equipment and instrumentation that are made of iron.
1.5.2 Scale Accumulation
Calcium sulfides come also into the picture of precipitation and development of scale
in heating tubes. Scaling or precipitation causes the following problems.
• Reducing heat transfer in heaters, causing more fuel consumption and higher cost.
• Creating Hot Spots in heating tubes, which reduces their operational expected life.
• Increasing flow rates excessively, which overloads pumping units making them
less efficient.
• Causing blockage in tubes and thus lowering their capacities and efficiencies.
1.5.3 Catalyst Activity
Salts have negative effects on catalysts, which are used in cracking plants and
hydrogen processing units for heavy oil products. As the processing temperatures are
high in these units, salt could deposit on catalysts in high concentrations and therefore
could lower catalyst activity or could cause poisoning of the catalyst and thus could
reduce the life cycle of the processing unit.
11
1.6. Research Objectives
This piece of work will focus on the development of desalting operation in an old
refinery. The current capacity of the refinery is 60,000 BPSD and the refinery is
planning to increase the capacity to 70,000 BPSD. The refinery currently uses
different crude blends from different sources. Design conditions will be based on 80
vol% Maya and 20 vol% Brent crudes. In addition to increasing the capacity and
changing the crude slate, based on the economic studies done by the refinery, it is
advantageous to further process the bottom of the barrel and turn the low value
Vacuum Tower Bottoms (VTB) product to more valuable products such as Naphtha,
Kerosene and Diesel by building a grass-root Delayed Coking Unit (DCU) in the
plant. This addition to the refinery, requires the VTB to have a low salt content as salt
can accumulate in the furnace tubes of the DCU feed heater and cause operational
problems.
Due to the above modifications in the refinery there is a need for full revamp of the
Crude Distillation Unit (CDU) as well as the desalting unit, which is an integrated
part of the CDU. Currently there is only one single desalter in the unit. The salt
concentration in the desalted crude stream should be 1 PTB. The current operation
allows up to 10 PTB salt in the crude stream. A second stage desalter is needed to
achieve this design spec on the desalted crude.
Following are the main objectives of this study and will form chapters of this thesis:
1. Investigate the effect of different variables on the desalting process.

2. Compare different industrial technologies for desalting operation.
3. Understand and develop a model to predict the optimum operating temperature
of the Maya crude.
4. Develop heat integration scheme to achieve the required temperature in the
desalter.
5. Develop HYSYS simulation for the two stage desalting process.
6. Develop Process Flow Diagrams for the desalting process.
12
Chapter 2: Process Design Parameters
13
2.1. Introduction and Background
Emulsions play a great role in our daily life. They are of great practical interest
because of their widespread occurrence in most aspects of our daily usage and
consumption. Some familiar emulsions include those found in foods (mayonnaise,
milk, etc.), cosmetics (lotions and creams), pharmaceuticals (hormone products and
soluble vitamins), and agricultural products (herbicide emulsion formulations).
However, petroleum and water emulsions are one of many problems directly
associated with the oil industry, during both field production and in the refinery
environment. Whether these emulsions are created along the process or are
unavoidable, as in the oil-field production area, or are deliberately induced, as in
refinery desalting operations, the economic necessity to eliminate emulsions or
maximize oil-water separation is always present.
14
2.2. Nature of Petroleum Emulsions
Oil production is associated with the simultaneous production of formation water
from petroleum reservoirs. In its early life, a production well produces water at rates
normally relatively low, whereas towards the end of the well’s lifetime the produced
water may be as high as 90% or more of the total liquid production. From a geological
point of view, formation water resides in crude oil principally because salt water
generally underlies the crude oil in the formation from which it is produced. As the
producing life of a field is extended, however, increasing proportions of formation
water are produced with the oil. Eventually, most producing wells, at some point in
their life spans, will produce water and oil simultaneously, either as a result of natural
formation conditions or as an effect of secondary or tertiary production methods.
Emulsification of the water and oil, by intimate mixing, may occur in the formations
themselves, or in mechanical equipment, such as chokes, pipeline network, separators,
and feed pumps.
Water intrusion normally starts at the edge of an oil field and progresses until the
production is predominantly water. Oil field waters vary widely in composition and
quantity of salt, which is usually dissolved in water, but their salinity is generally
greater than that of seawater. Generally, the concentrations of solids in oilfield waters
are much higher than in seawater. The total solid concentrations in formation waters
range from as little as 200 PPM to saturation i.e. approximately 250,000 PPM. Most
sea waters contain approximately 35,000 PPM total solids. The important point is that
the water contained in a producing formation has different composition compared
with any other brine, even those in the immediate vicinity of that formation.
Emulsions vary from one oil field to another simply because crude oil differs
according to its geological age, chemical composition, and associated impurities.
Furthermore, the produced water’s chemical and physical properties, which also are
specific to individual reservoirs, will affect emulsion characteristics. It should be
emphasized that formation waters from two different fields are never similar and they
vary widely in characteristics. Some have relative densities greater than 1.2, whereas
others are essentially non-saline. Ions presents usually include Na+, Ca²+, Mg²+, Cl-,
HCO3-, SO4²-, and sometimes Ba²+.
An emulsion can be defined as a system consisting of a mixture of two immiscible

liquids, one of which is dispersed as fine droplets in the other and is stabilized by an
emulsifying agent. The dispersed droplets are known as the internal phase. The liquid
surrounding the dispersed droplets is the external or continuous phase. The
emulsifying agent separates the dispersed droplets from the continuous phase. For an
oil field, the two basic types of emulsions encountered are water-in-oil and oil-in-
water. Oil-in-water emulsions are often termed reverse emulsions. More than 95% of
the crude oil emulsions formed in the oil field are the water-in-oil type. Ideally, there
are three components in a water-in-oil emulsion:
(1) Water being the dispersed phase.

(2) Oil being the continuous phase.
(3) Emulsifying agent to stabilize the dispersion.
15
Besides these three components, certain conditions must also be met before an
emulsion could form. Two conditions necessary to form stable emulsions are a) the
two liquids must be immiscible, and b) there must be sufficient agitation to disperse
the water as droplets in the oil. These emulsions may comprise varying proportions of
oil and water. Purchasing oil is always dependant on water content, which must be
reduced to as little as 2%, varying with specifications prevalent for the geological area
or dictated by the purchaser.
In oil field operations, two types of emulsions are now readily distinguished in
principle, depending on which kind of liquid forms the continuous phase.
(i) Oil-in-water (O/W) for oil droplets dispersed in water.

(ii) Water-in-oil (W/O) for water droplets dispersed in oil.
The emulsified water exists predominantly in the form of dispersed particles that vary
in size from large drops down to small drops of about 1 µm (0.0004 in.) in diameter.
The size distribution and stability of emulsions are usually determined by two factors
a) character of water and oil (gravity, surface tension, chemical constituents, etc.) and
b) production methods.
In field operations, oil and water are encountered as two phases. They generally form
a water-in-oil (W/O) emulsion, although as the water cut increases and secondary
recovery methods are employed, reverse or oil-in-water (O/W) emulsions are
increasing.
Further reference to emulsion in this research implies water-in-oil type emulsions,

which is the predominant type in crude oil production.
2.2.1 Role of Emulsifying Agents
Water-in-oil emulsions contain complex mixtures of organic and inorganic materials.

The compounds that are found along with water and oil are called emulsifying agents.
Those agents are surface-active materials that tend to stabilize emulsions to an even
greater degree. These include asphaltenes (Sulfur, Nitrogen, and Oxygen), resins,
phenols, organic acids, metallic salts, silt, clays, wax, and many others.
Emulsifying agents have surface-active preferences. Some have preference to oil, and
other are more attracted to water droplets. Ideally, an emulsifying agent has a head
and a tail. The head is hydrophilic, attracted to water droplets, and the tail is
Lipophilic, which attracts oil.
Some emulsifying agents may form a complex at the surface of droplets and thus
yield low interfacial tension and a strong interfacial film. Nevertheless, emulsifying
agents either tend toward insolubility in either liquid phase or have an approach
mechanism for both phases, but always found concentrated at the surface. In general,
the action of emulsifying agents can be visualized as one or more of the following:
(a) Reducing the interfacial tension of water droplets, thus causing smaller
droplets to form. Smaller droplets are difficult to coalesce into larger
droplets, which can settle quickly.
16
(b) Forming a viscous coating, physical barrier, on droplets that keeps
them from coalescing into larger droplets.
(c) Suspending water droplets. Some emulsifiers might be polar molecules
creating an electrical charge on the surface of the droplets causing like
electrical charges to repel and preventing them from colliding.
The type and amount of emulsifying agent would affect emulsion’s stability.
Temperature history of the emulsion is also an important effect on the formation of
some of the emulsifying agents, paraffin and asphaltene type. The strength of the
interface bond and the speed of migration of the emulsifying agents are important
factors.
2.2.2 Stability of Emulsions
The stability of emulsions and the contributing factors are of great importance to
production of oil from underground reservoirs. Although extensive studies have been
conducted in investigation of the destabilization of W/O emulsions, the actual
mechanisms are still not well understood.
Emulsions may be stabilized by the presence of a protective film around water

droplets. Protective films, created by emulsifying agents, act as structural barrier to
coalescence of water droplets. Nevertheless, the factors favoring emulsion’s stability
can be summarized as follows.
2.2.2.1 Type of emulsifying agent
When water and oil first mix, the emulsion may be relatively unstable. As time
goes by, emulsifying agents migrate to the interface of water-in-oil due to their
surface-active characteristics. Emulsifying agents’ activity is generally related
to two function-performance at the interface, and the speed of migration.
2.2.2.2 Droplet size
The more shearing action that is applied to an emulsion the more the water
will be divided into smaller drops, and the more stable the emulsion becomes.
2.2.2.3 Water content
As the percentage of water increases, the stability of the emulsion decreases.

Experience has shown that the lower the water percentage, the more difficult it
is to treat the emulsion. Generally, a water percentage above 60% increases
the chance of forming water as an external phase. Thus, when diluted with
fresh water, the emulsion may invert to O/W type. The amount of emulsifying
agents, which are mostly present at the water-oil interface, is concentrated if
water percentage is small.
The stability of an emulsion may also be subject to the following.
• Viscosity of the oil (high viscosity oils have high resistance to flow and thus
retarding water droplet movement to coalesce)
17
• Age of emulsion (in general, as oil and water are mixed the emulsifying agents
tend to go toward the interface).
This kind of action causes emulsions to age and become more difficult to treat, as well
as causing film strength (foreign materials present in emulsions tend to increase the
strength of the film surrounding a drop of water).
To break or rupture the film that surrounds a water drop, it is necessary to introduce
chemical action and, in many desalting plants, apply heat. The chemical used to break
the film is widely known as demulsifier, the subject of the next section.
2.2.3 Emulsion Breaking or Demulsification
The treatment of emulsions has been approached in a number of ways over the years.
Today, however, injecting chemicals (demulsifiers) is by far the most widely used in
the oil industry.
Demulsifiers are similar to emulsifying agents. Their action is always at the water-oil
interface and, therefore the faster the demulsifier gets there the best job can be done.
Demulsifiers reach the interface and then work on three steps a) flocculation b)
coalescence and c) solid wetting. Flocculation is joining together of the small water
drops, rupturing of the thin film and then uniting the water drops. As coalescence
takes place, the water drops grow large enough to settle down and be easily separated.
The solid wetting takes its course with solid emulsifying agents as iron sulfide, silt,
clay, drilling mud solids, paraffin, etc.
Generally, demulsifiers act to neutralize the effect of emulsifying agents. The cost-
effectiveness of a demulsifier program depends on proper chemical selection and
application.
18
2.3. Factors Affecting Desalting Performance
Treatment of emulsions involves allowing time for water drops to settle out and be
drained off. Settling time and draining are accomplished in wash tanks, separators,
and desalting vessels. However, settling and draining can be speeded up using one or
more of the following actions.
• Injecting chemicals (demulsifier)

• Application of heat
• Addition of diluents (fresh water)
• Application of electricity
The main objective of a desalting plant is to break the films surrounding the small
water droplets, coalescing droplets to form larger drops, and then allowing water
drops to settle out during or after coalescing.
The most important variables affecting desalting performance that have been
identified and studied include (1) settling time, (2) demulsifier injection, (3) heat, (4)
addition of fresh water, (5) effective mixing of oil and water as well as chemicals for
breaking the emulsion and (6) electricity.
2.3.1 Settling Time
The desalting process uses one or more of the above mentioned procedures so as to
increase the water weight making it faster to settle and be drained off. Thus, gravity
differential is the scientific principle that forms the basis for all emulsion treatment
plants.
Formation water could flow with crude in two forms: free and emulsified. The free
water is not intimately mixed in the crude and found in larger drops scattered
throughout the oil phase. This kind of water is easy to remove simply by gravity-oil-
water separators, surge tanks (wet tanks), and desalting vessels. On the other hand,
emulsified waters are intimately mixed and found scattered in tiny drops in the oil
phase. This kind is hard to remove by simple settling devices, so, further treatment
such as chemical injection, fresh water dilution, mixing, heating, and electricity.
The desalting process relies heavily on gravity to separate water droplets from the oil
continuous phase. However, a drag force caused by the downward movement of water
droplets through the oil always resists gravity. Adequate provision has then to be built
into the desalting and dehydration system to ensure better gravitational separation.
Gravitational residence time is based on Stokes’ equation as follows:
ν = 2πr2 (∆ρ)g / 9η (2.3.1)
Where ν is the downward velocity of the water droplet of radius r, ∆ρ is the difference
in density between the two phases, and η is the viscosity of the oil phase. This
equation implies that gravitational separation can be intensified based on:
(i) Maximizing the size of the coalesced water drops.
19
(ii) Maximizing the density difference between water drops and the oil phase.
(iii) Minimizing the viscosity of the oil phase.
Heating and addition of diluent (fresh water) can best achieve factors (ii) and (iii),
whereas applying electric field will enhance factor (i).
2.3.2 Chemical or Demulsifier Injection
Emulsions can be further treated by addition of chemical destabilizers. These surface-

active chemicals adsorb to the water-oil interface, rupturing the film surrounding
water drops and displacing the emulsifying agents back into the oil. Breaking the film
allows water drops to collide by natural force of molecular attraction. Basically for
effective chemical injection, the chemical must be able to dissolve in the surface film
surrounding the water drops and it must be made of polar molecules, attracted to
acidic or organic skins surrounding water drops, which are also of polar materials.
Emulsifying agents envelop water drops with thin films preventing them from
colliding. The films are polar molecules, and the attraction between two water drops
become much like two bar magnets being drawn to each other. A demulsifier contacts
the emulsifying agent or the film, reacts with it and causes it to weaken or break.
Time and turbulence aid diffusion of demulsifiers through the oil to the interface. The
demulsifier, having caused the natural skin or film to recede from the entire water-oil
interface, exposes a thin film susceptible to rupture by the water-to-water attraction
forces at very close distances.
Chemical/demulsifier calculations are based on the following three assumptions:
• The continuous phase is oil.

• The chemical/demulsifier acts and travels in the continuous phase.
• The chemical/demulsifier is water insoluble but oil soluble.
The lower the water percentage in an emulsion the more difficult it is to treat. Reasons
for such a rule are as follows.
• The distribution of water drops in the continuous phase depends on the water
percentage. As the water percentage increases, the closer the water drops
become to each other.
• Emulsifying agents are more concentrated at the water-oil interface if the
water percentage is small.
• Dispersed drops are difficult to coalesce compared to the ones close-by. In
addition, the rate at which water drops will coalesce is a function of the
droplet radius.
2.3.3 Heating
Heat decreases the viscosity, thickness, and cohesion of the film surrounding water
drops. Heat also reduces the continuous phase (oil) viscosity helping water drops to
move freely and faster for coalescing. Heat is applied so as to accomplish the
following functions.
20
• Dissolve the skin surrounding the water drops.
• Spread demulsifier throughout the continuous phase reacting with films.
• Create thermal current to collide water drops.
• Melt the emulsifying agents.
Controlling the temperature during operations is a very delicate job. Any excessive
heat might lead to evaporation, which would result not only in loss of oil volume, but
also reduction in price because of decrease in the API gravity. Furthermore, fuel gas is
a valuable product that should not be inefficiently wasted.
Heating depends on the amount of water in the oil, temperature rise, and flow rate.
The water percentage plays a great role in fuel consumption. It requires about half as
much energy to heat oil as it does to heat water. For that reason, it is essential to
remove as much water as it is permissible prior to heating. In general, as the water
content of the emulsion increases the temperature difference between the inlet, to a
heater, and the outlet streams decreases.
Excessive heating might also result in many operational problems. Such problems
include:
• Increase in fuel cost.

• Maintenance problems and cost.
• Scale development.
• Increase in oil volume loss and API decrease.
2.3.4 Dilution with Fresh Water
Salts in emulsion could come in solid crystalline form. So, the need for fresh water to
dissolve these crystal salts arises and so the dilution with fresh water has become a
necessity in desalting/dehydration processes. Fresh water is usually injected before
heat exchangers, so as to increase the mixing efficiency and prevent scaling inside
pipes and heating tubes.
Fresh water is injected so that water drops in emulsions can be washed out and then
be drained off, hence the term “wash water” is used. The quantity or ratio of fresh
water injected depends on the API gravity of the crude. Generally the injection rate is
3-10% of the total crude flow.
2.3.5 Mixing
As discussed earlier, high shear actions form emulsions. Similarly, when dilution
water or fresh water is added to an emulsion, one needs to mix them so as to dissolve
the salt crystalline and to aid in coalescing finely distributed droplets. Mixing takes
place in a mixing valve designed to provide a high shear force in the range of 10-25
psi differential pressure. Mixing aids in the following steps:
• Smaller drops join together more easily.

• Chemical or demulsifier mixes with the emulsion.
21
• Free injected volume of wash water is broken into emulsion sized drops for
even distribution.
2.3.6 Electrostatic Field
The applied electrical voltage gradient has a large affect on desalting efficiency.
However, this is set at the design stage, since the transformer sends a constant voltage
to the electrical grid, and the separation of the electrical grids inside the desalter
vessel is not easily changed.
Inside the desalter vessel, the water droplets in the emulsion have positively and
negatively charged ends. The electrical grid distorts the originally spherical droplets
to more elliptical shapes. Droplets will be attracted by the positive and negative
electrodes, based on their internal charges and their position in the desalter. The
positive end of one droplet will be close to the negative end of another droplet, thus
providing an electrostatic attraction.17 This is illustrated in Figure 2.3.6.
+ + +
Water/Crude Oil Emulsion Before Water/Crude Oil Emulsion Just After Crude Oil Emulsion in Desalter Vessel
Wash Water Addition Wash Water Addition and Mixing Showing Coalescence of Water Droplets
- - - - - -
-- - Water Droplets -- - Wash Water Droplets -- - -
- - -
++ + ++ + -
+ ++ in Crude Oil + ++ - ++ +-
- - - - - - + ++ - - -
-- - -- - -- -
+ + +
++
+ ++
+ + ++
+ + ++ ++
+ ++
+
Crude Oil Crude Oil Crude Oil
- - -
Figure 2.3.6 - Microscopic Representation of Attraction and Coalescence of Water Droplets
The electrostatic attraction between droplets can be represented by the following

equation17:
KE 2 d 6
F= (2.3.6a)
S4
F Electrostatic force between two adjacent droplets (N)
E Voltage gradient (V/m)
K Dielectric constant for crude oil-water system
D Diameter of water droplets
S Centre to centre distance between two adjacent droplets
As can be seen in Equation 2.3.6a(2.3.6a) if the voltage gradient is increased, the

electrostatic force between two adjacent water droplets will increase. However, there
are a number of limitations on the voltage gradient. First, transformers can only
supply a certain amount of voltage to the electrical grids. Multiple transformers could
be installed to supply voltage to the grids, but the initial capital cost of these
transformers may outweigh the economic benefit achieved by a higher separation
efficiency. Secondly, at a certain voltage, water droplets will begin to rupture,
forming smaller water droplets.17 These droplets will have a higher interfacial tension,
thus causing a more stable emulsion. This occurs at the critical voltage gradient
defined by Equation 2.3.6b.17
22
T
Ec = k (2.3.6b)
d
Ec Critical voltage gradient (V/m)
K Dielectric constant for crude oil-water system
T Surface tension
d Diameter of droplet
As can be seen in equation 2.3.6b the critical voltage gradient decreases as the droplet
diameter increases. Thus, the critical voltage gradient must be based on the expected
droplet diameter when enough water droplets have coalesced together to settle out of
the oil phase.
2.3.7 pH
Crude oil contains a number of organic acids and bases which act as emulsifiers by
modifying surface charges at the oil/water interface.22 The ionizability of these
components is controlled by the emulsion pH, which can have a large effect on the
physical structure of the emulsion and hence the emulsion stability. Fortunately, the
addition of a demulsifier can greatly broaden the range of pH over which successful
separation can be achieved.19
Figure 2.3.7a - Effect of pH and Demulsifier Concentration on Emulsion Stability
The composition of the water phase can also have a large effect on emulsion stability.
Due to ionic interactions between salts and the acids and bases at the oil-water
interface, higher concentrations of brine in the water phase reduces the optimum pH at
which separation occurs, as well as broadens the overall peak as Figure 2.3.7a
exhibits.19
23
Figure 2.3.7b - Effect of Brine and pH on Emulsion Stability
The industry standard for measuring the acid content of crude oils is the Total Acid
Number (TAN) as defined in Equation 2.3.7 below:
mg KOH
TAN = required to neutralize all free acids (2.3.7)
g Crude
Crude oils with TANs higher than 1.0 are called high TAN crudes. The total base
number (TBN) is correspondingly defined as the amount of perchloric acid required to
neutralize all of the bases in the crude.
24
2.4. Comparison between Desalting Technologies
During this study, two desalter vendors, Cameron and NATCO, were contacted to
understand their concepts for designing desalters. The two vendors provide different
technologies for desalting operation. Cameron Petreco provides Bilectric Desalter
technology whereas NATCO uses the Dual Polarity technology for their desalters.
Each technology has its strengths and special considerations. Below are some
characteristics of the two technologies.
2.4.1 Cameron’s Bilectric Technology
The Bilectric design47 uses Alternating Current to polarize the water molecules, which
promotes coalescence of the water droplets. Figure 2.4.1, shows Cameron’s Bilectric
desalter design. The Bilectric design utilizes a three-grid electrode system and
horizontal emulsion distribution for superior oil/water separation performance.
These units have proven reliable for many years in the refinery application. Since the
existing desalter uses the Bilectric desalting technology, it may be an advantage to use
the same technology for the second stage desalter.
Figure 2.4.1 - Cameron Bilectric® Dehydrator/Desalter
2.4.2 NATCO’s Dual Polarity Technology
In place of the AC current electrical system, the Dual Polarity technology48 uses a
system with both AC and DC fields. The crude oil emulsion enters the Dual Polarity
equipment and flows upward through the AC field. Free water separates immediately
and falls to the water section of the vessel. Larger water droplets coalesce due to the
AC field and separate, while smaller water droplets continue with the oil as it flows
into the DC section. These remaining water droplets are subjected to the DC
electrostatic field, which causes them to coalesce and settle in the bottom of the
vessel.
25
Using the same dependable AC power supply as a conventional electrostatic desalter,
the Dual Polarity technology splits the high voltage, with rectifiers, into positive and
negative components. Pairs of electrode plates are charged in opposition. Water
droplets entering the field are elongated and attracted to one of the plates, accepting
the charge of the electrode plate they are approaching.
The dual polarity electrostatics provide for more complete dehydration.48

Consequently, it can process at higher viscosities, which means less heat is required to
lower the viscosity of the oil at processing conditions. In Figures 2.4.2a and 2.4.2b
NATCO provides performance comparison between utilizing the AC field only as
opposed to combination of AC and DC for desalters.
Figure 2.4.2a - Temperature Requirement vs. API Gravity
Figure 2.4.2b - Throughput vs. API Gravity
As per NATCO, the Dual Polarity electrostatic desalter requires less space because
the vessel can handle much higher flow rates than conventional desalters. The AC/DC
process creates larger droplets than conventional AC units, which makes it easier for
26
these droplets to fall through the opposing emulsion flow, so more oil can be
processed in a given size vessel.
27
2.5. Electrical System for Desalters
As mentioned earlier two desalter vendors, Cameron and NATCO, have been
consulted for desalter technology in order to choose a new desalter for revamp of the
crude distillation unit. Each vendor is applying different technologies to achieve the
required desalting. The brief overview of each vendor electrical system is outlined
below.
2.5.1 Cameron’s Bilectric System
As explained earlier, the Bilectric system is based on a technology using AC field for
removal of particulates. In an AC field, the rapid reversal of the current causes the
chemical reaction to be reversed before the corrosion products can be removed from
the reaction site by diffusion. Therefore, no net corrosion is observed.
The Bilectric design utilizes a three-grid electrode system and horizontal emulsion
distribution.47 The basic configuration of this process is shown in Figure 2.5.1.
Figure 2.5.1 - AC Electrostatic Coalescer
As per Cameron, the electrical portion of Bilectric system will consist of three 100
KVA, 60 Hz, single phase power units (transformers), level indicator, switchboard
panel with three AC voltmeters/ammeters, start/stop pushbutton in explosion proof
housing, three voltage/current transmitters (4-20 mA) in explosion proof housing and
a junction box for customer interface.
Similar to most conventional electrostatic oil dehydration systems, Bilectric system

employs reactance transformers to achieve protection of the electrical power supply.
An internal reactor produces a voltage drop in series with the primary winding of the
transformer which limits the power to the transformer windings. The demand load for
Cameron’s Bilectric system is 300 KVA and expected load is 60KW for 2nd stage
desalter.
28
2.5.2 NATCO’s Dual Polarity System
NATCO’s Dual Polarity system utilizes a combination of AC and DC fields to gain

the benefits of both while avoiding the corrosion problems that are associated with
just DC field. The basic configuration of this process is shown in Figure 2.5.2.
Figure 2.5.2 - Dual Polarity AC/DC Field
By using rectifiers, Dual Polarity system splits the high voltage into positive and
negative components. Pairs of electrode plates are charged in opposition. Water
droplets entering the field are elongated and attracted to one of the plates, accepting
the charge of the electrode plate they are approaching.
As per NATCO the electrical portion of Dual Polarity system will consist of one 100
KVA, 60 Hz, single phase transformer with built-in firing board SCR and rectifiers,
circuit breaker, level switches, primary circuit voltmeters, and PC-Load Responsive
Controllers (PC-LRC). Built-in firing board SCR and PC-LRC are optional and
according to the vendor will provide tuning capabilities of power supply properties
and higher tolerance for conductive crude.
Similar to most conventional electrostatic oil dehydration systems, Dual Polarity

system employs reactance transformer to achieve protection of the electrical power
supply. An internal reactor produces a voltage drop in series with the primary winding
of the transformer, which limits the power to the transformer windings. As per the
above mentioned, the demand load for Dual Polarity system is 100 KVA and expected
load typically is around 30% of demand load.
29
2.6. Interface Level Control
The second important control function for a desalter is the interface level control. The
current trend to operate on heavy crudes can lead to heavier rag layers in desalters,
which makes it difficult to control the interface level.
Measurement of the water/oil interface position has commonly been attempted with
analog type capacitance level transmitters.46 However, the measuring probe of this
type of device could become coated with carbon, water emulsions, and other material.
This coating and buildup creates interface position errors and eventually renders the
output signal meaningless. As can be seen in Figure 2.6a the probe cannot measure oil
in a water continuous mixture, and a high water cut near the top of the tank causes
capacitance probes to read full scale.
Figure 2.6a - Level Control in the Desalter Using Capacitance Probe
Another more advanced method for controlling the level is AGAR Interface Control.
A better control system not only helps in the effective control of the equipment but
also helps prevent any oil carryover to the brine system, which goes to effluent
treatment. Figure 2.6b depicts a typical AGAR level control system.
Figure 2.6b - Level Control in the Desalter Using AGAR System
30
The AGAR Concentration Control gives a current output proportional to water
content over the full scale of 0 -100%. This tells the operators about the width of the
emulsion pad and also in which direction the rag is growing. It also enables operators
to control the level accurately, in the desired direction.
31
Chapter 3: Determination of Optimum Temperature for
Desalting Operation
32
3.1. Introduction
With the decreasing light crude resources and advancements in the delayed coking
technology the heavier crude types are becoming more important options in terms of
crude oil refining. The objective of this section is to determine the optimum
temperature of the Maya crude, which is to be used in the plant for which this study is
being done.
A detailed understanding of the properties of Maya crude is essential in order to

determine the optimum temperature required for desalting of this type of crude. The
main concern is determination of the dependence of Maya crude oil properties on
temperature. The knowledge of this dependence, in addition to providing valuable
information about Maya crude, can be used to explore the effect of temperature in the
desalting process.
33
3.2. Analysis of Effect of Temperature on Desalting Process
Based on Stokes’ Law, Equation 3.2, settling rate depends highly on temperature.
Vs = 2 gr2(dw-do) / 9 µ2 (3.2)
Where:
Vs = settling rate, m.s-1

g = gravity, m.s-2
r = radius of droplet, m
dW = density of water, kg.m-3
do = density of oil, kg.m-3
µo = dynamic viscosity of oil, kg.m-1.s-1
Liquid density and viscosity usually decrease with temperature. The effect is even
greater regarding viscosity as the dependence is exponential. This means that
increasing operation temperature will raise settling rate and therefore, improve
separation. In a given desalter, separation improvement means that a larger quantity of
oil can be desalted in the same time.
This would suggest that a higher temperature is more convenient. However, crude oil
conductivity increases with temperature and so does the power requirement of the
process. Additionally, higher temperatures imply an increase of heating costs.
Given these opposing facts, it is expected that there is an optimum temperature. In the
case of Maya feedstock it is necessary to know the dependence of density, viscosity
and conductivity on temperature in order to determine the optimum temperature.
34
3.2.1 Density as a Function of Temperature
The dependence of Maya crude density on temperature is given in Figure 3.2.1. Based
on the lab data provided, the correlation that best fits the data behavior is given below.
Figure 3.2.1 - Maya Density vs. Temperature
d0 = –0.7902 T + 1204.6 (3.2.1)
Where:
T is temperature in Kelvin.
35
3.2.2 Viscosity as a Function of Temperature
Based on the available data for the viscosity of Maya crude at few different
temperatures a curve was plotted based on the best fit for the points given. Figure
3.2.2 shows the resulting equation for dependence of viscosity on temperature for a
sample Maya crude.
Figure 3.2.2 - Maya Viscosity vs. Temperature
ν = 6.8x1011 e(–0.075T) (3.2.2)
Where:
ν is Kinematic viscosity in cSt.
36
3.2.3 Electrical Conductivity as a Function of Temperature
Based on the available data for electrical conductivity of the Maya crude at a few
different temperatures a curve was plotted based on the best fit for the points given.
Figure 3.2.3 shows the resulting equation for dependence of electrical conductivity on
temperature for a sample Maya crude.
Figure 3.2.3 - Maya Electrical Conductivity vs. Temperature
κ = 0.02 e(0.0269T) (3.2.3)
Where:
κ is electrical conductivity in µS.m-1
Results from these tests show that the properties of Maya are highly dependent on
temperature. These equations where used to estimate input data for the mathematical
model that determines optimum temperature.
37
3.3. Mathematical Modeling of Optimum Temperature
The model designed to study the effect of temperature on process economics was
developed considering a change in current desalting operating temperature. In order to
calculate changes in process economics, the model should include a way of estimating
oil inflow based on temperature. The equations presented in the previous sections
allow for calculation of the water droplets settling rate from temperature. It is
assumed that at a given or fixed operating voltage the droplets population and average
size are fixed. Hence, the amount of water separated from oil is distributed in an equal
number of equally-sized drops, at any given temperature. An increase in temperature
will only cause the drops to move faster across the water-oil interface, increasing the
desalter water outflow. From equations 3.2.1 and 3.2.2 equation 3.2 can be
transformed into a temperature-dependant equation. Hence, it is possible to know the
drop’s settling rate by fixing the temperature. For calculation purposes, the drop’s
residence time within the desalter is defined as the time it takes for a single drop to
fall a given distance from the oil phase into the water phase. This is shown in equation
3.3a.
θd = h /Vs (3.3a)
Where:
θd = Drop’s residence time, s

h = Distance covered by the drop
Also drop flow was defined as the volume of water contained in a drop, which flows
within the desalter while falling into the water phase. Mathematically, drop flow is
defined in equation 3.3b.
Fd = Vd / θd (3.3b)
Where:
3 -1
Fd = Drop flow, m s
3
Vd = Volume of water in drop, m
Because drop flow is the amount of water moved through the desalter by a single
drop, the total water flow through the desalter can be calculated by knowing the
number of drops. In order to do this, drop flow is estimated for the current operating
temperature, at which the total water flow out of the desalter is known. As mentioned
before, water size and number are considered to be constant at any give temperature,
so the following relation can be assumed.
Fw (out) / Fd = Fw (out)* / Fd* (3.3c)
Where:
3 -1
Fw (out) = Water outflow, m year
38
3 -1
Fd* = Drop flow at current operating temperature, m s
3 -1
Fw (out)* = Water outflow at current operating temperature, m year
Finally, equations 3.3a and 3.3b can be substituted in equation 3.3c to obtain the
following linear relation between settling rate and water outflow.
Fw (out) = [Vs / Vs*] . Fw (out)* (3.3d)
It is to be noted that knowledge regarding the size and number of drops, as well as the
distance covered by them while settling is not required to estimate water outflow at a
given temperature. The water outflow can then be readily related to oil inflow by
considering the desalter dehydration efficiency and the water/oil feed ratio, as shown
by the following equations 3.3e and 3.3f.
Fw (in) = Fw (out) / ε (3.3e)
Where:
3 -1
Fw (in) = Water inflow, m year
ε = dehydration efficiency
And
Fo (in) = Fw (in) / Rwo (3.3f)
Where:
3 -1
Fo (in) = Oil inflow, m year
Rwo = Water/oil feed ratio
Once the oil inflow has been established for a certain temperature, the changes in
costs and benefits can be computed. The main elements considered in the model are
given in the following sections.
39
3.3.1 Benefit Due to Flow Increase (BFI)
As increase in temperature increases the settling rate of water, a larger amount of

crude oil can be treated and produced by increasing the desalter temperature. To this
end BFI is defined for economic evaluation of desalting.
BFI = [Fo (in) – Fo (in)*] . [∑in xiPi - PIM] (3.3.1)
Where:
-1
BFI = Benefit due to flow increase, USD year
Fo (in)* = Oil reference inflow, m3year-1
xi = Fraction of oil that corresponds to product i
-3
Pi = Market price of product i, USD m
-3
PIM = Price of crude oil in international market, USD m
n = Number of distillation fractions considered in the evaluation
The information used to compute BFI is presented in Table 3.3.1.
Table 3.3.1 - December 2003 Price of Crude Products

Product Vol. Fraction (xi) Price (Pi), USDm-3
Gasoline 0.156 485.45
Kerosene 0.020 236.28
Gas Oil 0.161 154.54
Lubricants 0.130 1136.36
Residue 0.505 64.82
Losses 0.028 0.00
*Note: 142.15 USD m-3 was used for the international market price (PIM) for Maya
crude in 2003.
40
3.3.2 Costs Due to Power Requirements (CP)
An increase in crude oil conductivity implies that more electric current is used,
maintaining voltage constant. This means that, while coalescence does not increase,
the power consumption does. CP was estimated as follows:
CP = (P - P*).t.Ckwh (3.3.2)
Where:
-1
CP = Power Costs, USD year
P = Power at operating temperature, kW
P* = Power at operating reference temperature, kW
-1
t = Desalter operating time, hours year
-1
Ckwh = Cost of Power, USD kWh
3.3.3 Pumping Costs (CB)
A larger flow requires additional pumping, both for oil and for water. This cost is
estimated according to the following expression.
CB = {[Fo (in) – Fo (in)*] + [Fw (in) – Fw (in)*]} .Cp (3.3.3)
Where:
-1
CB = Pumping costs USD year
3 -1
Fw (in)* = Water reference inflow, m year
-3
Cp = Unit pumping cost, USD m
3.3.4 Preheating Costs (CC)
Increasing temperature generates extra cost due to preheating either oil or water.
These costs are calculated as follows:
CC = Q.Cj (3.3.4)
Where:
CC = Preheating costs, USD year-1

Q = Quantity of heat required, J year-1
Cj = Unitary cost of heating energy, USD J-1
41
3.4. Results and Conclusions
The functions described above can be combined into a single Profit Function, which
was used to determine the optimum temperature.
P = BFI – (CP + CB + CC) (3.4)
The results obtained from the mathematical model show that there is a temperature
where the difference between total costs and total income is maximum and hence the
profit is maximized. This is shown graphically in Figure 3.4a and the maximum
difference is observed at 408.15 K (135°C or 275oF), which is the optimum
temperature for desalting operation of a typical Maya crude.43
Since Maya crude forms the major part of the blend for the refinery and there may be
periods that the refinery under study would use Maya crude only, the optimum
temperature of the desalter is determined based on the optimum temperature for
desalting of Maya only and not that of blends.
Profit Trend vs. Temperature
1.2
0.8 408.15
USD/barrel
0.6
0.4
BFI
0.2
Total Costs
0
360 380 400 420 440 460 480 500 520
Temperature (K)
Figure 3.4a - Costs and benefit trends
Figure 3.4b shows the profit curve vs. the temperature, which is another
representation for the maximum profit point.43 Any other operating temperature in the
desalter would not produce the most economic results.
42
Profit vs. Temperature
0.6
408.15
0.5
Profit (USD/bbl)
0.4
0.3
0.2
0.1
0
360 380 400 420 440 460 480 500 520
Temperature (K)
Figure 3.4b - Profit trend vs. Temperature
In view of the achieved result and the fact that current operation temperature is lower
than the optimum temperature it is advised to increase the temperature of the desalter
to 135ºC or 275 ºF, which is the optimum temperature. This modification will result in
maximum profit from the operation. Such a change can be achieved in several
different ways. In order to achieve the optimum temperature of 275ºF, a detailed study
needs to be done with regards to heat sources available and the limitations thereof. To
this end a full fletch simulation of the CDU has been prepared to study the unit
operation. The simulation and heat integration options are presented in Chapter 4.
43
Chapter 4: Process Design, Simulation, and Integration of
the Desalter in the Crude Distillation Unit of a Refinery
44
4.1. Introduction to Modeling the Process in HYSYS
Simulations are needed for generation of Heat and Material Balances (H&MB),
design of equipment and in order to predict and plan the operations. So, it is needless
to say how important simulations are and what the consequences could be if the
simulation results are incorrect. The errors in simulations could come from many
different sources including wrong initial assumptions, wrong or insufficient data
input, use of inappropriate thermodynamic package, inconsistencies in the model,
non-convergence of numerical solution, and other reasons.
A mathematical model or a simulation can be only used for a certain range of

operating conditions and may not cover all operating conditions in a processing unit
or plant, as there may be so many conflicting constraints and variables. This is also
the situation with the crude distillation unit and hence the crude desalting operation.
The heat and material balance for the crude distillation unit under study in this thesis
is developed in HYSYS. Two thermodynamic packages have been used to simulate
the crude distillation unit, BK-10 package to model the vacuum section and Peng-
Robinson package to simulate the rest of the unit; however, none of these two
packages and no other thermodynamic package built-in HYSYS has the capability to
predict the salt balance for the desalting operation. Therefore, assumption has been
made that the water in the crude as well as brine leaving the desalter is pure water and
heat and material balances for the desalting unit are based on pure water. Salt balance
has been done in Excel based on initial salt content reported in the crude, the desalter
vessel efficiency and specification required for the desalted crude.
As the desalting process has been simulated in HYSYS, from here on, modeling or
simulation refers to HYSYS simulation. In order to model the process, the very first
step is to define the composition of the feed to the unit. To this end there is a very
comprehensive built-in databank in HYSYS, from which the chemical components
can be picked to build up the feed components. Usually, for natural gas and very light
hydrocarbon feeds, it is easy to select the constituents as they are readily available
from the HYSYS component databank. However, in most often cases, for crude and
other complex chemical compounds where it is hard to identify all the components,
the feedstock needs to be prepared based on pseudo-components, which are not
readily available in the databank. In order to accurately prepare the pseudo-
components for the crude slate, detailed lab data and analysis is needed. Once the data
is made available for different cuts in the crude, the pseudo-components can be
formed and named. The detailed data analysis is referred to as the Crude Assay. The
more accurately the crude assay is prepared, the more accurately the crude can be
simulated and hence the more reliable the results from the simulation are. Preparation
of the crude composition and its properties is also called Crude Characterization,
which will be discussed in detail in later sections of this chapter.
If there is more than one crude type in the feed, which is the case in this study, each
crude needs to be separately characterized and then the blend feature in HYSYS will
be used to make the required feedstock to the unit.
Once the crude characterization and blending process is done the next step would be
to select a proper thermodynamic package. As explained earlier, two thermodynamic
packages have been used for the crude distillation unit to predict the process more
45
accurately. BK-10 is selected to model the vacuum tower and equipment in that unit
as this thermodynamic package predicts low pressure hydrocarbon processes
accurately. Peng-Robinson is used to model the other parts of the CDU and is a more
generalized model, covering a wide range of temperatures and pressures for
hydrocarbon processes.
After crude characterization and selection of thermodynamic packages are complete,

Unit Operation in HYSYS should be set up in the flowsheeter, where different
equipments are connected through streams. To complete this part of the simulation a
comprehensive Front-End knowledge of the plant is needed. The simulation for this
study is a revamp simulation and therefore, UAs (total energy transferred) for heat
exchangers, number of trays, height of packed beds, internals and other information
for distillation columns, capacities for pumps, compressor curves, and other
equipment information are available. Therefore the performance of the plant can be
simulated under different operating conditions as the equipments are fixed. The
simulation for revamp projects is benchmarked against actual plant data so that
different operating envelopes could be covered and predicted, as done in this case.
The new equipments are specified based on the spec requirements for the new
operation in the plant.
Specifying and rating the equipments, needs a lot of time and attention as at this stage
all the parameters for the equipments are inputted, controls and recycle streams are
put in place and details of simulation are completed. Once all the equipment
information and spec required are inputted, the Unit Operation data entry is complete
and the flowsheeter can be run to achieve the results required.
Before starting the simulation work it is absolutely crucial to understand the process
scope, the feed composition and the operating conditions and constraints. These topics
are covered in the next section before starting the actual modeling process.
46
4.2. Overview of Crude Distillation Unit (CDU)
The major components of the CDU under study include the preheat train, desalter,
flash column, crude tower and the vacuum tower. As far as the scope of this thesis,
only the operation around the desalter will be considered as highlighted in the
following Block Flow Diagram (BFD). However, in order to properly design and
simulate the desalting process, the whole crude distillation unit should be simulated as
the desalting process is an integrated part of the CDU and hence the operating
conditions in the CDU will affect the desalter performance, specially the temperature
in the desalter, which is one of the key parameters for this study.
80 vol% Maya Cold Preheat 1st and 2nd Stage Hot Preheat
20 vol% Brent Train Crude Desalters Desalted Crude Train
70,000 BPSD 70,000 BPSD 70,000 BPSD
Stripped Sour Water

300 USGPM Brine to Sewer
Vacuum Condensate
25 USGPM 325 USGPM
River Water (NNF)
Offgas
Pre-flash Column
Hot Crude Light Naphtha Offgas
Atmospheric
Hvy. Naphtha
Column
DeC4 Pre-flash
OVHD Bottoms Kerosene
Diesel
AGO
Offgas
Vacuum LVGO
ATB Column
HVGO
VTB to DCU
Figure 4.2 - Block Flow Diagram for Crude Distillation Unit
47
4.3. Overall Project Scope
A North American refinery is going to revamp some of its processing units, including
the crude distillation unit. The existing capacity of the existing refinery is 60,000
BPSD and after revamp the refinery will be able to process 70,000 BPSD of crude. In
addition to increasing the capacity, the crude type will change from an average API
crude to a low API crude. As a result, the crude distillation unit will have to process a
higher capacity and a heavier crude blend. Therefore the crude distillation unit will be
revamped based on 70,000 BPSD of 80 vol% Maya and 20 vol% Brent. Maya crude is
a heavy crude and could contain significant salt in it. For revamp of the crude and
distillation unit, it is assumed that the Maya crude received by the refinery could
contain as high as 100 PTB of salt in current and future operation during the life cycle
of the plant. The revamp spec on salt is 1 PTB in the desalted crude stream, in order to
meet the requirement for the Vacuum Tower Bottoms (VTB) salt concentration.
Obviously with this spec on salt, other revamp criteria for desalting operation,
including protection of downstream equipment against corrosion and fouling as well
as protecting the reactor catalysts against poisoning, are satisfied.
The existing refinery configuration is based on a single stage electrostatic desalter,

which provides approximately 90% salt removal efficiency on medium crudes and
85% salt removal efficiency on heavy crudes such as Maya. To design equipments for
revamp of the desalting unit, it is assumed that the salt concentration in the crude,
received by the refinery, is 100 PTB. Therefore, given the best performance of the
existing single stage desalter, the desalted crude from 1st stage desalter would still
contain some 10 PTB salt, if only one stage of desalting is used. When Maya crude is
used in the refinery, after a single stage desalting operation, some 15 PTB of salt will
still remain in the desalted crude. Assuming a salt removal efficiency of 85% for the
1st stage desalter, the second stage desalter needs to be 92.5% or more efficient in
order to meet spec of 1 PTB, given there is 100 PTB of salt in the crude into the first
stage desalter. New electrostatic desalters can provide efficiencies as high as 99%
guaranteed by the vendor. Therefore, by installing a second stage desalter in the crude
distillation unit, the desalted crude spec of 1 PTB will be met. More details on salt
balance calculation is given further in this chapter.
In addition to installation of a new desalter in the crude distillation unit, it is required

to increase the desalter operating temperature to reach the optimum temperature for
desalting of Maya crude as discussed in chapter 3. These changes will conclude the
scope of revamp for crude distillation unit.
4.3.1 Process Design Criteria for Desalting Operation
The objective of crude desalting is removal of salts and solids and the formation water
from unrefined crude oil before the crude is introduced to the downstream equipment
in the refinery. The followings are the design parameters for revamp of the desalter:
• As per the results of the model in chapter 3, the temperature in the desalter
should be kept around 275ºF for optimum desalting operation of the Maya
crude.
48
• It is necessary to adjust the pH of the brine and keep it between 6 and 7 as this
range of pH will give the best results for breaking up the emulsions and hence
better efficiency in the desalter.
• The source of wash water for desalting is stripped sour water and as such it
may contain phenols in it. In order to minimize carryover of phenols, wash
water injection to the desalter should be maximized to reabsorb the phenols
into the crude streams.
• Desalting efficiency of 93.3% or more should be achieved through the 2nd
stage desalter to reduce the salt content of desalted crude stream to 1 PTB.
• Maximum of 0.5 vol% insoluble water content in the oil stream leaving the
desalter.
• Maximum of 200 ppmwt insoluble oil content in the effluent water stream
leaving the desalter system.
• The new desalter vessel should be designed for 100% unit capacity of 70,000
BPSD for the identified crude oil blend.
4.3.2 Feedstock
The following table shows the feedstock properties for Maya and Brent crudes.
Table 4.3.2: Feedstock properties for Crude Distillation Unit

Units Of 80/20 Blend
Feed Properties Maya Brent
Measure (HYSYS Sim.)
Unit Throughput BPSD 56,000 14,000 70,000
Volumetric Flow Rate VOL% 80 20 100
Specific Gravity @ 60ºF SG 0.93 0.84 0.88
API Gravity °API 21.3 38.0 24.1
TAN Mg KOH/g 0.3 0.05 -
Sulfur Content wt% 3.42 0.44 2.8
Nitrogen Content ppmwt 3600 1033 4633
Nickel ppmwt 55 2 57
Vanadium ppmwt 280 7 287
Salt Content lb/1000 bbl 30-80 22 125
Base Sediment and Water (BS&W) LV% 5 0.1 1.0
Viscosity @ 104ºF(40ºC): cSt 96 3.54 -
Viscosity @ 122ºF(50ºC): cSt 60 2.94 -
49
4.4. Crude Characterization
The crude for which the simulation has been prepared is a blend of 20 vol% Brent and
80 vol% Maya. As seen from the feedstock properties Table 4.3.2, Brent is a
relatively light and sweet crude from the North Sea and Maya is a heavy and sour
crude from Mexico. In order to blend the crudes in HYSYS, the assay for each crude
needs to be separately prepared. Then the two crudes are blended through the Basis
Environment Oil Manager in HYSYS. However, before this could be done in
HYSYS, the lab data on crude properties such as density, viscosity, etc. need to be
analyzed and prepared for use in HYSYS. Data analysis and preparation as well as lab
test results for properties of Brent and Maya crudes are presented in the following
sections.
4.4.1 Brent crude
Crude assay data was provided in two forms; vendor lab data and calculated data,
CAL II, used for LP analysis. Since the vendor data provided complete set of data for
analysis, this information was used to develop the Brent crude characterization.
The True Boiling Point (TBP) data for the crude assay is shown in Figure A1.0.
Figure A1.0 – Brent Characterization – Crude Assay – TBP EP vs. Cumulative LV%
This curve was also plotted on logarithmic coordinates, as shown in Figures A1.1 and
A1.2, to identify any irregularities. As shown by these curves, the data follows a
smooth distribution across each fraction and requires no adjustment.
50
Figure A1.1 – Brent Characterization – Crude Assay – TBP vs. Log Cumulative LV%
Figure A1.2 – Brent Characterization – Crude Assay – TBP vs. Log Residual LV%
51
Standard curve fitting procedures available in Excel were used to obtain a curve of
best fit through the data points. Due to the nature of the distribution, it was not
possible to adequately represent the whole crude with a single equation. Two separate
equations were required and the resulting curve is shown in Figure A1.3.
Figure A1.3 – Brent Characterization – Crude Assay – TBP EP vs. Cumulative LV%
The predicted crude curve was input to HYSYS. As shown in Figure B1.0, the crude
curve of Figure A1.3 matches the curve generated by HYSYS, which would ensure
that HYSYS simulation will behave as per this crude characterization.
The provided gravity data was plotted against the volume average true boiling point as
shown in Figure A2.0. The same data was plotted on logarithmic coordinates in
Figure A2.2, which shows that the provided gravity data is well behaved and follows
nearly a straight line. Next, the gravity curve was extrapolated linearly to find the best
fit through the raw data to cover the full range of the crude as shown in Figures A2.4
and A2.5. The curved portion of the gravity distribution was obtained by trial and
error, using the n-paraffin line (from the API Technical Database) as reference, to
converge with the n-paraffin line as shown in Figure A2.3. These steps were repeated
until reasonable weight balance on the whole crude was obtained. Figures A2.0, A2.1,
A2.2, A2.3, A2.4 and A2.5 are given below.
52
Figure A2.0 – Brent Characterization – Crude Assay – Raw Density vs. TBP
Figure A2.1 – Brent Characterization – Crude Assay – Vol Ave TBP vs. Ray Density
53
Figure A2.2 – Brent Characterization – Crude Assay – Log Vol Ave TBP vs. Raw Density
Figure A2.3 – Brent Characterization – Crude Assay – TBP vs. Raw Density
54
Figure A2.4 – Brent Characterization – Crude Assay – Log Vol Ave TBP vs. Raw Density
Figure A2.5 – Brent Characterization – Crude Assay – Raw Density vs. Log Vol Ave TBP
55
The gravity distribution across the whole crude as a function of percent distilled was
obtained from the fitted gravity curve of Figure A2.5. The resulting gravity
distribution curve is shown in Figure A3.0, which also shows crude assay data for
reference. The calculated API curve in Figure A3.1 shows good agreement with the
provided data.
Figure A3.0 – Brent Characterization – Crude Assay – Raw Density vs. Mid Cum LV%
Figure A3.1 – Brent Characterization – Crude Assay – Calculated Density vs. Cum LV%
56
The generated API density curve was input to HYSYS and compared with the curve
generated by HYSYS. As shown in Figure B2.0, the API density curve of Figure A3.0
matches the curve generated by HYSYS, except at the light ends and the 99% boiling
point of the curve. The light ends consist of 7 LV% and API density should
correspond to defined components such as propane, butane, etc. Since HYSYS
calculates density of these components from a standard database, the deviation is
acceptable. The deviation at the 99% boiling point will have no significant impact on
the process simulation.
The provided viscosity data are shown in Figure A4.0. The same viscosity data were
transformed to logarithmic coordinates and plotted against the provided gravity data
as shown in Figure A4.1. Each set of viscosity data can be represented with a straight
line of best fit through the data as shown in Figure A4.2. The viscosity distribution
across the whole crude was generated using Figure A4.2. The resulting distribution
curve is shown in Figure A4.3 and A4.4.
Figure A4.0 – Brent Characterization – Crude Assay – Raw Viscosity vs. Mid Cum LV%
57
Figure A4.1 – Brent Characterization – Crude Assay – Log Viscosity vs. API Density
58
Figure A4.4 – Brent Characterization – Crude Assay – Log Viscosity vs. Cumulative LV%
59
The viscosity curve was also compared to the curve generated by HYSYS. As shown
in Figure B3.0 and B3.1, the viscosity curve matches the HYSYS curve well, except
for the light ends. The light ends consist of 7 LV% and viscosity should correspond to
defined components such as propane, butane, etc. Since HYSYS calculates viscosity
of these components from a standard database, the deviation is acceptable.
Figures B1.0, B2.0, B3.0 and B3.1 are presented below.
Figure B1.0 – Brent Characterization – Comparative Plot – TBP vs. Cum LV%
60
Figure B2.0 – Brent Characterization – Comparative Plot – Density vs. Cum LV%
Figure B3.0 – Brent Characterization – Comparative Plot – Log Viscosity vs. Cum LV%
61
Figure B3.1 – Brent Characterization – Crude Assay – Log Viscosity vs. Cum LV%
The calculated Watson Characterization Factor (Kw) distribution is shown in Figure

C1.0. The crude assay did not provide values of Kw but the curve shape displays a
typical characterization factor curve. The usual curve starts with the high values of Kw
for the light ends, decreases to a minimum, and then increases.
Values for several petroleum properties were calculated in similar fashion and are
shown in Figures C2.0 through C5.0. Predicted values show a degree of agreement
with the reported data for Cetane Index and Freezing Point as shown in C2.0 and
C5.0. On the other hand, the calculated values of Cloud and Pour Points compare
poorly to the provided values, as shown in Figures C3.0 and C4.0. No further action
was taken. As a final check of the TBP and API curves, the weight of the whole crude
used for the assay was calculated and compared against the actual weight. Table 4.4.1
shows a series of volume fractions and the corresponding TBP as determined from
figure A1.4. For each TBP, the corresponding gravity was determined from the curve
of Figure A3.1. The weight of each volume fraction was added to obtain the
calculated weight of the whole crude.
62
Table 4.4.1 - Calculated Weight for Brent Crude
LV% Cum LV% API Calc SG Cut Vol mL Cut Avg. Wt g Cum Wt g
0 0 117.60 0.57 0.00 0.00 0.00
2 2 102.73 0.60 2.00 1.21 1.21
2 4 91.22 0.64 2.00 1.27 2.48
2 6 82.36 0.66 2.00 1.32 3.80
2 8 75.58 0.68 2.00 1.37 5.17
2 10 70.40 0.70 2.00 1.40 6.57
2 12 66.40 0.72 2.00 1.43 8.00
2 14 63.27 0.73 2.00 1.45 9.45
2 16 60.75 0.74 2.00 1.47 10.93
2 18 58.65 0.74 2.00 1.49 12.41
2 20 56.81 0.75 2.00 1.50 13.92
2 22 55.13 0.76 2.00 1.52 15.43
2 24 53.53 0.76 2.00 1.53 16.96
2 26 51.97 0.77 2.00 1.54 18.51
2 28 50.42 0.78 2.00 1.56 20.06
2 30 48.88 0.78 2.00 1.57 21.63
2 32 47.36 0.79 2.00 1.58 23.21
2 34 45.85 0.80 2.00 1.60 24.81
2 36 44.39 0.80 2.00 1.61 26.42
2 38 42.98 0.81 2.00 1.62 28.04
2 40 41.63 0.82 2.00 1.63 29.67
2 42 40.37 0.82 2.00 1.65 31.32
2 44 39.19 0.83 2.00 1.66 32.98
2 46 38.09 0.83 2.00 1.67 34.65
2 48 37.06 0.84 2.00 1.68 36.33
2 50 36.10 0.84 2.00 1.69 38.01
2 52 35.18 0.85 2.00 1.70 39.71
2 54 34.28 0.85 2.00 1.71 41.42
2 56 33.39 0.86 2.00 1.72 43.14
2 58 32.49 0.86 2.00 1.73 44.86
2 60 31.57 0.87 2.00 1.74 46.60
2 62 30.62 0.87 2.00 1.75 48.34
2 64 29.65 0.88 2.00 1.76 50.10
2 66 28.71 0.88 2.00 1.77 51.86
2 68 27.84 0.89 2.00 1.78 53.64
2 70 27.11 0.89 2.00 1.78 55.42
2 72 26.25 0.90 2.00 1.79 57.22
2 74 25.33 0.90 2.00 1.80 59.02
2 76 24.40 0.91 2.00 1.82 60.84
2 78 23.49 0.91 2.00 1.83 62.66
2 80 22.61 0.92 2.00 1.84 64.50
2 82 21.74 0.92 2.00 1.85 66.35
2 84 20.85 0.93 2.00 1.86 68.21
2 86 19.91 0.93 2.00 1.87 70.07
2 88 18.85 0.94 2.00 1.88 71.96
2 90 17.58 0.95 2.00 1.90 73.85
2 92 16.01 0.96 2.00 1.92 75.77
2 94 14.03 0.97 2.00 1.94 77.72
2 96 11.48 0.99 2.00 1.98 79.70
2 98 8.23 1.01 2.00 2.03 81.72
2 100 4.10 1.04 2.00 2.09 83.81
In view of the good weight balance and reasonable predictions for several petroleum
properties, further adjustment of the TBP and gravity curves was deemed
unnecessary.
The provided values for sulfur were also plotted and are shown in Figure C6.0.
Figures C1.0, C2.0, C3.0, C4.0, C5.0 and C6.0 are presented here.
63
Figure C1.0 – Brent Characterization – Crude Assay – Calculated Kw vs. Log Cum LV%
Figure C2.0 – Brent Characterization – Crude Assay – Cetane Index vs. Log Mid Cum LV%
64
Figure C3.0 – Brent Characterization – Product Assays – Cloud Point vs. Mid Cum LV%
Figure C4.0 – Brent Characterization – Product Assays – Pour Point vs. Mid Cum LV%
65
Figure C5.0 – Brent Characterization – Product Assays – Freeze Point vs. Mid Cum LV%
Figure C6.0 – Brent Characterization – Crude Assay – Sulfur Content wt% vs. Mid Cum LV%
66
4.4.2 Conclusions for Brent Crude
• The calculated crude assay data is entered into HYSYS.

• The adjusted gravity distribution is entered into HYSYS.
• Viscosity assays at 122oF and 212oF is entered into HYSYS.
• During process simulation, crude viscosity in the preheat train to be checked
against this crude characterization for consistency.
67
4.4.3 Maya Crude
Crude assay data was provided in two forms: vendor lab data and calculated data or
CAL II data. Since the vendor lab data provided is incomplete, CAL II is used to
develop the Maya crude characterization.
The PCP CAL II data consists of an extensive set of boiling point curves and
petroleum properties for distillate and residue cuts that closely match refinery
products. These were generated from the crude assay by the PCP CAL II library,
which is also used as input for the refinery’s LP modeler.
This set of data is, in effect, an extrapolation of the raw crude assay data. The
provided data was analyzed for consistency and accuracy in predicting selected
petroleum properties. The result of this effort is represented in figures below.
The TBP data for each of the cuts is shown in Figure D1.0. This data was also plotted
on logarithmic coordinates, as shown in Figures D1.1 and D1.2, to identify any
irregularities. As shown by these curves, the data follows a smooth distribution across
each fraction and requires no adjustment.
Each of the True Boiling Point (TBP) distributions was plotted to construct the whole
crude, as shown in Figure D1.3. The composite TBP distribution curve for the whole
crude was constructed by joining the mid-volume percent of the individual curves
shown in Figure D1.3. Standard curve fitting procedures available in Excel were used
to obtain a curve of best fit through the data points. Due to the nature of the
distribution, it was not possible to adequately represent the whole crude with a single
equation. Three separate equations were required and the resulting curve is shown in
Figure D1.4.
The composite curve of Figure D1.4 and the raw distillation data from the crude assay
are shown in Figure D1.5 for comparison. There is a perfect agreement between the
two curves. The composite curve tends to over predict in the distillate range by up to 3
LV%, which is within the range of experimental error.
The composite curve was also compared to the blended curve generated by HYSYS
from each of the cuts. As shown in Figure D1.6, the composite curve of Figure D1.4
matches the HYSYS curve well.
68
Figure D1.0 – Maya Characterization – CALII Cuts – TBP vs. Cumulative LV%
Figure D1.1 – Maya Characterization – CALII Cuts – TBP vs. Log Cumulative LV%
69
Figure D1.2 – Maya Characterization – CALII Cuts – TBP vs. Log Residual LV%
70
Figure D1.5 – Maya Characterization – Comparative Plot – TBP vs. Cumulative LV%
71
Figure D1.6 – Maya Characterization – Comparative Plot – TBP vs. Cumulative LV%
The provided gravity data was plotted against the volume average true boiling point of
each cut as shown in Figure D2.0.
Figure D2.0 – Maya Characterization – CALII Cuts – Volume Ave TBP vs. Density
72
The same data was plotted on logarithmic coordinates in Figure D2.1, which shows
that the provided gravity data is well behaved and follows very nearly a straight line.
Figure D2.1 – Maya Characterization – CALII Cuts – Log Vol Ave TBP vs. Density
Next, the gravity curve was extended by a straight-line best fit through the raw data
and extrapolation to 100 LV% to cover the full range of the crude. The curved portion
of the gravity distribution was obtained by trial and error, using the n-paraffin line as
reference, to converge with the n-paraffin line as shown in Figure D2.3. These steps
were repeated until reasonable predictions for petroleum properties were obtained
simultaneously with a weight balance on the whole crude.
73
Figure D2.2 – Maya Characterization – CALII Cuts – Log Vol Ave TBP vs. Density
Figure D2.2a – Maya Characterization – CALII Cuts Linear Segment – Log Vol Ave TBP vs. Density
74
Figure D2.2b – Maya Characterization – CALII Cuts Curved Segment – Log Vol Ave TBP vs. Density
Figure D2.3 – Maya Characterization – CALII Cuts – Vol Ave TBP vs. Density
75
The gravity distribution across the whole crude as a function of percent distilled was
obtained from the fitted gravity curve of Figure D2.2. The resulting gravity
distribution curve is shown in Figure D3.0, which also shows the CAL II data for
reference. The calculated API curve shows good agreement with the provided data.
Since many of the property prediction methods are based on empirical correlations
between TBP and API, the adequacy of the fitted curves were checked through
calculation of a series of petroleum properties as outlined in the API Technical Data
Book.
Figure D3.0 – Maya Characterization – CALII Cuts – Density vs. Cumulative LV%
The Watson Characterization factor (Kw) was calculated at each LV% distilled from
the appropriate TBP curve shown in Figure D1.4 and the corresponding value of
gravity from the curve of Figure D2.2. The calculated Kw distribution is shown in
Figure D4.0. The provided values of Kw were plotted on the same figure for
comparison. The results display a typical characterization factor curve and show good
agreement between the calculated and provided data.
76
Figure D4.0 – Maya Characterization – CALII Cuts – Kw vs. Cumulative LV%
Values for several petroleum properties were calculated in similar fashion and are
shown in Figures D5.0 through D8.0.
Figure D5.0 – Maya Characterization – CALII Cuts – Cetane Index vs. Cumulative LV%
77
Figure D6.0 – Maya Characterization – CALII Cuts – Cloud Point vs. Cumulative LV%
Figure D7.0 – Maya Characterization – CALII Cuts – Pour Point vs. Cumulative LV%
78
Figure D8.0 – Maya Characterization – CALII Cuts – Freeze Point vs. Cumulative LV%
With the exception of Cetane Index, each of the predicted values shows good
agreement with the reported data in the range of interest that is 30-60 LV% distilled.
This range corresponds to distillate fractions that have product specifications.
As a final check of the TBP and API curves, the weight of the whole crude used for
the assay was calculated and compared against the actual weight. Table 4.4.3a shows
a series of volume fractions and the corresponding TBP as determined from Figure
D1.4. For each TBP, the corresponding gravity was determined from the curve of
Figure D2.2. The weight of each volume fraction was added to obtain the calculated
weight of the whole crude. The actual weight of the whole crude was determined from
the measured whole crude gravity provided with the crude assay as shown in Table
4.4.3b. The results are summarized in Table 4.4.3c, which shows weight balance
within 0.56%.
In view of the good weight balance and reasonable predictions for several petroleum
properties, further adjustment of the TBP and gravity curves was deemed
unnecessary.
79
Table 4.4.3a - Calculated Weight for Maya Crude
LV% Cum LV% API Calc SG Cut Vol mL Cut Avg. Wt g Cum Wt g
0 0 266.00 0.36 0.00 0.00 0.00
2 2 98.22 0.62 2.00 1.23 1.23
2 4 80.18 0.67 2.00 1.34 2.57
2 6 71.73 0.70 2.00 1.39 3.96
2 8 64.21 0.72 2.00 1.45 5.41
2 10 59.95 0.74 2.00 1.48 6.89
2 12 56.42 0.75 2.00 1.51 8.39
2 14 53.29 0.77 2.00 1.53 9.92
2 16 50.48 0.78 2.00 1.56 11.48
2 18 47.92 0.79 2.00 1.58 13.06
2 20 45.58 0.80 2.00 1.60 14.65
2 22 43.40 0.81 2.00 1.62 16.27
2 24 41.37 0.82 2.00 1.64 17.91
2 26 39.46 0.83 2.00 1.66 19.56
2 28 37.64 0.84 2.00 1.67 21.24
2 30 35.91 0.85 2.00 1.69 22.93
2 32 34.24 0.85 2.00 1.71 24.64
2 34 32.63 0.86 2.00 1.72 26.36
2 36 31.07 0.87 2.00 1.74 28.10
2 38 29.54 0.88 2.00 1.76 29.86
2 40 28.05 0.89 2.00 1.77 31.63
2 42 26.58 0.90 2.00 1.79 33.42
2 44 25.13 0.90 2.00 1.81 35.23
2 46 23.69 0.91 2.00 1.82 37.05
2 48 22.27 0.92 2.00 1.84 38.89
2 50 20.85 0.93 2.00 1.86 40.75
2 52 19.44 0.94 2.00 1.87 42.63
2 54 18.03 0.95 2.00 1.89 44.52
2 56 17.44 0.95 2.00 1.90 46.42
2 58 16.40 0.96 2.00 1.91 48.33
2 60 15.31 0.96 2.00 1.93 50.26
2 62 14.17 0.97 2.00 1.94 52.20
2 64 13.00 0.98 2.00 1.96 54.16
2 66 11.81 0.99 2.00 1.97 56.13
2 68 10.60 1.00 2.00 1.99 58.13
2 70 9.39 1.00 2.00 2.01 60.14
2 72 8.17 1.01 2.00 2.03 62.16
2 74 6.97 1.02 2.00 2.04 64.21
2 76 5.77 1.03 2.00 2.06 66.27
2 78 4.59 1.04 2.00 2.08 68.35
2 80 3.43 1.05 2.00 2.10 70.44
2 82 2.29 1.06 2.00 2.12 72.56
2 84 1.18 1.07 2.00 2.13 74.69
2 86 0.10 1.08 2.00 2.15 76.84
2 88 -0.96 1.08 2.00 2.17 79.01
2 90 -1.98 1.09 2.00 2.18 81.20
2 92 -2.97 1.10 2.00 2.20 83.40
2 94 -3.92 1.11 2.00 2.22 85.62
2 96 -4.84 1.12 2.00 2.23 87.85
2 98 -5.72 1.12 2.00 2.25 90.10
2 100 -6.56 1.13 2.00 2.27 92.36
Table 4.4.3b - Actual Cumulative Weight of the Whole Maya Crude

LV% Cum LV% API Raw SG Cut Vol mL Cut Avg. Wt g Cum Wt g
100 100 21.3 0.93 100 92.60 92.60
Table 4.4.3b – Summary of Results for Maya Crude Actual Weight

Calculated Wt g Actual Wt g Delta g Delta %
92.36 92.60 0.24 0.26
80
The provided viscosity data are shown in Figure D9.0. The same viscosity data were
transformed to logarithmic coordinates and plotted against the provided gravity data
for each cut as shown in Figure D9.1.
Figure D9.0 – Maya Characterization – CALII Cuts – Viscosity vs. Cumulative LV%
Figure D9.1 – Maya Characterization – CALII Cuts – Kinematic Viscosity vs. CAL II Density
81
Figure D9.1B – Maya Characterization – CALII Cuts – Kinematic Viscosity vs. CAL II Density
Figure D9.2 – Maya Characterization – CALII Cuts – Viscosity vs. Density
Each set of viscosity data can be represented with a straight line of best fit through the
data as shown. The viscosity distribution across the whole crude was generated using
Figures D2.2 and D1.4. The resulting distribution curve is shown in Figure D9.3.
82
Figure D9.3 – Maya Characterization – CALII Cuts – Viscosity vs. Cumulative LV%
The provided values for sulfur, metals, flash point and molecular weight were also
plotted and are shown in Figures D10.0 through D12.0.
Figure D10.0 – Maya Characterization – CALII Cuts – Sulfur Content wt% vs. Cumulative LV%
83
Figure D11.0 – Maya Characterization – CALII Cuts – Flash Point vs. Cumulative LV%
Figure D12.0 – Maya Characterization – CALII Cuts – Molecular Weight vs. Cumulative LV%
84
Figures E1.0 through to E5.0 show comparison of HYSYS generated properties vs.
data generated in this crude characterization document.
Figure E1.0 – Maya Characterization – Product Assays – Molecular Weight vs. Cumulative LV%
Figure E2.0 – Maya Characterization – Product Assays – Log Viscosity vs. Cumulative LV%
85
Figure E3.0 – Maya Characterization – Product Assays – Log Viscosity vs. Cumulative LV%
Figure E4.0 – Maya Characterization – Product Assays – Density vs. Cumulative LV%
86
Figure E5.0 – Maya Characterization – Product Assays – Kw vs. Cumulative LV%
4.4.4 Conclusions for Maya Crude
• The adjusted CAL II TBP assays for the distillate and residue cuts is entered into
HYSYS.
• The adjusted gravity distribution shown in Figure D3.0 is entered into HYSYS to
generate consistent property data, as outlined here.
• Viscosity assays at 104oF and 212oF are entered into HYSYS based on the best fit
lines shown in Figure D9.1.
87
4.5. Thermodynamic Package
When setting up the simulation, a great deal of attention should be paid to selection of
thermodynamic package and the limitation of operating envelope in which one
package can predict the simulation results. As far as crude distillation unit, the most
widely used thermodynamic package in HYSYS is Peng-Robinson (PR). This
package was used for the simulation of the crude distillation unit. However, this
package does not produce accurate results for the vacuum section and hence the BK-
10 thermodynamic package was used to simulate the vacuum section. Components of
the vacuum tower that utilized the BK-10 thermo package are built in a sub-flowsheet
and the rest of the process is simulated in the main flow sheet using PR
thermodynamic package. The result of the simulation using these thermodynamic
packages have been checked and benchmarked against actual operational data and
satisfactory results have been achieved.
The Peng-Robinson Equation of State (EOS) uses binary interaction to tune the model
to experimental data. The semi-theoretical basis for these factors is that they account
for shape and size differences between components but they are effectively fudge
factors used to tune the model to experimental data. There are two interaction
coefficients for each pair of components in the mixture; the assumption is made that
these interaction coefficients are not composition or temperature dependent though in
practice they are. HYSYS contains a library of binary interactions coefficients. As the
values used are often proprietary, different programs using the same EOS can give
different Vapor/Liquid Equilibria (VLE). Discrepancies between programs may also
be due to differences in the components mixing rules used. Interaction coefficients are
close to unity for large non-polar, binary pairs such as alkanes, but deviate strongly if
one or both of the components is small or polar, or both. Methane, Hydrogen,
Nitrogen, Carbon Dioxide and Hydrogen Sulfide all fall into this category. Since the
library binary interaction coefficients must cover the full range of mixtures, pressures
and temperatures to which the simulation may be applied, the values may not be the
best available for a specific system.
Binary interaction coefficients are calculated by regression analysis of the

experimental data. HYSYS also has the capability to develop sets of binary interaction
coefficients from experimental data.
88
4.6. Process Description
The process description entailed here only covers the desalting operation as per scope
of this study. For better understanding of the process description, refer to Process
Flow Diagrams in Appendix A.
Crude from the storage tanks is pumped through charge pumps P-1001A/B/C and
goes through a number of heat exchangers in the pre-heat train to absorb the heat from
hot products and the pump-around streams to elevate the temperature of the crude to
the required temperature for desalting operation. Preheated crude will then enter the
1st stage desalter V-1001 and subsequently goes through to the 2nd stage desalter V-
1002 for the salt content to be reduced to 1 PTB at the outlet of the 2nd stage desalter.
The desalted crude will then go through the hot preheat train to be further heated
against pump-around streams and VTB product to reach the required temperature
before being fed to the pre-flash column C-1001.
In the desalting operation, the main supply of wash water is the Stripped Sour Water
(SSW) coming from the Sour Water Treatment Unit. The SSW has traces of
hydrocarbon and is rich in phenols. Using the SSW as wash water will allow the
phenols to get reabsorbed in the crude oil before disposing of the wash water. The
idea is to maximize use of the SSW in the desalting operation in order to minimize
brine treatment and hence minimize any potential environmental impact. Some 300
USGPM of SSW is available in the plant and hence the idea is to use it as wash water
for the desalting operation. The other source of water, which also contains phenols, is
25 USGPM of condensate coming from vacuum tower barometric condenser.
Therefore 325 USGPM of water is available as wash water for the desalting operation.
Stripped sour water will go through wash water pumps P-1020A/B. Vacuum Steam
Condensate will also be available for wash water and supplied to the discharge of
wash water pumps and then the combined water will pass through the shell side of the
wash water/brine exchangers E-1020A/B/C. The wash water is heated from about
100ºF to 236ºF through E-1020A/B/C and then subsequently through the wash water
steam heater E-1021, raising the wash water temperature to about 373ºF by control on
High Pressure (HP) steam supply from the header. The flow of wash water through
the wash water steam heater can be adjusted by means of a bypass control valve.
In normal operation mode, the hot wash water from the outlet of E-1021 will be split
and half of the flow will be fed to the crude stream just before going through the 1st
stage mixing valve. The other half of the wash water flow is injected to the crude
stream coming out of the 1st stage desalter and going through the 2nd stage mixing
valve. The wash water and crude are mixed through the mixing valves and emulsified
into fine droplets to be effectively demulsified in the desalter.
The desalters use electrostatic DC field to polarize the water droplets to increase the
coalescence and hence the efficiency of the desalter. By removing the water from the
crude, the crude is stripped of salts which are dissolved in the water. The brine will go
to the bottom of the desalters and the crude will leave the desalter from the top. The
brine streams from 1st and 2nd stage desalters will flow under desalter level control
and will join together before going through the tube side of the wash water and brine
exchanger E-1020A/B/C to be cooled down to 130ºF. The brine exit temperature is
89
determined by the constraints in the plant and is not allowed to be higher than 130oF
for safety reasons. The brine effluent will then leave the battery limit to go to the oily
water sewer.
Desalter water can also be run in recycle mode. This will be done particularly when
the SSW is in short supply. In this mode about 150 USGPM of SSW will be injected
by pump P-1023 upstream of the 2nd stage desalter mixing valve through exchangers
E-1021A/B/C and E-1021 and wash water from bottom of 2nd stage desalter will be
injected by desalter recycle pump P-1023 to upstream of the 1st stage desalter mixing
valve. The 2nd stage desalter will have a 2nd stage desalter mud wash pump P-1022 for
mud washing.
4.7. Process Flow Diagrams (PFDs)

The Process Flow Diagrams are developed for better understanding of the process.
The clouds indicate changes as part of the revamp project. Unclouded parts indicate
unchanged existing facilities. The PFDs are presented in the Appendix A.
4.8. Heat and Material Balance (H&MB)

The result of simulation is a full H&MB for the crude distillation unit and is presented
below in the next few pages. The H&MB contains the cold and hot preheat train
streams, as well as the desalting unit streams. As discussed earlier, HYSYS cannot be
used to produce a salt balance for the desalting operation. Therefore, Excel
spreadsheet has been used to prepare Table 4.8.2a and Table 4.8.2b to present the salt
and water balance for different streams of the desalting unit. For simulation of brine
or salty streams in HYSYS, pure water assumption has been made. The energy or heat
balance still holds for the salty water or brine streams as far as HYSYS simulation.
90
Table 4.8.1a: H&MB for CDU – Streams 2, 3, 4, 5, 6 and 7
Stream Name 2 3 4 5 6 7
Stream Description FEED CRUDE CRUDE CRUDE CRUDE CRUDE
Phase Liquid Liquid Liquid Liquid Liquid Liquid
Total Stream Properties
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 55.0 56.1 61.2 90.3 158.7 167.4
Pressure psia 37.7 421.5 399.9 380.5 355.0 339.7
Molar Flow lbmol/hr 3774.2 3774.2 3774.2 3774.2 3774.2 3774.2
Mass Flow lb/hr 927806.7 927806.7 927806.7 927806.7 927806.7 927806.7
Std. Ideal Liquid Volume Flow barrel/day 70001 70001 70001 70001 70001 70001
Molecular Weight 245.8 245.8 245.8 245.8 245.8 245.8
Heat Flow Btu/hr -8.7860E+08 -8.7706E+08 -8.7412E+08 -8.6273E+08 -8.3384E+08 -8.3000E+08
Molar Enthalpy Btu/lbmole -232791 -232381 -231605 -228585 -220931 -219912
Mass Heat Capacity Btu/lb-F 0.408 0.407 0.414 0.434 0.479 0.485
Mass Density lb/ft3 56.80 57.04 56.99 56.25 54.54 54.30
Watson K 11.312 11.312 11.312 11.312 11.312 11.312
Vapor Phase Properties
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Actual Volume Flow ft3/hr
Mass Density lb/ft3
Compressibility
Thermal Conductivity Btu/hr-ft-F
Viscosity cP
Molecular Weight
Liquid Phase Properties
Mass Flow lb/hr 927806.7 927806.7 927806.7 927806.7 927806.7 927806.7
Thermal Conductivity Btu/hr-ft-F 0.09 0.09 0.09 0.08 0.08 0.08
Viscosity cP 83.00 80.00 68.00 28.00 7.80 6.80
Surface Tension dyne/cm 74.21 74.10 30.50 29.24 26.36 25.97
Actual Volume Flow ft3/hr 16335.7 16266.5 16280.3 16494.3 17012.8 17085.4
Aqueous Phase Properties
Molar Flow lbmol/hr
Mass Flow lb/hr
Std. Ideal Liquid Volume Flow barrel/day
Heat Flow Btu/hr
Viscosity cP
Surface Tension dyne/cm
91
Table 4.8.1b: H&MB for CDU – Streams 8, 8A, 8B, 9A, 9B and 10A
Stream Name 8 8A 8B 9A 9B 10A
Stream Description CRUDE CRUDE CRUDE CRUDE CRUDE CRUDE
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 215.8 215.8 215.8 215.9 215.9 266.2
Pressure psia 317.7 317.7 317.7 291.6 291.1 281.6
Mass Flow lb/hr 927806.7 473181.4 454625.3 473181.4 454625.3 473181.4
Watson K 11.312 11.312 11.312 11.312 11.312 11.312
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 927806.7 473181.4 454625.3 473181.4 454625.3 473181.4
Viscosity cP 3.90 2.04 3.90 3.90 3.90 2.40
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
92
Table 4.8.1c: H&MB for CDU – Streams 10B, 10, 111, 112, 113 and 118
Stream Name 10B 10 111 112 113 118
Stream Description CRUDE CRUDE NAPHTHA NAPHTHA NAPHTHA KERO.
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 266.1 266.2 339.2 244.0 219.2 395.3
Pressure psia 281.6 281.6 220.9 210.9 200.9 222.1
Mass Flow lb/hr 454625.3 927806.7 409399.0 409399.0 409399.0 117952.1
Watson K 11.312 11.312 11.684 11.684 11.684 11.581
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 454625.3 927806.7 409399.0 409399.0 409399.0 117952.1
Viscosity cP 2.40 2.40 0.05 0.10 0.12 0.05
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
93
Table 4.8.1d: H&MB for CDU – Streams 119, 128, 129, 208, 209 and 220
Stream Name 119 128 129 208 209 220
Stream Description KERO. DIESEL DIESEL LVGO P/A LVGO P/A HVGO P/A
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 231.7 458.7 160.0 243.6 164.1 223.7
Pressure psia 212.1 222.8 212.8 200.3 190.3 180.4
Mass Flow lb/hr 117952.1 22465.7 22465.7 70436.8 70436.8 278991.3
Watson K 11.581 11.538 11.538 11.546 11.546 11.502
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 117952.1 22465.7 22465.7 70436.8 70436.8 278991.3
Viscosity cP 0.19 0.09 0.62 0.25 0.45 1.74
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
94
Table 4.8.1e: H&MB for CDU – Streams 232 and 240
Stream Name 232 240
Stream Description VTB VTB
Phase Liquid Liquid
Vapour Fraction 0.00 0.00
Temperature F 405.8 420.6
Pressure psia 180.5 180.5
Molar Flow lbmol/hr 358.8 395.0
Mass Flow lb/hr 196115.9 215886.9
Std. Ideal Liquid Volume Flow barrel/day 12934 14238
Molecular Weight 546.6 546.6
Heat Flow Btu/hr -1.5322E+08 -1.6681E+08
Molar Enthalpy Btu/lbmole -427045 -422349
Mass Heat Capacity Btu/lb-F 0.578 0.586
Mass Density lb/ft3 56.91 56.59
Watson K 11.435 11.435
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 196115.9 215886.9
Mass Heat Capacity Btu/lb-F 0.578 0.586
Std. Ideal Liquid Volume Flow barrel/day 12934 14238
Thermal Conductivity Btu/hr-ft-F 0.09 0.09
Viscosity cP 13.32 12.12
Surface Tension dyne/cm 22.83 22.50
Actual Volume Flow ft3/hr 3446.3 3814.8
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
95
Table 4.8.1f: H&MB for CDU – Streams 129, 208, 209, 220, 232 and 240
Stream Name 129 208 209 220 232 240
Stream Description DIESEL LVGO P/A LVGO P/A HVGO P/A VTB VTB
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 160.0 243.6 164.1 223.7 405.8 420.6
Pressure psia 212.8 200.3 190.3 180.4 180.5 180.5
Mass Flow lb/hr 22465.7 70436.8 70436.8 278991.3 196115.9 215886.9
Watson K 11.538 11.546 11.546 11.502 11.435 11.435
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 22465.7 70436.8 70436.8 278991.3 196115.9 215886.9
Viscosity cP 0.62 0.25 0.45 1.74 13.32 12.12
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
96
Table 4.8.1g: H&MB for CDU – Streams 11, 12, 14, 70, 70A and 70B
Stream Name 11 12 14 70 70A 70B
DESALTED WASH WASH WASH
Stream Description CRUDE WW CRUDE WW CRUDE WATER WATER WATER
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 269.9 270.0 272.1 100.0 100.0 101.4
Pressure psia 281.6 263.6 243.6 114.7 29.7 370.7
Mass Flow lb/hr 1008692.5 1008692.5 936887.5 149913.5 149913.5 149913.5
Watson K 11.312 11.312 11.312
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Molar Flow lbmol/hr 3857.0 3857.2 3868.4
Mass Flow lb/hr 930428.6 930432.1 931763.2
Molecular Weight 241.2 241.2 240.9
Heat Flow Btu/hr -7.9172E+08 -7.9175E+08 -7.9207E+08
Mass Heat Capacity Btu/lb-F 0.549 0.549 0.550
Std. Ideal Liquid Volume Flow barrel/day 70203 70203 70316
Mass Density lb/ft3 51.63 51.61 51.52
Thermal Conductivity Btu/hr-ft-F 0.07 0.07 0.07
Viscosity cP 1.29 1.29 1.27
Surface Tension dyne/cm 18.62 18.62 18.54
Actual Volume Flow ft3/hr 18021.9 18028.4 18084.2
Mass Flow lb/hr 78263.9 78260.4 5124.3 149913.5 149913.5 149913.5
Std. Ideal Liquid Volume Flow barrel/day 5369.8 5369.5 351.6 10285.7 10285.7 10285.7
Thermal Conductivity Btu/hr-ft-F 0 0 0 0 0 0
Viscosity cP 0.207 0.207 0.205 0.680 0.680 0.670
97
Table 4.8.1h: H&MB for CDU – Streams 71, 72, 73, 74, 75 and 78
Stream Name 71 72 73 74 75 78
VAC. WASH WASH WASH WASH WASH
Stream Description COND. WATER WATER WATER WATER WATER
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 102.3 101.4 236.1 236.1 236.1 373.6
Pressure psia 328.7 328.7 318.7 318.7 318.7 308.7
Mass Flow lb/hr 11858.2 161771.6 161771.6 56620.1 105151.6 56620.1
Watson K 11.646 11.646 11.646 11.646 11.646 11.646
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 2566.4 2564.5 2567.5 898.6 1668.9 904.8
Viscosity cP 0.29 0.29 0.12 0.12 0.12 0.05
Mass Flow lb/hr 9291.8 159207.1 159204.1 55721.4 103482.6 55715.2
Viscosity cP 0.664 0.670 0.244 0.244 0.244 0.141
98
Table 4.8.1i: H&MB for CDU – Streams 79, 82, 83, 84, 85 and 86
Stream Name 79 82 83 84 85 86
WASH WASH WASH WASH CRUDE
Stream Description WATER WATER WATER WATER CRUDE WW
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 236.1 285.0 285.0 285.0 270.0 272.1
Pressure psia 308.7 308.7 281.6 263.6 263.6 263.6
Mass Flow lb/hr 105151.6 161771.6 80885.8 80885.8 930432.1 1011317.9
Watson K 11.646 11.646 11.646 11.646 11.312 11.312
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 1668.9 2571.0 1285.5 1285.5 930432.1 931759.3
Viscosity cP 0.12 0.09 0.09 0.09 1.29 1.28
Molar Flow lbmol/hr 5744.2 8837.0 4418.5 4418.5 4416.2
Mass Flow lb/hr 103482.6 159200.6 79600.3 79600.3 79558.6
Std. Ideal Liquid Volume Flow barrel/day 7100.1 10923.0 5461.5 5461.5 5458.6
Heat Flow Btu/hr -6.8739E+08 -1.0496E+09 -5.2478E+08 -5.2478E+08 -5.2557E+08
Thermal Conductivity Btu/hr-ft-F 0 0 0 0 0
Viscosity cP 0.244 0.194 0.194 0.194 0.205
Surface Tension dyne/cm 56.035 50.626 50.626 50.626 52.085
99
Table 4.8.1j: H&MB for CDU – Streams 87, 88, 90, 91, 91A and 92
Stream Name 87 88 90 91 91A 92
CRUDE
Stream Description WW BRINE BRINE BRINE BRINE BRINE
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 272.1 272.1 272.1 270.0 270.0 271.0
Pressure psia 243.6 243.6 238.6 263.6 238.6 238.6
Mass Flow lb/hr 1011317.9 74430.4 74430.4 78260.4 78260.4 152690.8
Watson K 11.312 9.718 9.718 10.206 10.206 9.894
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Molar Flow lbmol/hr 3868.4
Mass Flow lb/hr 931763.2
Molecular Weight 240.9
Heat Flow Btu/hr -7.9207E+08
Mass Heat Capacity Btu/lb-F 0.550
Std. Ideal Liquid Volume Flow barrel/day 70316
Mass Density lb/ft3 51.52
Thermal Conductivity Btu/hr-ft-F 0.07
Viscosity cP 1.27
Surface Tension dyne/cm 18.54
Actual Volume Flow ft3/hr 18084.2
Mass Flow lb/hr 79554.6 74430.4 74430.4 78260.4 78260.4 152690.8
Viscosity cP 0.205 0.205 0.205 0.207 0.207 0.206
100
Table 4.8.1k: H&MB for CDU – Streams 93, 94, 98 and 99
Stream Name 93 94 98 99
Stream Description BRINE BRINE HP STEAM STEAM COND.
Phase Liquid Liquid Vapour Liquid
Vapour Fraction 0.00 0.00 1.00 0.00
Temperature F 130.0 130.0 520.0 399.7
Pressure psia 228.6 25.0 254.7 244.7
Molar Flow lbmol/hr 8475.7 8475.7 493.5 493.5
Mass Flow lb/hr 152690.8 152690.8 8890.7 8890.7
Std. Ideal Liquid Volume Flow barrel/day 10476 10476 610 610
Molecular Weight 18.0 18.0 18.0 18.0
Heat Flow Btu/hr -1.0305E+09 -1.0305E+09 -4.9542E+07 -5.7543E+07
Molar Enthalpy Btu/lbmole -121587 -121587 -100386 -116598
Mass Heat Capacity Btu/lb-F 0.999 0.999 0.571 1.080
Mass Density lb/ft3 61.51 61.48 0.46 53.02
Watson K 9.894 9.894
Actual Volume Flow ft3/hr 19267.9
Mass Density lb/ft3 0.46
Compressibility 0.95
Thermal Conductivity Btu/hr-ft-F 0.02
Viscosity cP 0.02
Molecular Weight 18.0
Molar Flow lbmol/hr
Mass Flow lb/hr
Molecular Weight
Heat Flow Btu/hr
Mass Heat Capacity Btu/lb-F
Mass Density lb/ft3
Viscosity cP
Mass Flow lb/hr 152690.8 152690.8 8890.7
Std. Ideal Liquid Volume Flow barrel/day 10476.3 10476.3 610.0
Thermal Conductivity Btu/hr-ft-F 0 0 0
Viscosity cP 0.505 0.505 0.131
101
Table 4.8.1l: H&MB for CDU – Streams 14A, 14B, 14C, 15A, 15B and 15C
Stream Name 14A 14B 14C 15A 15B 15C
Phase Liquid Liquid Liquid Liquid Mixed Mixed
Vapour Fraction 0.00 0.00 0.00 0.00 0.07 0.17
Temperature F 272.1 272.1 272.1 291.9 334.8 372.0
Pressure psia 243.6 243.6 243.6 144.0 191.3 118.7
Mass Flow lb/hr 435465.3 255957.7 245464.5 435465.3 255957.7 245464.5
Watson K 11.312 11.312 11.312 11.312 11.312 11.312
Mass Flow lb/hr 3076.4 10340.8
Actual Volume Flow ft3/hr 3108.4 13154.5
Compressibility 0.92 0.94
Thermal Conductivity Btu/hr-ft-F 0.02 0.02
Viscosity cP 0.01 0.01
Mass Flow lb/hr 433083.6 254557.7 244122.0 433294.5 252727.5 235123.7
Viscosity cP 1.27 1.27 1.27 1.15 1.01 0.99
Mass Flow lb/hr 2381.8 1399.9 1342.6 2170.8 153.8
Std. Ideal Liquid Volume Flow barrel/day 163.4 96.1 92.1 148.9 10.6
Thermal Conductivity Btu/hr-ft-F 0 0 0 0 0
Viscosity cP 0.205 0.205 0.205 0.189 0.160
Surface Tension dyne/cm 52.077 52.077 52.077 49.838 44.828
102
Table 4.8.1m: H&MB for CDU – Streams 16A, 16B, 16C, 17, 17A, 17B
Stream Name 16A 16B 16C 17 17A 17B
Phase Mixed Mixed Mixed Mixed Mixed Mixed
Vapour Fraction 0.08 0.15 0.19 0.14 0.14 0.14
Temperature F 284.5 327.5 370.1 317.4 317.4 317.4
Pressure psia 102.5 102.5 102.5 102.5 102.5 102.5
Mass Flow lb/hr 435465.3 255957.7 245464.5 936887.5 468443.7 468443.7
Watson K 11.312 11.312 11.312 11.312 11.312 11.312
Mass Flow lb/hr 6897.2 8200.3 11830.3 27170.8 13585.4 13585.4
Compressibility 0.95 0.95 0.94 0.95 0.95 0.95
Viscosity cP 0.01 0.01 0.01 0.01 0.01 0.01
Mass Flow lb/hr 427782.4 247757.4 233634.2 909716.7 454858.3 454858.3
Viscosity cP 1.28 1.19 1.09 1.23 1.23 1.23
Std. Ideal Liquid Volume Flow barrel/day 53.9
Thermal Conductivity Btu/hr-ft-F 0
Viscosity cP 0.195
Surface Tension dyne/cm 50.684
103
Table 4.8.1n: H&MB for CDU – Streams 18A, 18B, 18, 20, 77 and 123
Stream Name 18A 18B 18 20 77 123
Stream Description CRUDE CRUDE CRUDE CRUDE DE-C4 DIESEL
Phase Mixed Mixed Mixed Mixed Mixed Liquid
Vapour Fraction 0.24 0.25 0.32 0.32 0.00 0.00
Temperature F 378.6 385.2 375.2 372.1 110.2 521.1
Pressure psia 78.5 78.5 43.3 43.3 45.0 123.0
Mass Flow lb/hr 468443.7 468443.7 936887.5 947140.3 10252.9 163879.4
Watson K 11.312 11.312 11.312 11.314 11.717 11.540
Mass Flow lb/hr 32003.3 33858.5 101334.3 101961.2 2.7
Actual Volume Flow ft3/hr 53940.9 56231.2 268801.2 269019.8 26.4
Mass Density lb/ft3 0.59 0.60 0.38 0.38 0.10
Compressibility 0.95 0.95 0.96 0.96 1.00
Thermal Conductivity Btu/hr-ft-F 0.02 0.02 0.02 0.02 0.05
Viscosity cP 0.01 0.01 0.01 0.01 0.01
Molecular Weight 64.6 66.1 75.2 75.3 13.9
Mass Flow lb/hr 436440.5 434585.2 835553.2 845179.1 10250.2 163879.4
Viscosity cP 1.13 1.10 1.34 1.32 0.20 0.05
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
104
Table 4.8.1o: H&MB for CDU – Streams 124, 132, 133, 218, 219 and 231
Stream Name 124 132 133 218 219 231
Stream Description DIESEL AGO P/A AGO P/A HVGO P/A HVGO P/A VTB
Vapour Fraction 0.00 0.00 0.00 0.00 0.00 0.00
Temperature F 368.3 606.8 281.4 427.9 368.7 512.0
Pressure psia 113.0 124.0 114.0 200.4 190.4 190.5
Mass Flow lb/hr 163879.4 23028.3 23028.3 278991.5 278991.5 196115.7
Watson K 11.540 11.498 11.498 11.502 11.502 11.435
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 163879.4 23028.3 23028.3 278991.5 278991.5 196115.7
Viscosity cP 0.14 0.12 0.56 0.44 0.60 7.40
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
105
Table 4.8.1p: H&MB for CDU – Streams 236, 237 and 239
Stream Name 236 237 239
Stream Description VTB VTB VTB
Phase Liquid Liquid Liquid
Vapour Fraction 0.00 0.00 0.00
Temperature F 512.6 512.6 512.6
Pressure psia 190.5 190.5 190.5
Mass Flow lb/hr 218067.4 2180.7 215886.8
Molar Enthalpy Btu/lbmole -391838 -391838 -391838
Watson K 11.435 11.435 11.435
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Mass Density lb/ft3
Compressibility
Viscosity cP
Molecular Weight
Mass Flow lb/hr 218067.4 2180.7 215886.8
Thermal Conductivity Btu/hr-ft-F 0.09 0.09 0.09
Viscosity cP 7.38 7.38 7.38
Actual Volume Flow ft3/hr 3991.7 39.9 3951.7
Molar Flow lbmol/hr
Mass Flow lb/hr
Heat Flow Btu/hr
Viscosity cP
106
Table 4.8.2a: System Salt Balance - Parallel Wash Water Injection (Normal Operation)
Stream Salt Salt
Stream Description # lb/hr bbl/d PTB lb/hr
Crude Oil Stream 10 927806.7 70001.0 100.0 291.7
Stripped Sour Water 70 149913.5 10286.0 0.0 0.0
River Water (NNF) - 0.0 0.0 0.0 0.0
Vacuum Steam Condensate 71 11858.2 857.0 0.0 0.0
Wash Water to 1st Stage 83 80885.8 5571.0 0.0 0.0
Crude+Wash Water at Inlet to 1st Stage 11 1008692.5 75573.0 92.6 291.7
Wash Water to 2nd Stage 84 80885.8 5571.0 0.0 0.0
Crude from 1st Stage Desalter 85 930432.1 70203.0 13.9 40.6
Crude+Wash Water at Inlet to 2nd Stage 86 1011317.9 75774.0 12.9 40.6
Desalted Crude Oil 14 936887.5 70668.0 1.0 2.8
Brine from 1st Stage Desalter 91 78260.4 5370.0 1121.9 251.0
Brine from 2nd Stage Desalter 90 74430.4 5107.0 177.6 37.8
Brine from Both Desalters 93 152690.8 10476.0 661.7 288.8
1st Stage Desalter Efficiency 85.00%

2nd Stage Desalter Efficiency 92.50%
Table 4.8.2b: System Salt Balance - Recycle Wash Water Injection (Counter-current Mode)
Stream Salt Salt
Stream Description # lb/hr bbl/d PTB lb/hr
Crude Oil Stream 10 927806.7 70001.0 100.0 291.7
Stripped Sour Water 70 0.0 0.0 0.0 0.0
River Water (NNF) - 74956.8 5143.0 0.0 0.0
Vacuum Steam Condensate 71 11858.2 857.0 0.0 0.0
Wash Water to 1st Stage 83 77734.1 5333.0 0.0 0.0
Crude+Wash Water at Inlet to 1st Stage 11 1005540.9 75334.0 92.9 291.7
Wash Water to 2nd Stage 84 86815.0 6000.0 0.0 0.0
Crude from 1st Stage Desalter 85 927806.7 70001.0 13.9 40.7
Crude+Wash Water at Inlet to 2nd Stage 86 1014621.7 76001.0 12.8 40.7
Desalted Crude Oil 14 936887.5 70668.0 1.0 2.8
Brine from 1st Stage Desalter 91 77734.1 5333.0 1129.6 251.0
Brine from 2nd Stage Desalter 90 77734.1 5333.0 165.9 37.8
Brine from Both Desalters 93 77734.1 5333.0 1295.6 288.8
1st Stage Desalter Efficiency 85.00%

2nd Stage Desalter Efficiency 92.50%
107
4.9. Equipment Design Consideration
The existing desalter is chosen to be the 1st stage desalter followed by the new desalter
to be the 2nd stage desalter. To maximize the phenolic water consumption in the
desalters, the following two configurations are considered for the injection of wash
water. For schematic view of the configurations, refer to the PFD in Appendix A.
4.9.1 Parallel Wash Water Injection to Both Desalters
In this configuration the wash water from Battery Limit is heated through brine wash
water and VTB wash water exchangers to reach the required operating temperature of
275oF. It is then split in equal amounts (7 vol% each) and injected to crude streams
just before the mixing valves for each desalter. This configuration will insure
maximum consumption of wash-water and hence maximum absorption of phenols
from the wash water into the crude oil.
4.9.2 Counter-Current (Recycle) Injection of Wash Water
In this configuration the wash water is heated through brine wash water and VTB
wash water exchangers to reach the required operating temperature of 275oF. The
heated wash water (7 vol%) is injected to the 2nd stage desalter and the salty water
effluent from the 2nd stage desalter is routed to the 1st stage as wash water for the 1st
stage desalter, from where it will exit as brine stream going out of the system. This
configuration is the most typical configuration used in refinery operations.
Under normal conditions, operation will be based on parallel injection mode. In case
of unavailability of wash water, the desalters will operate in recycle mode.
4.9.3 Heat Exchange for Increased Desalter Temperature
As per the results in Chapter 3, the optimum temperature for desalting of Maya crude
is 275ºF. The current operating temperature in the desalter is 256 ºF. Therefore, the
temperature of the desalter should be increased. There are a few ways to achieve a
higher temperature in the desalter, as following:
a) Increased pump-around flow rates; By increasing the pump-around flow rate

the temperature of the crude in the pre-heat train could increase as there would
be more heat available through the increased flow rate for given temperatures.
However, increasing the pump-around flow rates will affect the atmospheric
and vacuum column reboiler duties as the product specification should be met.
Also increased pump-around flow rates, would result in additional capacity on
the pumps, increased column diameter, and increased pipe sizes. These
changes are very costly and not considered as a suitable option by the refinery.
b) Relocation of existing and new desalters further downstream in the preheat

train where the temperatures are higher; By moving the desalters from the
existing location further downstream in the pre-heat train where there is higher
crude temperature it is possible to meet the required temperature in the
desalters. However this option requires major change in piping as well as plot
preparation for moving around existing equipment. This is too costly for the
108
refinery. Also the additional pressure drop through moving the exchangers
upstream of the desalters will decrease the pressure in the desalter and this is
not favorable as the desalter works under pressure control to prevent vapor
production in the desalter. If the existing pressure is to be kept under the new
proposed configuration, the charge pumps should be changed and increased in
size. Overall, this is an expensive option and therefore not considered by the
refinery.
c) Addition of utility exchangers to heat up crude or water; A third option could

be increasing the temperature of the wash water or crude stream to achieve the
optimum temperature of 275oF in the desalter. Adding utility exchanger to the
crude stream will cause pressure drop in the crude preheat train and hence
reduced pressure in the desalter. Also there is no space available in the plot for
fitting in an exchanger next to the existing desalter. Therefore this option is
not feasible. Addition of utility exchanger to the wash water section seems to
be the only feasible option as the above problems are not encountered. Also
due to increased wash water flow rates the existing wash water/brine heat
exchanger needs to be revamped. Therefore integration of wash water/brine
exchanger together with new utility exchanger is considered for revamp design
of the desalters.
4.9.4 Heat Integration
In the current operation, the wash water is first heated through the brine wash water
exchanger before mixing with crude oil at the entrance of 1st stage desalter V-1001.
The size of this exchanger is not sufficient for the new service as the wash water flow
rate will increase in new operation, The increased wash water rate means increased
brine flow rate and hence a bigger wash water/brine exchanger required as the
existing exchanger would produce too much pressure drop under new conditions and
also wouldn’t be able to meet the optimum heat transfer required.
Even when the wash water/brine exchanger is sized sufficiently for the new service it
wouldn’t be able deliver the required (optimum) temperature for the desalter on itself
because the wash water is being heated by the brine stream. In the new operation the
brine is coming out of the desalter at 275oF, so assuming a temperature approach of
35oF the wash water temperature coming out of the wash water/brine exchanger
would be at 240oF. To increase the temperature of wash water from 240oF to optimum
temperature of 275oF, a utility exchanger would be required.
The utility available in the refinery is High Pressure (HP) steam and therefore the new
exchanger would be designed as a steam pre-heater to increase the temperature of the
wash water sufficiently so that after wash water is mixed with the crude oil, the
optimum temperature in the desalter is met.
Although the mathematical model in chapter 3 shows that the optimum temperature of
the Maya crude is 275oF, operating temperature chosen for 1st stage desalter is 270oF
in order to have a margin to prevent vapor production in the desalter as the pressure of
the existing desalter cannot be increased. Also a reasonably lower temperature would
decrease the size of new utility exchanger required.
109
The results of HYSYS simulation for new temperatures are given in the H&MB
section in this chapter.
4.10. Environmental Considerations

The potential environmental concerns for revamp of the desalting unit are as follows:
• Carry-over of phenols into the salty water (brine) stream in an upset condition
• Carry-over of oil into the brine
These potential environmental concerns are considered when designing the new
desalters explained below.
4.10.1 Loss of Phenols into Brine
The wash water used will be phenolic stripped sour water from the refinery. One of
the primary reasons for using this stream as wash water in the desalter is that phenols
can be reabsorbed from the water into the oil and the brine going to the effluent
treatment will be free from phenols.
One of the concerns is that upset in the desalter operation can lead to phenols going
into the brine. This malfunction is controlled by shutting off the stripped sour water as
source of water to the desalters and switching the desalters operation from a parallel
wash water injection mode into countercurrent mode and use of river water instead of
the stripped sour water. This change will temporarily prevent rejection of phenols into
the brine stream until normal operation is restored. In addition to this control scheme,
the rag layer in the desalter is physically removed by draining and therefore any
emulsion material including phenol will be removed from the desalter. Vendors will
provide special piped systems to drain the rag layer. Disposal of this emulsion is also
an environmental issue.
4.10.2 Loss of Oil into Brine
In the event of upsets or while mud washing, there is a strong possibility of some loss
of oil into the brine. This could be prevented by level control of the interface layer.
AGAR control could potentially help with minimizing the potential risks with this
matter and hence prevent economical and environmental concerns. Therefore to
control the level in the desalter, AGAR control has been chosen.
110
Chapter 5: Conclusions
111
In this thesis work, a history of dehydration and desalting, its importance, the global
trend in crude quality and sources of wet crudes were reviewed. As a result it was
concluded that as the oil industry is maturing, the crude sources are becoming more
heavy and more salty and also the specifications by refiners are becoming more
stringent. Therefore, the need for desalting is increasing.
As crude oil emulsions make the desalting process more difficult and more energy
intensive, it is of utmost importance to understand the fundamentals of emulsion and
how to remove them in an oil system. This subject was studied in Chapter 2. The
information provided is based on experimental data from other works. A few common
techniques to break up emulsion or to facilitate separation in emulsion systems are
discussed in this chapter. The two most important factors in breaking up the emulsion
are heating and application of electrostatic field, which become the cornerstone of
later chapters of this thesis.
Different technologies were compared to understand the best available means for
desalting the heavy crude oils. To this end, two major desalter vendors, Cameron and
NATCO, were contacted for their technologies. Based on the experience history of the
two vendors it was concluded that Cameron provides a more suitable technology for
desalting the Maya crude. Details of the two technologies as well as pros and cons for
each technology are provided in sufficient detail in Chapter 2.
The effect of different variables on the desalting process were investigated and
studied in Chapter 3. As a result, it was found that the properties of Maya crude i.e.
density, viscosity and electrical conductivity seem to be strong functions of
temperature. This functionality needs to be further studied through statistical data
analysis and further lab data are needed to support the results. This could be subject of
future studies.
The functionality of Maya crude on temperature was used to mathematically model

and determine the optimum operating temperature for this type of crude. The
mathematical model prepared for this study is a linear model with profit and cost
terms and the optimum temperature is achieved at maximum profit or minimum cost
for the overall system. The results show that the objective function is satisfied at
operating temperature of 275 ºF. It is worthwhile to prepare a detailed mathematical
model to forecast the optimum desalting temperature of any crude in the market and
this could be the subject of future research work.
To achieve the optimum temperature in the desalter and in order to maximize the heat
integration in the system the available heat sources in the process were looked into.
The Cold Preheat Train and the Hot Preheat Train in the crude distillation unit were
both studied to find potential heat sources. Although it was found that heat sources are
available or could be made available by reconfiguration of equipment, it was not
justified to make these modifications due to system constraints such as pressure drop
and additional costs for the project. Therefore, to achieve the optimum temperature in
the desalter, it was proposed to increase the temperature of the wash water to
effectively elevate the temperature of the crude and wash water mix in the desalter to
that of the optimum temperature. This can be achieved by installing a new steam pre-
heater online for the wash water and reconfiguring the existing wash water heat
exchangers. The heat integration and equipment design considerations are discussed
112
in Chapter 4. As a future complimentary work for this study, Pinch Technology can
be used to analyze and quantify the heat integration in the cold and hot preheat trains
of the crude distillation unit and conclude whether the current heat integration is
optimum regardless of the constraints in the system.
As far as desalting specification, it was concluded that a second stage desalter is

needed for achieving the 1 PTB at the outlet of the desalter for the Maya crude in the
refinery. Two configurations were studied; Parallel injection of wash water to both
desalters and/or counter current operation of the two desalters. Based on the
availability of the wash water and the need for maximizing the wash water
consumption in the desalters, parallel injection of wash water to desalters was chosen
as normal operation. In the meantime the system is designed to operate in recycle or
counter current mode at wash water turn down.
In order to start the HYSYS simulation of the crude distillation unit, a full
characterization of Brent and Maya crudes, which form the blend for the refinery,
were accomplished in Chapter 4. The two sets of data i.e. lab data and calculated data
were utilized to predict the key properties of each crude and their blend. The results of
the crude characterization are integrated in the HYSYS model.
In the end a full HYSYS simulation was developed for the Crude Distillation Unit to
study different options for desalting of the Maya crude. A full heat and material
balance was developed from this simulation and the results are given in Chapter 4.
One of the shortcomings of this piece of work is that it is impossible to

mathematically model the behavior of the crude emulsion in the desalter and hence
very little can be predicted in this regard without having experimental data. Future
research in the field of emulsion, to better understand the properties of emulsion, will
definitely help in predicting the behavior of difficult crudes.
The other drawback in this study is that most of the information available on
electrostatic desalting, which is THE way for desalting of heavy crudes, is with
vendors and unless technology is bought from the vendors, details of the experimental
work will not be made available due to confidentiality of the information.
The other area that can be improved in the future is the simulation of the desalting
operation, as HSYSY is not a suitable tool to model electrolyte processes, and in order
to model the desalting operation in HYSYS, the associated water or formation water
with crude was assumed to be pure water. However, this assumption produces
inaccuracies in predicting the stream properties of the desalting operation. For future
research, it is recommended to either develop a suitable model to better predict the
desalting parameters or use a commercially available software that will have better
capabilities to model hydrocarbon electrolyte processes.
113
Appendix A: Process Flow Diagrams (PFDs)
114
PFD1 – The Cold Preheat Train for Crude Distillation Unit
115
PFD2 – 1st and 2nd Stage Desalters
116
PFD3 – The Hot Preheat Train for Crude Distillation Unit
117
References
1. Noïk C, Chen J, and C Dalmazzone. Electrostatic Demulsification on Crude Oil: A
State-of-the-Art Review. Society of Petroleum Engineers. SPE 103808. 2006.
2. Johnson, T. Oil Market Volatility. Council on Foreign Relations. December 10, 2007.
Available online: http://www.cfr.org/publication/15017/ [Accessed February 5, 2008.]
3. Crude Benchmark Analysis. Energy Economist. March 7, 2008. Available online:

http://www.platts.com/Oil/Resources/News%20Features/crudeanalysis/index.xml
[Accessed February 8, 2008]
4. U.S. Refinery Crude Oil Input Qualities. Energy Information Administration. January
29 2008. Available online: http://tonto.eia.doe.gov/dnav/pet/pet_pnp_crq_dcu_nus_m.htm
[Accessed February 19, 2008]
5. Sweet and Sour Crudes. Econbrowser. August 21, 2005. Available online:
http://www.econbrowser.com/archives/2005/08/sweet_and_sour.html [Accessed February
5, 2008]
6. World Crude Oil Prices. Energy Information Administration. February 13, 2008.
Available online: http://tonto.eia.doe.gov/dnav/pet/pet_pri_wco_k_w.htm [Accessed
February 20, 2008]
7. Annual Energy Outlook 2006 with Projections to 2030. Energy Information

Administration. Report #: DOE/EIA-0383(2006). Available online:
http://www.eia.doe.gov/oiaf/aeo/otheranalysis/aeo_2006analysispapers/tri.html [Accessed
February 5, 2008]
8. R.D. Kane, ASM Handbook
9. P.E. Krystow, Corrosion in Crude and Vacuum Units, Report submitted to NACE Task
Group T-8-5 on 2-14-69
10. Smith HV and KE Arnold. Petroleum Engineering Handbook. Chapter 19: Crude Oil
Emulsions. Society of Petroleum Engineers. 1987.
11. Khatib ZI, Faucher MS, and El Sellman. Field Evaluation of Disc-Stack Centrifuges
for Separating Oil/Water Emulsions on Offshore Platforms. Society of Petroleum
Engineers. SPE 30674. 1995.
12. Dhuldhoya N, Mileo M, Faucher M, and E Sellman. Dehydration of Heavy Crude Oil
Using Disc Stack Centrifuges. Society of Petroleum Engineers. SPE 49119. 1990.
13. Meikrantz DH, Macaluso LL, and MM St. George. Advances in Liquid/Liquid
Centrifuge Design Provide New Options for Petroleum Production. Society of Petroleum
Engineers. SPE 56709. 1999.
14. Bai ZS and HL Wang. Crude Oil Desalting Using Hydrocyclones. Chem Eng Res
Design. 85(A12): 1586-1590. 2007.
118
15. Egloff G, Nelson EF, Maxutov CD and C Wirth. Desalting Crude Oils. Oklahoma
City Meeting. SPE-938048-G. October 1937.
16. Filtration method efficiently desalts crude in commercial test. J Oil Gas. 91(20): 59-
60. 1993.
17. Chawla, M.L. Field Desalting of Wet Crude in Kuwait. Kuwait Oil Co. Society of
Petroleum Engineers, SPE 15711.
18. Mandal, K.K. Improve Desalter Control. Hydrocarbon Processing. April 2005.
19. Kokal, S. Crude Oil Emulsions: A State-of-the-Art Review. Society of Petroleum

Engineers, SPE 77497. 2005.
20. Manning, Dr. F.S., Thompson, Dr. R.E. Oilfield Processing, Volume 2: Crude Oil.
Pennwell Books, 1995.
21. Paragon Engineering Services. Desalting. IHRDC Production Facility Bookware

Series.
22. Buckley JS and Y Liu. Some mechanisms of crude oil/brine/solid interactions. J Petro
Sci Eng. 20: 155-160. 1998.
23. Shafizadeh A, McAteer G and J Sigmon. High Acid Crudes. New Orleans Meeting
Presentation, Crude Oil Quality Group. 2005.
24. Peters, MS, and KD Timmerhaus. Plant Design and Economics for Chemical
Engineers, Fourth Edition. McGraw-Hill Inc, 1991.
25. Arnold K, and M Stewart. Surface Production Operations, Volume 1 – Design of Oil
Handling Systems and Facilities. Elsevier, 1999.
26. Eddy HC. Discussion on Electrical Dehydration of Crude Oil. J Indust Eng Chem.
November, 1921.
27. ID-205 Interface Detectors. Agar Corporation. 2001. Available online:

http://www.agarcorp.com/detectors.html [Accessed March 2, 2008]
28. Anderson WA. Process Equipment Sizing and Selection. Department of Chemical
Engineering, University of Waterloo. 2004.
29. ADMIX. Sizing the ADMIXERTM Static Mixer and Sanitary Blender, Tech. Note
No.102. Manchester, NH, November 1998. Available online: http://www.admix.com
30. Mukherjee, R. Effectively Design Shell-and-Tube Heat Exchangers. Chemical

Engineering Progress. February 1998.
31. World Crude Oil Prices. Energy Information Administration. 2008. Available online:
http://tonto.eia.doe.gov/dnav/pet/pet_pri_wco_k_w.htm [Accessed February 14, 2008]
119
32. Escobedo, E.R.B., Garfias, F.J, Pruneda, E.F. Optimum Temperature in the
Electrostatic Desalting of Maya Crude Oil. March 15, 2005. Facultad de Ouimica,
Unversidad Nacional Autonoma de Mexico, Ciudad Unversitaria, Mexico.
33. Gibson, D. Oil Industry – Questions and Answers. Gibson Consulting. 2007.
Available online: http://www.gravmag.com/oil3.html#barrel [Accessed February 17,
2008]
34. Nelson, W.L. Petroleum Refinery Engineering, Fourth Edition. McGraw-Hill Book
Company Inc, 1958.
35. Arnold, K. and Stewart, M.: Surface Production Operations, Vol. 1; Design of Oil
Handling Systems and Facilities, Gulf Publishing Co., Houston (1986)
36. Becher, P.: Principles of Emulsion Technology, Reinhold Publishing Corp., New
York City (1955)
37. Blair, C.M.: Handling the Emulsion Problem in the Oil Fields, Magna Corp.,
Santa Fe Springs, CA (1971)
38. Sams, G.W. and Warren, W.: New Methods of Application of Electrostatic Fields,
Natco Group, New Orleans, Louisiana (2004)
39. Bailes, Philip J., Freestone, D. and Sams, G.W.: Pulsed DC Fields for Electrostatic
Coalescence of Water-in-Oil Emulsions, The Chemical Engineer (1997)
40. Pruneda, E.F., Escobedo, E.R.B., Vazquez, F.J.G.: Optimum Temperature in the
Electrostatic Desalting of Maya Crude Oil (2005)
41. Kim, Young H., Wasan, Darsh T.: Effect of Demulsifier Partitioning on the
Destabilization of Water-in-Oil Emulsions (1996)
42. Bhardwaj, A., Hartland, S.: Demulsification of Water-in-Oil Emulsions, J.

Dispersion Sci. Technol. (1993)
43. Pruneda, E. F., Escobedo E. R. B., Vazquez F. J. G.: Optimum Temperature in

the Electrostatic Desalting of Maya Crude Oil, J. Mex. Chem. Society (2005)
44. Lucas, N. Roy: Dehydration of Heavy Crudes by Electrical Means, Petrolite

Corp., Petreco Div., Houston, TX (1966)
45. Pearce, C.A.R.: The mechanism of the Resolution of Water-in-Oil Emulsions by

Electrical Treatment (1953)
46. AGAR Control, AGAR Corporation, http://www.agarcorp.com
47. Petreco Bilectric Dehydrators and Desalters, Cameron Petreco Process Systems
http://www.c-a-m.com/content/products/product_detail.cfm?pid=2630
120
48. Natco Dual Polarity Dehydrators and Desalters, Natco Oil Treating Technologies,
http://www.natcogroup.com/Content.asp?t=ProductPage&ProductID=100
121

Desalter PDF

Uploaded by

Copyright:

Available Formats

Desalter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Desalter PDF

Uploaded by

Copyright:

Available Formats

Modeling and Optimization

presented to the University of Waterloo

thesis requirement for the degree of

Master of Applied Science

Waterloo, Ontario, Canada, 2009

© Shahrokh Ilkhaani 2009

1. Salt causes corrosion in the equipment.

I am also grateful to fellow researchers and potentially life-long friends and

List of Figures .......................................................................................................................................viii

Chapter 1: Introduction to Crude Oil Desalting........................................................................................1

Chapter 2: Process Design Parameters ................................................................................................... 13

Chapter 3: Determination of Optimum Temperature for Desalting Operation....................................... 32

Chapter 5: Conclusions......................................................................................................................... 111

Appendix A: Process Flow Diagrams (PFDs) ...................................................................................... 114

References ............................................................................................................................................ 118

Figure 1.3.2 - Price Differential between Brent and Maya Crudes...........................................................5

Figure 1.4.1c - Water Coning Phenomenon .............................................................................................8

Figure 1.4.1d - Water Encroachment/ Early Water Breakthrough ...........................................................9

Figure 1.4.1e - Water Fingering Phenomenon..........................................................................................9

Figure 1.4.2 - An Example of a Casing Failure ...................................................................................... 10

Figure 2.3.7a - Effect of pH and Demulsifier Concentration on Emulsion Stability.............................. 23

Figure 2.3.7b - Effect of Brine and pH on Emulsion Stability ............................................................... 24

Figure 2.4.1 - Cameron Bilectric® Dehydrator/Desalter........................................................................ 25

Figure 2.4.2a - Temperature Requirement vs. API Gravity.................................................................... 26

Figure 2.4.2b - Throughput vs. API Gravity .......................................................................................... 26

Figure 2.5.1 - AC Electrostatic Coalescer .............................................................................................. 28

Figure 2.5.2 - Dual Polarity AC/DC Field.............................................................................................. 29

Figure 2.6b - Level Control in the Desalter Using AGAR System......................................................... 30

Figure 3.2.1 - Maya Density vs. Temperature ........................................................................................ 35

Figure 3.2.2 - Maya Viscosity vs. Temperature ..................................................................................... 36

Figure 3.2.3 - Maya Electrical Conductivity vs. Temperature................................................................ 37

Figure 3.4a - Costs and benefit trends .................................................................................................... 42

Figure 3.4b - Profit trend vs. Temperature ............................................................................................. 43

Figure D9.2 – Maya Characterization – CALII Cuts – Viscosity vs. Density........................................ 82

PFD2 – 1st and 2nd Stage Desalters ....................................................................................................... 116

Table 4.3.2: Feedstock properties for Crude Distillation Unit................................................................ 49

Table 4.4.1 - Calculated Weight for Brent Crude................................................................................... 63

Table 4.4.3a - Calculated Weight for Maya Crude................................................................................. 80

Table 4.8.1a: H&MB for CDU – Streams 2, 3, 4, 5, 6 and 7.................................................................. 91

Many commercial installations nowadays are employing chemically aided electrical

Figure 1.3.2 - Price Differential between Brent and Maya Crudes

1.4.1 Primary Causes

Figure 1.4.1c - Water Coning Phenomenon

Figure 1.4.1e - Water Fingering Phenomenon

1.4.2 Secondary Causes

• Inter-communication between tubing and casing strings.

1.4.3 Tertiary Causes

1.5.2 Scale Accumulation

1.5.3 Catalyst Activity

1. Investigate the effect of different variables on the desalting process.

An emulsion can be defined as a system consisting of a mixture of two immiscible

(1) Water being the dispersed phase.

(i) Oil-in-water (O/W) for oil droplets dispersed in water.

Further reference to emulsion in this research implies water-in-oil type emulsions,

2.2.1 Role of Emulsifying Agents

Water-in-oil emulsions contain complex mixtures of organic and inorganic materials.

2.2.2 Stability of Emulsions

Emulsions may be stabilized by the presence of a protective film around water

2.2.2.1 Type of emulsifying agent

Fw (out) = [Vs / Vs] . Fw (out) (3.3d)

CB = {[Fo (in) – Fo (in)] + [Fw (in) – Fw (in)]} .Cp (3.3.3)