Desalter optimisation strategies:

Part 2
Analysis and optimisation strategies implemented for a two-stage desalter
processing light to medium API crude blends at a Southeast Asian refinery
Venkatesan Mani
Veolia Water Technologies and Solutions

A
s discussed in Part 1, published in PTQ Q4 2024, medium API) is pumped to the crude distillation unit (CDU).
crude oil desalting plays a pivotal role in refinery The feed undergoes preheating in a cold preheat train to
operations by removing salts and impurities that can a targeted desalter temperature of 138-145°C to reduce
wreak havoc on downstream equipment through corrosion, viscosity. The feed is evenly distributed to the two-stage
fouling, catalyst poisoning, and product quality degradation. electrostatic desalters in Trains A and B (see Figure 1).
Despite significant technological advancements in desalter At the crude unit battery limit, a wash water concentra-
design, optimising desalter performance remains an intricate tion of 2.5-3.0% is introduced, while 4.5-5.0% is fed before
challenge due to the complex interplay of multiple factors, the mix valve. The wash source is a blend of raw water and
including crude oil composition, operational conditions, stripped sour water. The crude and water are mixed in the
equipment design, and chemical treatment. This interde- first-stage mix valve at a pressure drop of ~1.2-1.4 bar,
pendence among factors often hinders achieving overall while the second-stage mix valve operates at a pressure
performance goals through isolated optimisation efforts. drop range of ~0.8-1.0 bar. The resulting emulsion down-
stream of the mix valve facilitates the removal of water-sol-
Optimisation of key parameters: Southeast Asian uble impurities from the crude.
refinery case study The water-in-oil emulsion is distributed between the elec-
Desalter operation overview tric grids at an applied field of 150 kilovolt-amps (KVA) with
From the crude tank farm, the feed crude blend (light to an operating voltage of 400-360 volts and an amperage of

Train A
Embreak* Level Level Desalter crude
primary control Transformer control Transformer
demulsifier
3 to 5 ppm

1st stage 2nd stage

Mix valve
Embreak*
1 to 3 ppm
Mix valve
2nd stage water to 1st stage
Crude
tank Brine H2O Ex.

Stripper SWS wash water to 2nd stage ProChem* pH modifier Effluent water Train A

ProChem* solids wetting agent

3 to 5 ppm Train B
Embreak* Level Level
primary control Transformer control Transformer
demulsifier
3 to 5 ppm

1st stage 2nd stage

ETP
Embreak* Mix valve
1 to 3 ppm
Mix valve
2nd stage water to 1st stage
Brine H2O Ex.
ProChem* pH modifier
Stripper SWS wash water to 2nd stage Effluent water Train B
ProChem* solids wetting agent
3 to 5 ppm

Figure 1 Desalter process flow diagram

www.digitalrefining.com February 2025 31

 Crude Stability Model Output
RELATIVE INSTABILITY INDEX (RX)
Parameter Value
Stable Unstable Critical Severe

Severity
RIX 1.70 –
<1.5 1.5 - 3.0 3.0 - 6.0 6.0 - 10.0
CPI 3.93 –
Stabiliser 2168
Demand <5 ppm* Stable emulsion, oily water & solids

Unit Issues
Potential
API Gravity 31.3 ˚ Desalter
Viscosity 11.0 cSt
Cold preheat, Hot preheat & Heater fouling
TAN 0.3 mg KOH/gm
Sulphur 1.8 % w/w
FOULING POTENTIAL INDEX (FPX)
YC7-Asphaltenes 2.1 % w/w
<1.0 1.0-2.0 2.0-3.0 3.0 - 6.0 6.0 - 10.0

Severity
YSaturates 51.9 % w/w
Low Moderate Medium Critical Severe
YAromatics 31.8 % w/w
YResins 10.6 % w/w
Conventional Fouling Unconventional Fouling
YC5-Asphaltenes 5.6 % w/w

Unit Issues
Potential
YCII 1.4 – Hot Preheat fouling
True RIX 1.7 –
Heater fouling
*Based on total charge rate

Figure 2 CrudePlus study: crude stability and emulsion and fouling predictions

80-120 amps for most crude blends. The design residence basic sediment and water (BS&W) as free water, with 8%
time for the desalter is 12 minutes for crude oil and 160 wash water and <12 ppm chloride in the wash water.
minutes for water. The crude oil inlet is designed for 20 PTB salt and 0.5
The dosage strategies followed split feed technologies.1 vol% BS&W. However, most of the crude blends processed
The proprietary emulsion breaker (Embreak) feeding rate have a salt concentration of <10 PTB. Chloride in the over-
will be 5-10 ppm, with the dosage split into two locations: head is controlled to <30 ppm without any caustic injection
one before the cold preheat and another at the mix valve into the desalted crude, ensuring sodium levels are main-
of each stage. The solids wetting agent feed rate will be tained below 1 ppm in the atmospheric residue.
3-5 ppm based on incoming crude solids (>60 ppm in raw The following section covers the basics of key parame-
crude). These chemistries help reduce interfacial film ten- ters and how each key parameter optimisation approach
sion, promoting oil and water separation. was followed, considering interdependent variables.
The desalted crude achieves the key performance indica-
tors (KPIs). Specifically, it achieves a crude outlet salt con- Crude oil characteristics and benchmarking study
centration of <0.5 per thousand barrels (PTB) and 0.2 vol% Any desalter optimisation strategy begins with under-
standing the characteristics
Feed crude characteristics study of the crude oil before moving
on to operational parameter
Results Ref method optimisation. A detailed crude
Feed crude API 29.6 ASTM D287-22
blend quality analysis was
Viscosity at 40C, cst 10.73 ASTM D7945-21a
conducted periodically based
Total chloride, ptb 8.0
on crude blend changes and
Inorganic chloride, ptb (extractable) 7.9 ASTM D6470
whenever KPIs were not in
Organic chloride, ppm (non-extractable) <1 ASTM D4929-19a
control for any short period.
Solids in crude, ptb 104
The crude characterisation
Filterable solids, ptb (>0.42 micron) 104 ASTM D4807-88
testing focused on desalter
Iron, ppm 4.3 XRF*
Crude compatibility
impact variables, such as
RIX – Relative instability index 1.7 CrudePlus* crude oil API/density, vis-
CPI – Crude precipitation index 3.9 CrudePlus* cosity, salt content, water
FPX – Fouling potential index 2.02 CrudePlus* content, metal content, con-
Crude emulsification study ductivity, compatibility, foul-
Emulsification tendency High PED ing potential, and filterable
Emulsification precursor Iron/filterable solids PED solids.
Emulsion resolved with solids wetting agent Synergistic performance PED An emulsion potential study
was also conducted using
Table 1 a portable electric desalter

32 February 2025 www.digitalrefining.com

 Emulsion breakers screening (8% wash water) Emulsion breakers screening (5% wash water)
100 100
Demulsification efficiency (DE), %

Demulsification efficiency (DE), %

90 Embreak 2W157i 5 ppm 90 Embreak 2W157i 5 ppm
Embreak 2W157i 9 ppm Embreak 2W157i 9 ppm
80 80
Embreak 2W197i 9 ppm + Embreak 2W197i 9 ppm +
70 Embreak 2163 3.5 ppm 70 Embreak 2163 3.5 ppm

60 60
50 50
40 40
30 30
20 20
10 10
0 0
BLANK Embreak 2W157i BLANK Embreak 2W157i

Figure 3 Emulsion stability test study using portable electric desalter

(PED) to validate process operating conditions and opti- standard viscosity temperature charts for liquid petroleum
mise chemical dosages for the emulsion breaker and solids products (D341 Chart VII).
wetting agents. The impacts of wash water quality, pH, and It is important to remember that emulsion viscosity
chlorides were also assessed. Additionally, desalter emul- increases exponentially with lower temperatures, particu-
sion layer samples were analysed when there was growth larly in oil-water emulsions with varying water-to-oil ratios,
in the desalter emulsion layer. The brief test results are as referenced in emulsion journals. Typically, emulsion vis-
summarised in Table 1. cosity >100 centipoise will lead to a stable emulsion layer
Crude samples were subjected to compatibility and foul- in the desalter. Increasing the emulsion breaker dosage can
ing assessments using field-proven technologies, such as help resolve short-term issues. However, with intense sol-
the proprietary CrudePlus tools (see Figure 2). ids and destabilised asphaltenes stabilisation conditions,
The emulsification tendency of the crude was evaluated increasing temperature will be the best approach for long-
in the PED. The crude was confirmed to have high emulsion term sustainable operations. Hence, without temperature
potential based on poor water separation with no chemi- optimisation based on crude viscosity, optimisation of other
cal treatment. With the right emulsion breaker dosage, the parameters will not help achieve consistent KPIs.
emulsion broke down, and water separation was observed Another limitation of high desalter temperature operation
with treatments using solids wetting agents. These results will be an increase in water solubility in the desalted crude
were used for desalter optimisation, as shown in Figure 3. oil. To retain the BS&W <0.3 vol% in the desalted crude
and keep the viscosity in control, the desired desalter oper-
Desalter temperature ating temperature target was >138-145°C, where Figure 4
Desalter temperature = f {crude oil viscosity and density, shows the results achieved after temperature optimisation.
water solubility in the crude} Predominantly, refineries often report total BS&W <0.05
To achieve maximum desalting efficacy, a widely followed vol% in desalted crude, even at high desalting tempera-
approach is to increase the desalter temperature, which can tures. The low BS&W is subject to debate based on the fact
enhance salt removal efficiency. However, this method typ- that 60% is due to sampling errors (such as hot sampling of
ically adopted and pushed the desalter’s maximum oper- desalted crude without proper cooling), 30% is due to the
ating temperature limit (typically 155°C) or the limit based lack of water solubility at the desalter operating tempera-
on the transformer bushing design temperature and some- ture, and 10% is due to other analytical factors. Therefore,
times corrosion risk limits. While higher temperatures can it is important to validate the BS&W in desalted crude with
reduce crude oil viscosity and aid water separation, beyond the water yield in the overhead based on different sources
certain temperature increases, water yield diminishes the rather than solely relying on BS&W analytical results, as
returns in separation efficiency. Hence, increasing to the shown in the following example:
maximum desalter temperature is not a cure-all and comes BS&W in desalted crude = Water yield in CDU overhead
with limitations and risks. boot – CDU column stripping steam – CDU side cuts
Temperature optimisation often starts based on the rec- steam – Overhead wash water before coolers.
ommended operating viscosity of the desalter. Even though
there are no well-defined design limits for desalter operat- Mix valve DP
ing viscosity, the desalter is targeted to operate at a viscos- Mix valve ΔP = f {crude characteristics, temperature,
ity <2 cSt based on best practices guidelines for effective wash water, electric grid, interface level}
desalting. Hence, with the analysis of crude blend viscosity The function of the mix valve extends beyond simply mix-
and API, the temperature required to achieve the desired ing crude oil and water. It is designed to control the water
operating viscosity can be estimated from the ASTM population, droplet size, and distribution within the crude

www.digitalrefining.com February 2025 33

 emulsions are less stable. Efficient desalter designs require
160 0.80 less wash water. To achieve stringent salt content targets,
such as salt <0.5 PTB in desalted crudes, a higher percent-
155 0.70
age of wash water of up to 10% is needed. Deploying <3

% BS & W in desalter crude

Desalter temperature (˚C)

150
Temp > 150 ˚C
0.60 vol% wash water drastically affects desalter performance.
145
BS & W > 0.45 vol%
0.50 It is recommended to split the wash water feed rate at
Temp < 145˚C
BS & W < 0.3 vol% the crude battery limit, specifically before the cold preheat
140 0.40
exchanger (to prevent solids fouling in the cold preheat
135 0.30 and increase the contact time between crude and water to
130 0.20 effectively remove upstream chemical-based crude impu-
Post mix valve
optimisation rities) and upstream of the mix valve. Stripped sour water,
125 Desalter temperature 0.10
Water yield from crude BS & W free from chloride (<10 ppm), dissolved oxygen <7 ppb, no
120 0.00 dissolved carbon dioxide, and pH 7-8, is best for desalting
operations.
Figure 4 Water carryover in desalted crude vs temperature While higher wash water percentages improve desalting,
the availability of wash water sources from plant overhead
oil, effectively washing salts from the crude. Globally, stripped sour water is also limited. An increase in desalter
globe valves are the most common mix valves deployed for wash water increases effluent generation and treatment
desalters due to their precise control, versatility, durability, costs in the wastewater treatment plant. In crude desalting,
reliability, and ease of maintenance. Mix valves typically it is important to remember that salt removal from 10 PTB
accommodate pressure drops ranging from 10 to 50 psi. to <2 PTB will be comparatively simpler than achieving a
For light to medium API crude, the recommended pressure reduction of salt from 2 PTB to <0.5 PTB.
drop operation is 10-35 psi (0.8-2.4 bar). Too low produces At low salt levels, specifically <2 PTB with 8% wash
more coarse droplet sizes, leading to poor desalting, and operation conditions, the conductivity of water droplets
too high produces fine droplet sizes that can make the dispersed in crude will be low. This leads to a poor force
emulsion difficult to break. of attraction between water droplets, compromising water
The droplet size and distribution are not limited to the coalescence efficacy. Hence, maintaining a minimum con-
pressure drop in the mix valve. They depend on the elec- ductivity of 1,000 µS/cm in the desalter outlet brine water
tric field, temperature, wash water, and crude density.2 In is crucial to ensure a healthy force of attraction between
refinery day-to-day desalter optimisation, simple baseline the particles.
monitoring of salt outlet vs mix valve DP will help establish When the desalted brine conductivity ranges from 2,000
the safe operating range for the mix valve DP. to 5,000 µS/cm, water carryover in the desalted crude
The best practices followed for operating the two-stage ceases, and consistent salt removal efficiency is achieved.
desalting mix valve include: Hence, it is important to monitor the conductivity of the
u Operate the first stage of the desalter at a high-pressure desalter brine water from each stage of the desalter for a
drop to achieve intense mixing, maximising the removal of holistic desalter optimisation approach. With all the previ-
salts during the desalting process. ously noted best practice adoptions, deploying low chloride
v Operate the second stage at a relatively lower DP to pro- wash water (<10 ppm) with a wash water rate of 4.5-5%
mote dehydration, ensuring that all the salt-washed water at the mix valve led to achieving the salt outlet KPI (see
carried from the first-stage crude will be fully dehydrated. Figure 6).
Figure 5 illustrates how to achieve the desired crude The desalter brine pH is equally important compared
salt outlet range with a sharper interface layer at high to the wash water pH for overall desalter performance.
voltage and low amperage conditions in the electric grid. Operating the brine pH slightly towards the neutral to
This ensures there is no fine emulsion formation and water slightly acidic side (6-7) is recommended for effective
carryover in the desalted crude. Best practices such as desalting operation, as it helps minimise the corrosion envi-
mix valve calibration, check valve opening, and actuator ronment. High pH in the wash water and desalted brine can
response were adopted annually for reliable mix valve deprotonate emulsifying agents, enhancing their ability to
operations. For high solids crudes processing timeframes, stabilise water droplets in oil.
the mix valve was cleaned every two years to remove any For crudes with chemical impurities like phosphate
deposits around the valve positions. esters, amine chlorides, calcium naphthenates, or sodium
naphthenates, it is recommended to operate the desalter
Wash water brine pH slightly acidic (5-6) to wash away these impurities
Wash water = f {wash water %, wash water quality, salt in the desalter brine. It is important to inform the down-
& BS&W, electric grid, brine pH, chemicals} stream wastewater treatment plant to monitor chemical
A common industry practice is to deploy 3-10% wash water oxygen demand (COD) control, as it can increase with low
relative to the volume of crude oil.3 Typically, medium to pH operations. To overcome this issue, a pH modifier with
heavy crude oils require more wash water (6-10%) due to a corrosion inhibitor, such as Prochem or Predator, was
higher viscosity and stable emulsions, while lighter crudes deployed to control brine pH.
(4-6%) need less wash water as they are less viscous and In case service water or raw water is used as wash water

34 February 2025 www.digitalrefining.com

 1.40 3.00 3.0 7.0%

1.20 2.50 6.0%

2.5

1.00
Mix valve DP (bar)

5.0%

Wash water (%)

Delta b/w first &
second stage 0.2 bar 2.00 2.0

Salt out (ptb)

Salt (ptb)
0.80 4.0%
1.50 1.5
Post adopted strategy of Consistent 5 to 4.5 vol% of wash water
0.60 1st stage mix valve high DP (Desalting focus) into mix is key for effective desalting, 3.0%
2nd stage mix valve low DP (Dehydration focus) helped to achieve salt <0.5 ptb
1.00 1.0
0.40 2.0%

0.20 0.50 0.5 1.0%

0.00 0.00 0.0 0.0%

Train A - 1st stage Train A - 2nd stage Train A Salt out, Ptb Train A - Salt out, Ptb Train A - 1st stage Train A - 2nd stage
Mix valve DP, bar Mix valve DP, bar Wash water % Wash water %

Figure 5 Mix valve strategies for two-stage desalter Figure 6 Wash water per cent vs salt outlet

during refinery CDU start-up or due to stripped sour water All desalter designers commonly provide three to five
challenges (which is not recommended for long-term use), interface layer sample points, with 6in between each sam-
high levels of dissolved CO₂ and O₂ in the raw water can ple point. Typically, emulsion layer growth of more than 6in
acidify the desalter brine. Sulphur-rich crudes produce in the desalter vessel indicates strong emulsion formation,
more acidic byproducts, and higher desalter temperatures which significantly compromises the crude oil and water
accelerate oxidation reactions, increasing the acidity of the residence time in the desalter. This affects the efficacy of
desalter brine water. The desalter experienced a pH drop the desalting process and can create uncertainty, leading to
of 1.5-3 units while using raw water during high sulphur slugs of salt and water entering the desalted crude under
crude processing periods. The pH modifier Prochem was the event of small disturbances, causing fouling in down-
used to maintain the brine pH at 6-7 and sustain desalter stream heat exchangers.
performance. The best operating guidelines for interface level adjust-
ment for two-stage desalters are:
Interface level u Target the first-stage desalter operation at a higher
Interface level = f {try-line emulsion thickness, residence interface level (60-70%). It will increase the water phase
time, chemical dosage, mix valve, solids in crude, mud residence, helping to control oil carryover in the brine.
wash} v Target the second-stage desalter operation at a lower
The term ‘interface level’ in a desalter specifically refers interface level (40-50%). It will increase the crude resi-
to the boundary between the water and emulsion/oil lay- dence time, helping to prevent water carryover in the crude.
ers. Interface level and residence time are interdepend- Upon adopting these shared operating guidelines, the
ent parameters. A rising interface level often indicates salt outlet was in control, as the oil in brine remained <100
the growth of the emulsion layer. Hence, monitoring the ppm (within the design limit) 100% of the time.
interface level is critical, and it helps achieve the desired The presence of solids, such as clay particles, and heavy
residence time for both oil and water phases. Regular mon- organic compounds, like asphaltenes, can stabilise emul-
itoring can help identify trends and patterns in emulsion sions, creating a complex, heterogeneous layer that results
formation. An increased pressure drop across the desalter in challenges in interface level measurement. From the
can indicate the presence of a thick emulsion layer, causing literature and lab PED experiments, it was learned that
resistance to flow. Also, high chemical consumption often solids-stabilised emulsions have relatively high viscosities
correlates to a lack of proper interface monitoring. and densities compared to typical oil-water emulsions
Combining interface level measurements with other oper- without solids. This heterogeneity can confuse measure-
ational parameters like pressure drop and chemical usage ment devices. Hence, preventing solids stabilisation at the
is recommended to get a comprehensive view of emulsion desalter interface layer is crucial, as it poses challenges in
behaviour. Advanced capacitance and gamma densitome- interface level measurements. Controlling interface emul-
ters can provide detailed profiles of the layers within the sion layer solids to <1,000 ppm is critical to prevent solids
desalter. Therefore, conducting periodic visual inspections pickering stabilisation.
through try-line/try-cock sampling points to validate the As shown in Figure 7, a real-time study on desalter emul-
accuracy of interface measurement (as shared in detail in sion layer solids stabilisation helped to understand the sol-
the troubleshooting section) is crucial. ids distribution and growth of the emulsion layer in both
Typically, desalter try-line sample visual validation is the first and second stages. During this solids stabilisation,
the first step in desalter troubleshooting. Any deviations the interface level measurement showed 10-15% devia-
in measured vs observed levels in try-lines should be tions from the actual desalter level observations in the try-
addressed as part of the troubleshooting approach dis- line sample points. Figure 7 is a graphical representation
cussed in Part 1 of this article. of where samples were taken and the solids ppm levels at

36 February 2025 www.digitalrefining.com

 Train A (First stage) Train A (Second stage)
Oil collection header
Oil phase

5
9138 ppm Distributor
11986 ppm 5
4 4
3 Interface 3
685 ppm
2 2 1289 ppm
1 1
Water phase

Crude in 104 ppm

Train B (First stage) Train A (Second stage)

Oil collection header
Oil phase

5 2952 ppm Distributor

4 16680 ppm 5
3 4 1396 ppm
Interface 3
2
1 2 4318 ppm
Water phase 1

Crude in 104 ppm

Sample No. Description Filterable solids, ppm Ca, ppm Fe, ppm Ni, ppm S, wt% V, ppm
1 Feed crude 104 3.1 4.26 10.88 2.451 31.93

Desalter
Train A

2 First stage 11986 50.46 1590.1 14.43 1.945 26.92

3 First stage 9138 37.79 1151.3 13.89 2.077 28.92

4 Second stage 685 8.6 52.8 10.54 2.29 29.75

5 Second stage 1289 9.4 97.4 10.88 2.33 29.84

Desalter
Train B

6 First stage 16680 336.6 8090.3 41.54 2.163 58.85

7 First stage 2952 11.08 361.6 10.62 2.157 27.49

8 Second stage 4318 97.85 3413 49.3 2.70 15.85

9 Second stage 1396 13.84 450 14.88 2.276 30.04

Figure 7 Solids stabilisation influence on interface layer

the interface layer. The bottom picture is a detailed compo- • pH modifiers: These are selected and applied in wash
sition analysis of the emulsion layer to explain the nature water based on the desalter brine pH ~6-7. This will also
of solids stabilisation. Hence, monitoring the interface level help prevent corrosion from low pH and improve the effi-
during every crude blend change or crude tank changeover ciency of salt removal.
is critical for holistic desalter optimisation. The strategies adopted for chemical optimisation include:
• Selection and dosage: Figure 8 shows the PED study
Desalter performance chemicals results. Regularly analysing crude oil quality and emulsion
Desalter performance chemical = f{try-line emulsion studies using a PED helps study different chemistries and
thickness/interface level, electric grid, mix valve, temper- downselect the best demulsification agents. The right dos-
ature, wash water quality, solids in crude} age rates of 5-10 ppm are applied to achieve 100% oil and
A comprehensive chemical treatment programme is crucial water separation within the desalter residence time.
for optimising crude oil desalting processes in refineries. • Monitoring and adjustment: Continuous monitoring of
This programme typically includes the use of demulsifiers, desalter performance and adjusting chemical dosages as
wetting agents, and pH modifiers as follows: needed to maintain the interface emulsion layer thickness
• Demulsifiers: It is recommended to evaluate and select below 6in.
the Embreak chemistry based on the emulsion separation • Troubleshooting: Time-to-time optimisation of pH
study with a PED to mimic real-time desalter functionality modifiers to control brine pH plays a critical role in trou-
• Solids wetting agents: These are crucial for crude con- bleshooting. Measuring solids at the interface and main-
taining more than 60 ppm filterable solids taining levels below 100 ppm are necessary to prevent

www.digitalrefining.com February 2025 37

 PED study with 8% wash water @ 3 KVA PED study with 5% wash water @ 3 KVA
8.0 5.0
Blank Blank
7.0 Embreak 2W157i (9) Embreak 2W157i (3)
Water separation from crude oil

Water separation from crude oil

Embreak 2W157i (9) 4.0 Embreak 2W157i (9)
6.0 + Embreak 2163 (3.5) Embreak 2W157i (9)
+ Embreak 2163 (3.5)
5.0
3.0

4.0

2.0
3.0

2.0
1.0
1.0

0.0 0.0
0 1 2 4 6 8 16 32 0 1 2 4 6 8 16 32
Residence time (minutes) Residence time (minutes)

Figure 8 PED study with emulsion breaker, wetting agent, and wash water per cent change

emulsion layer growth. Mostly, split dosage of demulsifi- Asian refinery, the applied power from the transformer is
ers into the battery limit and at the mix valve is of prime 150 KVA with a designed applied voltage of 400 volts and
importance for desalter optimisation. a frequency of 50 Hz. The surface area under the grid is
Figure 8 explains how the emulsion separation study 60 m². Hence, the applied electric field will be 2.5 KVA/m².
helps to understand the emulsion’s stability without chem- Since the applied electric field is within the recommended
icals. The emulsion stability reduces with chemical optimi- range of best practice windows, with brine conductivity
sation, with an increase in 5-8% wash water. An increase (water droplets in the crude) in the range of 2,000-5,000
in dosage with solids wetter helped resolve the emulsion micro Siemens (discussed in detail in the wash water sec-
100% within the given residence time of the desalter. tion), good desalting efficacy could be achieved for light to
medium API crude blends operation periods without many
Electric grid challenges. Based on monitoring, the desalter could oper-
Electric grid = f {crude conductivity, brine conductivity, ate at the recommended voltage of 400-380 volts and an
interface level, water solubility in crude, temperature} amperage of 80-100 amps.
The electric grid is responsible for inducing dipole moments For more than two years, the desalter targeted perfor-
in water droplets dispersed in the crude oil, promoting the mance has been consistently achieved and sustained, with
coalescence of small water droplets into larger droplets, >90% performance indicators. Specifically, the desalted
which can then settle out of the oil due to gravity. The elec- crude salt has consistently remained <0.5 PTB, the over-
tric grid acts as the heart of electrostatic desalter operation. head chloride levels have been limited to <30 ppm without
High voltage is necessary to create a strong electrostatic caustic dosage in the desalted crude, and controlled sodium
field. The voltage should be maintained at the design spec- levels in the atmospheric residue have been controlled to
ifications to ensure effective coalescence, and monitoring <1 ppm to prevent downstream ARHDT catalyst poisoning.
amperage is equally important. Excessive amperage can To recap briefly, following a systematic optimisation
indicate issues such as short-circuiting, which can damage strategy is crucial. More importantly, focusing on day-to-
the grid and reduce its effectiveness. day monitoring, validating the desalter performance during
Also, high operating temperatures can affect the perfor- each blend change and crude tank farm change operation,
mance of the electric grid and the dielectric properties of the and slop oil management will lead to proactive optimisation
crude oil. Hence, optimal temperature is crucial to ensure the towards operational and chemical factors. Also, do not for-
grid operates within safe limits. In the desalter operation, get to conduct periodic data analysis with basic statistical
the emulsion produced by the mix valve is distributed via or desalter benchmarking tools, which will be key to identi-
the crude distributor between the electric grids. Between fying the influencing parameters.
70% and 90% of primary water coalescence occurs within
a few seconds under the applied electric field. Secondary Conclusion
coalescence occurs beyond the electric grid, where larger The cost of poor desalting is significantly high. From the
coalesced water droplets continue to settle due to gravity, desalter survey, it is evident that crude blend quality varia-
contributing significantly to overall water removal. Hence, tions drive the day-to-day desalter key parameter optimisa-
emulsion growth control, specifically interface monitoring, tion using basic data analysis tools. The brief overview and
is critical to control desalter upsets. key performance factor optimisation strategies followed in
The applied electric field for light to medium API crudes this real-time refinery case study exemplify the critical role
will be ~1-3 KVA/m². For the desalter in the Southeast of strategic parameter management in refining operations.

38 February 2025 www.digitalrefining.com

 Acknowledgment
Adopting a holistic approach to operational parameter
The author would like to express his gratitude to the Veolia team, includ-
optimisation, integrated with the best chemical treatment
ing Mike Dion, Sumalya Nag, Nimesh Kumar Patel, Martin Willis, Ching
programme, enabled the refinery to consistently meet
Yaw Lee, and Piravin Mageswaran, for their valuable contributions.
desalter performance indicators. This ensured sustained
Special thanks go to the Southeast Asian refinery CDU Process Manager
operational excellence and profitability in a dynamic refin-
and Specialist Distillation team, Saiful Dzulfitri, Azamuddin Jameran, Nor
ery market environment. The key takeaways from this case
Atifah Binti, and Chua Wee Kheng, for their intense support and motiva-
study include: tion during the on-field troubleshooting and optimisation activities.
• Understand the interdependencies among factors like
crude quality, operating conditions, equipment design, and References
chemical treatment in desalter optimisation. 1 Dion M, Desalting opportunity crude, NPRA Spring National Meeting,
• Establish safe operating boundaries for critical parame- March 1995.
ters through data analysis, troubleshooting discrepancies, 2 Desalter Training Manual, Veolia Water & Process Technologies.
and periodic calibrations. 3 Gutzeit J, Controlling unit overhead corrosion: rules of thumb for bet-
• Optimise key parameters synergistically, considering their ter crude desalting, Paper no. 07567, Corrosion 2007.
interdependencies (for example, temperature for viscosity 4 Shooshtari M, Karimi M, Panahi M, Modelling of an industrial mixing
and water solubility, mix valve DP for droplet size and dis- valve and electrostatic coalescer for crude oil dehydration and
tribution, wash water for conductivity and coalescing force, desalination, Separation Science & Technology 2023, Vol 58, No.7,
interface level for residence time and emulsion control). pp.1,306-1,318.
• Implement a comprehensive chemical treatment pro-
gramme, including demulsifier selection, wetting agents, and Venkatesan Mani is Senior Product Manager and Application Expert for
pH modifiers, based on emulsion studies and brine chemistry. Veolia Water Technologies and Solutions for Southeast Asia and Oceania
• Continuously monitor performance, validate results dur- region, responsible for speciality chemical solution design and application
ing crude changes, and make proactive adjustments to expertise in refinery and petrochemical process technologies. Mani holds
operational and chemical factors. a BTech in chemical engineering from Anna University, Chennai, India.

• Leverage basic statistical tools and benchmarking for iden-

tifying and addressing influencing parameters periodically. LINKS
• By following a systematic optimisation approach and
integrating best practices, refineries can overcome the More articles from the following categories:
challenges of crude oil desalting, minimise the associated Corrosion and Fouling Control
costs, and achieve sustained operational excellence, even Crude and Vacuum Units
in dynamic market conditions.

www.digitalrefining.com February 2025 39