Chimical Unit
Chimical Unit
Chimical Unit
A R T I C L E I N F O A B S T R A C T
Keywords: Ras Tanura Refinery desalination plants production has been historically limited for a number of
Multi-stage flash (MSF) factors. Major obstacles for low performance were heat transfer, plugging of spray nozzles,
Multi-effects desalination (MED) operating under severe scaling process, brine and distilled water level control, seawater velocity,
Scale
chemical dosing, and poor strainers performance.
Calcium carbonate (CaCO3)
Calcium sulphate (CaSO4)
Major challenges for the low performance in desalination plants were discussed for both Multi-
MED-TVC Stage Flash unit (MSF) and Multi-Effects Desalination unit (MED).
Brine circulation Based on the resulted outputs, discussion and analysis, it was concluded that thermal desali
nation process is sensitive process where many factors can lead to performance reduction.
However, thermal desalination process is easy to be maintained and restore its production
capacity.
The paper includes troubleshooting, analysis, actions taken, and the results. The conclusion
includes nine recommended tips to enhance the performance and restore the desalination units’
production.
1. Introduction
Arabian Gulf countries use desalinated water from thermal desalination units for human necessities and to generate high pressure
steam to generate power. Multi-Stage Flash (MSF) and Multi-Effect Desalination (MED) units are commonly used.
Multi-Stage Flash (MSF) and Multi-Effects Desalination (MED) units production rates are negatively affected over time by the high
salinity in the Arabian Gulf where it reaches 51,000 ppm in some locations. In Saudi Arabia, Saudi Aramco Ras Tanura Refinery (RTR)
has one MSF unit, commissioned in 1978, and four MED-TVC units, commissioned in 2002, where their production rates declined over
the years, because of scaling issues.
The purpose of this paper is to highlight and discuss the scaling factors, efficient chemical cleaning, scale and deposits control. The
paper includes troubleshooting, analysis, actions taken, and the results.
Ras Tanura Refinery (RTR) produces distilled water from seawater in an MSF, using 15 psig (1.03 bar) steam as the heat source with
a capacity of 300 GPM (1,635 m3/day). RTR MSF is a brine circulation type with 190 � F (87.8 � C) as top brine temperature (TBT). It has
* Corresponding author.
E-mail address: abuaha0e@aramco.com (H. Abu Al-Saud).
https://doi.org/10.1016/j.wri.2020.100124
Received 23 July 2019; Received in revised form 29 February 2020; Accepted 1 March 2020
Available online 9 March 2020
2212-3717/© 2020 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
2
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
Table 1
Scale samples analysis results.
Compounds Sample#1 (wt %) Sample#2 (wt %)
Aragonite-CaCO3 94 –
Calcite-CaCO3 3 –
Greigite-F3S4 2 –
Gypsum-CaSO4.2H2O 1 100
one brine heater and total of 15 stages in a series to produce the distilled water, where the vapor is created during flowing of the
seawater. The remaining hot brine is transferred freely from chamber to another where it is flashed there. The advantage of that is to
maintain the hot brine conditions outside and inside heating tubes in order to minimize scale formation on the tubes’ surface, which is
a major advantage over the original and simple concept of thermal evaporation (see Fig. 1).
Distilled water is also produced in four MED units using 150 psig (10.3 bar) steam with 525 GPM (2,865 m3/day) per unit, for a
total capacity of 1,200 GPM (6,540 m3/day). RTR MEDs are parallel with thermal vacuum compression. Each unit consists of four
effects and one distilled condenser. Each effect has a tube bank where the vapor formation takes place by suppling 150 psig (10.3 bar)
of steam in the first effect tubes, and spraying seawater into the top of the tube bank. The vapor generated from the first effect is
considered as the heating media for the second effect tubes. The vapor is condensed after spraying the seawater and sent to the
distillate condenser (see Fig. 2).
The units treat seawater with 51,000 ppm of Total Dissolved Solid (TDS) and minimizes solids carryover into the distillate for purity
of not more than 5 ppm of TDS.
Most thermal desalination units were not producing at their maximum capacity due to several contributing factors. The major
challenge is scale formation. This section discusses the major contributors to scale formation:
3
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
D. Injecting proper scale inhibitor: Scale inhibitor is used in MED units with a pH of 4.0. It is formulated for use in high “Top Brine Temperature”
desalination units, where the predominant scale potential is from calcium sulphate. It contains polymer and Phosphinosuccinic Oligomer (PSO),
wherein it functions as stabilizing and dispersing agents, and it can be used above the 266 � F (130 � C) top brine temperature. Also, this scale
inhibitor has a tracing material which allows special analyzers to measure it to monitor, control the dosage rate and optimize it. It was found that
calcium sulphate solubility significantly increased after injecting the acid base antiscalant.
E. Automation of antiscalant feed system: New antiscalant technology was implemented to eliminate scale in the tubes’ effect and to sustain higher
distillate production. The technology provides 24/7 online monitoring of plant operation key parameters to control both the water quality
analysis and anti-scalant dosage. The technology is represented by an online analyzer, control and dosage system. The online analyzer takes the
samples from three different locations in the unit: before and after the chemical injection point and analyze the water through tracing material in
the antiscalant chemical and then adjust the dosage rate to control the antiscalant concentration in the seawater inlet. Also, one sample is taken
from rejected brine to measure the rejected antiscalant through tracing material to control the Cycle of Concentration (COC).
F. Maintain brine concentration in the unit: The unit is designed for a certain make-up flow, which is considered the major contributing factor of
concentrated brine rejected by the unit. This ensures sufficient wetting of the tubes, and limits the risk of scale formation. Adjusting the flow on
the design decreases the risk of local overconcentration.
4
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
Table 2
Chemical solutions vs Scale removal efficiency.
Chemical solution Hydrochloric Acid (HCL) Sulfamic Acid (NH2SO3H) EDTA
Uses Descalant for metal surfaces from rust deposits and wide Descalant for cleaning of heat Descalant for removing
range of inorganic scales exchangers and pipe CaSO4
CaCO3 scale removal High High Low
efficiency
Mg(OH)2 scale removal High High Low
efficiency
CaSO3 scale removal Low Low High
efficiency
Seawater temperature and salinity play a major role in scale formation, where the salts tend to be crystallized at high temperature,
due to low solubility of calcium sulphate.
5
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
6
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
Debris always plugs some condenser tubes and can also plug spray nozzles, causing insufficient tube wetting or dry spots, which
result in accelerated scale formation. Seawater strainers have to be frequently back washed to minimize passing debris to the unit.
Checking spray nozzles during the shutdown opportunity and positioning them correctly, compared to the tube bundle, so that the
water is sprayed on the tubes.
The choice of chemicals (antiscalant and anti-foam) is very important to ensure long-term operation with few acids cleaning. The
dosing rate also needs to be sufficient for the local conditions. Using an anti-foam from the same chemical manufacturer as the anti-
scalant helps to have the most synergy and efficiency between the two products. The proper antiscalant has to be chosen based on the
seawater salinity and top brine temperature (TBT) of the unit.
7
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
It is caused usually caused by low performance of brine reject pump. Operating MED at high brine level leads to deposits on the
bottom side of the tubes.
Low tube velocity plays a major role in accelerating the formation of scale, since the carbonate molecules tend to deposit and
adhere on the tube surface at relatively low velocity. Low velocity is caused due to low performance of brine circulation pump.
Scaling is the most common type of deposition in desalination units, and is commonly associated with inverse solubility salts, such
as calcium carbonate (CaCO3), calcium sulphate (CaSO4), and magnesium hydroxide Mg(OH)2 found in seawater.
Two scale deposit samples collected from the desalination unit were submitted to the lab for chemical composition determination.
X-ray Powder Diffraction (XRD) technique was used to determine the composition of the submitted samples. The results revealed that
the both samples mainly consisted of calcium sulphate hydrate (gypsum-CaSO4.2H2O) and calcium carbonate scale (aragonite-CaCO3
a) Improper antiscalant dosing and monitoring caused frequent clogging in the brine heater tubes and stages. The dosing system was replaced by new
technology, where dosing is automatic.
8
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
b) Low seawater velocity (lower than the design velocity) inside the tubes as a result of the low performance of brine circulation pumps. Their
performance was restored after the overhaul and unplugging of the unit tubes, allowing the seawater to flow freely without obstructions.
c) Vacuum leaks were found — at different locations — which introduced air into the unit and disturbed unit parameters. A vacuum decay test was
applied for 2 h after fixing and tightening all leaks.
d) Production was found to be leaking into the brine pool, due to passing of the dump valves and troughs (distillate tray box) at different stages. All
dumps valves and damaged troughs were replaced and repaired to avoid further loss.
and calcite-CaCO3) with minor amounts of corrosion product (greigite-F3S4) and calcium sulphate hydrate (gypsum-CaSO4.2H2O).
(see Table 1).
4.1. Parallel feed multiple effect desalination with thermal vapor compression (MED-TVC)
Based on the scale found inside the units, the troubleshooting results showed and clarified the main constrains those limiting the
MED production, specifically:
As a results, the enhancement of scale reduction in RTR-MEDs were completed in two steps: inhabiting and descaling (see Figs. 3-5).
e) A large amount of material — mud, silt, biofouling, seaweed, and corrosion products was found inside the unit, which reduced the antiscalant
effectiveness. As mentioned previously, silt, mud, and biofouling were treated by fixing the strainers, cleaning the seawater intake canal, and
dosing with chlorine dioxide as a disinfecting agent.
9
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
f) Improper tube cleaning during shutdown, where the cleaning was limited to hydrojetting. A chemical cleaning procedure was developed to clean
the tubes using inhibited hydrochloric acid. This ensured the removal scale particles, because if these scales are not cleaned completely, the
remaining scale inside the tube surface may encourage scale deposition again. The chemical cleaning procedure consists of the following steps:
1) Preliminary water flushing: the inlet and outlet point of the tube side of the unit were connected to centrifugal pump (with capacity 0–4000
L/min) and mixing tank, flow reverse manifold, strainers, temperature, pressure gauges and flow meter and make a circulatory loop. Vents
were installed at high points and drains were installed at low points and connect these to the pump suction tank. The unit was filled with
distilled water. The circulation was started at 4.5 bar and continue flushing the system with water until the effluent water is free from loose and
soluble deposits.
2) Acid cleaning with inhibited Hydrochloric acid: the acid solution was prepared using water with chloride content<50 ppm. The inhibited
acid solution consists of:
Table 3
Inhibited acid concentration.
Chemicals Concentrations
A. Disinfection: The desalination units have a history of biological growth and fouling on the exchangers and effects. The disinfection
chemical type was replaced by chlorine dioxide (ClO2). ClO2 as a disinfectant had been evaluated over 10 months for treating
seawater feeding the units. The evaluation shows:
1) Microbiological analysis (GAP & Y/M) show improvement where the results decreased from 2.3 � 107 cells/ml to below 100
cells/ml after implementing ClO2, which will reduce the main corrosion mechanism.
2) Total of 16 seawater strainers (1,372 mm OD X 2,005 mm H, coated carbon steel) pressure back wash frequency was reduced
from 8 h to 35 h.
3) Improvement in heat transfer coefficient by 70% in exchangers.
B. Cleaning seawater intake: The seawater intake was full of sand and debris, which impact the MED tubes and heat transfer rate. To
clean the intake, a submergible pump was used to remove sand and debris that accumulated over the years at the bottom of the
intake canal.
C. Mechanical adjustment: All spray nozzle positions were corrected as per the design angle, and unplugged to get proper wetting of
the tube bundles, to avoid any hot spot on the tubes that might cause scale and tube failure.
It was found that before implementing the recommendations, the fouling resistance (R) increased by 0.0014 m2. K/W to reach
0.0035 m2. K/W within 365 days (3.9 � 10 6 m2. K/W per day). Fortunately, the fouling resistance (R) increasing rate was found
reduced to be 6.14 � 10 7 m2. K/W per day and reached 0.0025 m2. K/W within 730 days (see Figs. 6,7).
4.1.2. Descaling
Scale is difficult to be removed mechanically and needs a frequent chemical cleaning to maintain the plant efficiency. To evaluate
the best choice of the acid cleaner, the material composition of a scale deposit must be known to be cleaned, and the metallurgy used in
the MED plant, cost, safety concerns, and then suitable cleaning agent can be determined.
(see Table 2)According to the scale analysis, the majority of the scale is coming from calcium carbonate (CaCO3, more than 90%)
10
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
and minor calcium sulphate (CaSO4) and magnesium hydroxide (Mg(OH)2). The metallurgy in RTR MED is mainly: titanium tubes,
stainless steel 316 for distilled water pipes, fiber glass, and cemented pipes for brine water pipes and stainless steel SMO 254 for the
shells (see Figs. 8-10).
The descaling was completed by implementing two stages of chemical cleaning using 5% wt inhibited Sulfamic acid and 5% wt
inhibited Ethylene-diamine-tetra-acetic acid (EDTA). First, Sulfamic acid was circulated for 48 h and the samples were collected
hourly. The results were:
After that, the EDTA was circulated for 48 h after drain the acid and do water wash in the unit. The results were:
At the end of the second application of the process, a total of 354 kg of scale were removed, 234 kg of calcium sulphate, and 120 kg
of calcium carbonate. The following pictures show the difference between before and after the chemical cleaning processes:
The first sample (sample#[1] in the picture above) was taken in the beginning of the process. Sample [2] was taken after 12 h,
sample [3] was taken after 24 h and sample [4] was taken at the end of the cleaning process (after 30 h). The change in turbidity
between the samples reflects the effectiveness of the cleaning process inside the unit. The color of the entrained liquid, in the second
picture, after completing the process, shows that dirt and scale were removed from the unit (see Figs. 11-14).
After implementing all mentioned actions, MED production returned to the design capacity (see Figs. 15-18).
11
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
(see Table 3)RTR-MSF brine heater and stages#1–12 were found plugged by hard scale (calcium sulphate). As a result, the unit
production decreased. The scale formation and low production were investigated, and the results:
The acid was circulated at 4.5 bar and the temperature was adjusted to 60–72 � C. The circulation is continued until hydrochloric
acid concentration falls below 3.0% or if iron exceeds 10,000 mg/l. Then, the solution was drained (see Figs. 19-21).
3) Neutralization: solution of water, with chloride content<50 ppm, mixed with 0.5–1.0% w/w soda was circulated for 1 h to
neutralize the residual acid inside the system.
After implementing all mentioned actions, RTR-MSF production returned to the design capacity.
Based on the resulted outputs and the earlier discussion and analysis, it was concluded that desalination process is sensitive process
where many factors can lead to performance reduction. However, thermal desalination process is easy to be maintained and restore its
production capacity. Moreover, it was noticed that one failure led to multi failures in the unit, such as: unadjusted spray nozzles in MED
led to heavy scale on the tubes, tubes hot spots, high temperature and then low performance.
In addition, general actions are recommended to maintain thermal desalination unit production and minimize scale formation:
None.
12
H. Abu Al-Saud and J. Griman Water Resources and Industry 23 (2020) 100124
References
[1] H. Fath, M. Ismail, Operational performance of 2 X 5000 m3/day MSF sidi krir, desalination plant part I: enhancement of chemical leaning & brine heater
condensate process, in: Thermal Engineering &Desalination Technology Dept, King Abdul Aziz Univ., Jeddah, Saudi Arabia, 2003.
[2] H.T. El-Dessouky, H.M. Ettouney, Fundamentals of salt water desalination book, 2002. www.elsevier.com.
[3] K. Ba-Mardouf, O. Hamed, K. Al-Shail, F. Mondili, H. Al-Harthi, Impact of Dose Rate Reduction of Antiscalant Belgard Ev-2030 on the Performance of Alshoaiba
Phase I Desalination Plant, 2004. Technical Report No. APP 3808/20010 in May 2004.
[4] A.Y. Abdulgader, G. Mustafa, N. Nabil, K. Al- Hassani, M. Khafagy, Performance evaluation of A modified polymaleic acid as antiscalant in A reheat plant. Fourth
annual workshop on water conservation in the kingdom, Dhahran 2001 (2001), 23-25 April.
13