Modeling Simulation Part-II-Thermal...considered in the exergy analysis, the brine heater was in most cases responsible for the ... TBT of 90oC, the brine heater exergy losses were

MODELING AND SIMULATION OF MULTISTAGE FLASH DISTILLATION

PROCESS - PART II1

Osman A. Hamed, Khalid Ba-Mardouf, Hamed Al-Washmi, Khalid Al-Shail, Hylail Abdalla and Ali Al-Wadie

Research and Development Center

Saline Water Conversion Corporation PO Box 8328, Al-Jubail 31951, KSA

E-mail : [email protected]

SUMMARY The first part of the Project was completed on 17th October, 1999. A commercial

computer program was acquired and used for the simulation of the MSF process.

Operation data were collected from the seven selected MSF distillers representing Al-

Jubail Phase-II, Al-Khobar-II, Al-Khafji and Jeddah Phase-II, III & IV desalination

plants covering a one year period. The results of the study were published in Technical

Report No. TR 3808/APP 97002. Summary of the study results are shown in Appendix-

1. The comparative thermal analysis of the MSF distillers revealed that after more

than 16 years of continuous operation, their performance ratios were in most cases

higher than the design values. Detailed second law analysis was also carried out for

each distiller to determine the distribution of overall exergy losses among the various

subsystems of the MSF distiller and identify locations where losses of useful exergy

occur.

In the second part of the project, the thermal performance of SWCC large capacity

MSF desalination plants that were not included in the scope of work of the first part of

the project, which are : Yanbu, Shugaig, Shoibah, Al-Jubail I and Al-Khobar-III was

analyzed and simulated. Monitoring the thermal performance of seven MSF distillers

representing SWCC MSF plants on the East and West Coasts revealed that the

distillers instead of being deteriorated due to ageing, the production capacities and

performance ratios are still maintained within or, in most cases, higher than the design

values. This was attributed to the main developments which were introduced to SWCC

MSF plants during the last three decades which included the successful and economic

1 Issued as Technical Report No. TR 3808/97002-II in March 2004.

implementation of scale control techniques and the use of appropriate corrosion

resistant materials. Plants which were built more than twenty years back are still

operating with high availability and load factors which approaches or even sometimes

exceeds the original design figures. Such excellent situation provided the opportunity

to investigate the possibility of increasing water production from SWCC existing MSF

plants, further optimization of antiscalant dose rate and establish an unified ball

cleaning procedure.

Second law simulation studies revealed that specific exergy losses of the examined MSF

distillers ranged between 16.7 to 23.6 kWh/m3 compared to the expenditure of only 2

kWh/m3 of an ideal reversible process and around 5 kWh/m3 for an RO process. It has,

thus been recommended that research efforts are to be focused to lower the primary

energy consumption of the MSF distiller.

1. INTRODUCTION

The first part of the Project was completed on 17th October, 1999. A commercial

computer program was acquired and used for the simulation of the MSF process. The

algorithm and overall logic of the program were first studied and modified to suit the

seven selected MSF distillers which were representing Al-Jubail Phase-II, Al-Khobar-

II, Al-Khafji and Jeddah Phase-II, III & IV desalination plants. The program was also

developed to predict the heat transfer coefficients and fouling factor of the whole

distiller as well as those of the individual stages. Operation data were collected from

the seven selected MSF distillers covering a one year period. The results of the study

were published in Technical Report No. TR 3808/APP 97002 [1] and presented in IDA

World Congress on Desalination and Water Reuse which was held in San Diego,

California in 1999 [2]. Copy of the paper which was also published in the international

journal of Desalination [3] is enclosed here as Appendix-I. The comparative thermal

analysis of the MSF distillers revealed that after more than 16 years of continuous

operation, their performance ratios were in most time higher than the design values.

This was attributed to SWCC strict requirements of operation and maintenance which

resulted in extending the MSF plant life to more than 30 years.

Detailed second law thermodynamics analysis was also carried out for each distiller to

determine the distribution of overall exergy losses among the various subsystems of the

MSF distiller and identify locations where losses of useful exergy occur. It has been

found that the total exergy losses of the examined distillers varied between 50 and 80

kJ/kg which were much larger than that necessary for the infinitely slow reversible

desalination which is only around 7.2 kJ/kg. The excess exergy introduced was a result

of the irreversibilities associated with the different subsystems. Distillers that

generated high exergy losses were Jeddah-III and IV, and both were operating at high

TBT’s and a relatively low number of stages. Al-Khobar Phase-II distiller, which also

has a limited number of stages and was subjected to seawater of high salinity, exhibited

high exergy destruction. The Al-Jubail-II, Al-Khafji and Jeddah-II distillers

demonstrated comparable and relatively low exergy are characterized by a large

specific condensing area while the third has exceptionally very high number of stages

(34 stages).

Break down of the total exergy losses of the distiller among the major subsystems

revealed that the irreversibilities inherited in the recovery section was responsible for

51 to 64 % of the total exergy losses. The exergy losses in the heat rejection section,

brine heater and exergy losses wasted in leaving streams were to some extent

comparable. However, it was found that when the condensing areas of each of the three

main heat transfer sections (heat recovery, heat rejection and brine heater) were

considered in the exergy analysis, the brine heater was in most cases responsible for the

highest exergy destruction flux. It was also found that the brine heater exergy losses

were highly influenced by the temperature of steam entering the brine heater. For

example, it was found that when Al-Khobar Phase-III distiller which was operating at

TBT of 90oC, the brine heater exergy losses were 8 kJ/Kg when the steam entering

temperature was 95 oC and increased to 15 kJ/Kg when the steam temperature was

raised to 106oC.

It was concluded that the design and operating features of an MSF distiller with an

improved thermal performance are: high number of stages, high top brine temperature

and low specific condensing area.

The first part of this project did not cover all of the SWCC major MSF plants. It was

recommended that a comprehensive thermal performance analysis have to be carried

out for the remaining plants, which included Yanbu Phase-I, Shugaig, Shoibah Phase I,

Al-Jubail Phase-I and Al-Khobar Phase-III.

The improved simulation program can also be effectively utilized in carrying out a

comprehensive parametric study to determine design and operating conditions which

can lead to the reduction of MSF irreversibility and improve overall system efficiency.

Such analysis will enable developing better understanding of the MSF process and

provides guidance for future planning and selection of an improved and optimized

design. It will also provide relevant guidance for process improvement in existing

plants.

2. OBJECTIVES (1) Few of SWCC large capacity MSF desalination plants were not included in the

scope of work of the first part of the project, which are: Yanbu Phase-I, Shugaig,

Shoaibah Phase-I, Al-Jubail Phase-I and Al-Khobar Phase-III. It is thus essential

to analyze and simulate the thermal performances of these five plants and to

compare their performances with the MSF plants which have been covered in

part-I.

(2) To utilize the developed simulation program in Part-I, for carrying out a

comprehensive parametric simulation study.

3. METHOD OF ANALYSIS (1) The inhouse modified simulation computer program which was previously

reported and described in references [1,2] was used for carrying out the

comprehensive parametric study. The program solution flow chart and the

solution scheme were explained in Appendix-I. The simulation program has got

the capability to perform energy/exergy calculations of the MSF distiller. The

program was used as a tool to investigate the impact of variation of number of

stages, top brine temperature and terminal temperature difference on the

performance ratio and specific exergy losses.

(2) For the evaluation of the thermal performance of the commercial plants, a total of

seven MSF distillers, one distiller from each of Yanbu, Shugaig, Al-Jubail Plants

and two distillers from each of Shoaibah and Al-Khobar-III plants were selected

for thermal monitoring. Field visits were arranged to collect design and

operational data such as temperatures, pressures, salinity and flow rates of the

different involved streams. The variation of performance ratio, specific exergy

losses and exergetic efficiency with the most important operating temperatures

were evaluated. Subsystem exergy analysis for each distiller was performed.

4. OPERATIONAL PERFORMANCE

The criteria used to assess the thermal performance of each distiller on a macro scale

included the water production, performance ratio, overall heat transfer coefficient and

fouling factors as compared to corresponding design values.

4.1 Al-Shoaiba Phase-I Al-Shoaiba Phase-I Power/Water Desalination Plant was commissioned in 1989. It

consists of five power/water cogeneration units. Each of the five power/water

cogeneration units consists of two MSF distillers, a boiler, one back pressure turbine

generator, a deaerator and two feed water heaters. The maximum rated water

production capacity of each of the ten distillers is 5.84 MIGD at a maximum TBT of

101.5oC. The total installed power production of the plant is 312 MW. Two dump

condensers are also available which could be connected to two of the turbines in order

to condense exhaust steam, if the need arises. MSF distillers are designed to operate at

two different modes of operation 100% load at 90°C and 120% load at 101.5oC.

The performances of two MSF distillers (unit 5 and 6) of Al-Shoaiba Phase-I plant have

been examined during the period of 11 months from June, 2000 to May, 2001. The

operating data and test results are shown in Figures 1 and 2. Both figures show that

production capacity and performance ratio of the two distillers were above the design

values. The design fouling factor (FF) of the brine heater was 0.3 m2K/kW and was

consistently higher than the operating fouling factors. The reasons for the excellent

operational thermal performance of the two units could be attributed to the over-sizing

of the heat transfer area and good performance of antiscalant in conjunction with the

use of effective sponge ball cleaning. The satisfactory thermal performance of these

two distillers discloses that there is considerable scope for further increase in water

production.

4.2 Al-Shugaig Plant Assir (Al-Shugaig) Phase-I plant was designed to produce 128 MW power and 21

MIGD potable water. The plant consists of two dual-purpose cycles. Each cycle

comprises of a boiler, and back pressure turbine coupled to two multistage flash

evaporator (MSF) units, one feed water heater and a deaerator. The capacity of each

MSF unit is 23,500 m3/d at 90oC. The generator connected to each turbine can generate

up to 64 MW.

There are four evaporators multistage flash type with cross tube configuration. Each

evaporator consists of 16 stages for heat recovery and 3 stages for heat rejection. Brine

heater and each stage of heat recovery comprises of 2760 tubes. Heat rejection

comprises of 2067 tubes per stage. The design top brine temperature and performance

ratio are 101oC and 8.2 kg/2326 kJ, respectively.

Al-Shugaig plant uses Belgard 2030 as an antiscalant with dosing rate of 1.0 ppm and

Nalco 131 antifoam with dosing rate of 0.015 ppm. 600 cleaning balls were charged

with frequency of 8 cycles per shift and 16 cycles per shift at top brine temperatures of

90 oC and 101 oC, respectively.

Figure 3 shows the thermal performance of Unit # 3 of Al-Shugaig Plant. During the

one year period, the distiller was operating with a top brine temperature range of 83 and

95oC. In most cases, the top brine temperature was maintained at around 89 oC. the

distiller production was consistently above the design value except in the period when it

was operating at partial load. The fouling factor of the brine heater was consistently

maintained below the design value.

4.3 Al-Jubail Phase – I Al-Jubail Phase-I power/desalination complex consists of six MSF evaporators each

rated at 22, 960 m3/d (5.05 MIGD). The evaporators are of the brine recirculation type

with a maximum brine temperature of 90.60oC and a performance ratio of 8.0

kg/2326kJ. There are 19 heat recovery stages and 3 heat rejection stages arranged in a

cross flow configuration. Evaporator tubing is titanium in all sections. The internal

walls of the first three stages and last stage have stainless steel cladding and the

remaining stages are of carbon steel coated with epoxy painting. Brine velocity

through the tubing is 2.00 m/sec (6.56 ft/sec) at design conditions in the brine heater

and 1.87 m/sec (6.135 ft/sec) in the heat recovery sections. The brine heater and heat

recovery sections incorporate an on-line sponge rubber ball cleaning system.

Deaeration of the make-up seawater is performed in the last stage of the evaporator.

The operational performance of the examined distiller is shown in Fig. 4. In most

cases, the unit was operating with a TBT of 90oC with antiscalant dose rate of 0.8 ppm.

The unit production during this period was slightly exceeding the design value. It is to

be noted that when the test started, the distiller was in fouled condition and fouling

factors were relatively high and almost approaching the design value. However, during

the monitoring period, the overall heat transfer coefficient and fouling factors were

virtually maintained constant which indicate that there was no sign of scale build up.

4.4 Yanbu Phase-I Yanbu Phase-I consists of five extraction condensing turbines and each turbine is

coupled to an MSF distiller. The MSF plant was originally designed to operate with

acid treatment. The distiller production at TBT of 121 oC is 21614.8 m3/day and gain

output ratio of 10.05. Each distiller consists of 21 recovery stages and 3 rejection

stages.

Although, the plant was originally designed as an acid operated plant, it is at present

operating on a cyclic treatment mode. It operates for a two month period on additive

treatment using a dose rate of 3 ppm at TBT of 115 oC followed by a one month acid

treatment using around 110 ppm of sulphuric acid. The thermal performance of the

selected distiller (unit #5) is shown in Fig.5. During an almost one year period the

TBT ranged between 113 and 115oC. The unit water production was consistently

above the design water production at 115oC which is 800 m3/hr. The performance

ratio was always maintained within the vicinity of the design performance ratio of 9.9

Kg/2326 kJ. The overall heat transfer coefficient and fouling factor of the brine heater

were consistently maintained within the permissible range.

4.5 Al-Khobar Phase-III Al-Khobar Phase-III cogeneration plant is a relatively new plant and was commissioned

in 6.5.1421H. It consists of four steam turbines coupled to eight MSF distillers. The

MSF distillers (Unit # 17 and #18) were earmarked to observe and evaluate their

thermal performances for a one year period. Each distiller consists of 16 stages and of

rated design capacity of 34000 m3/day (7.5 MIGD) at TBT of 105oC and of

performance ratio 6.5.

The thermal performances of the two distillers were shown in Figs. 6 and 7. Both

distillers were operating at the maximum design top brine temperature of 105 oC using

antiscalant Belgard EV 2030 with a dose rate of 1.5 ppm. Water productions and

performance ratios of both units were in most cases, above the corresponding design

values. The overall heat transfer coefficients of brine heater were above the design

value and there was no indication of scale formation.

5. DISCUSSION

5.1 Comparison between the Operational Performance of SWCC MSF Distillers

The operational performance of the seven MSF distillers representing Jeddah Phase-II,

III & IV, Al-Jubail Phase-II, Al-Khobar Phase-II and Al-Khafji and which was

presented in the first part of this project [1] and the seven MSF distillers representing

Al-Jubail Phase-I, Al-Khobar Phase-III, Shoaiba Phase-I, Yanbu Phase-I and Shugaig

plants which were covered in this part of the project, were compared and analyzed.

Variation of the performance ratios with time covering a one year period for eleven

MSF distillers is shown in Figure 8. The dotted lines represent the design performance

ratio of each distiller. It can be seen that the distillers instead of being deteriorated due

to ageing, the performance ratios were maintained equal or in most cases, more than the

corresponding design values. The reasons for such good thermal performance could be

attributed to several design and operating conditions, such as:

(1) selection of high design fouling factors

(2) effective alkaline scale control and

(3) good selection of construction materials.

5.1.1 Design Fouling Factor

A major factor which was responsible for maintaining MSF distiller performance above

the guaranteed performance was the selection of high design fouling factors. Table 1

shows the design overall heat transfer coefficients, fouling factors and performance

ratio of SWCC MSF distillers.

The design fouling factors (FF) of the brine heaters and heat recovery sections for

SWCC’s additive plants range between 0.176 and 0.325 m2K/kW while for acid treated

plant range from 0.0861 to 0.14 m2K/kW. Due to the good performance of antiscalant

in conjunction with effective use of sponge ball cleaning, these FF values are very

conservative (larger than required). Selection of large FFs results in a heat exchanger

containing more surface area than required. Low values of design fouling factors such

as 0.15 m2K/kW can be safely employed in new additive MSF designs.

However, the selection of high design fouling factors for the existing MSF plants,

which resulted in over sizing of the heat transfer surface, will allow to operate these

plants at top brine temperature equal to or even higher than maximum design values.

This will result in the increase of water production. So, prospects for increasing water

production from existing plants have to be explored. To achieve these objectives a

number of feasible design and operating variables have first to be verified and

confirmed such as:

(1) Availability of extra heat transfer surface area.

(2) Increase of TBT for extra production might possibly overload the flash

chambers and demister pads and give higher and possibly unacceptable distillate

conductivities.

(3) Sizes of the various operating pumps (especially brine recycle, distillate and

condensate) meet the requirements of the increased production.

(4) Overloading of distillate troughs.

(5) Possibilities of blocking small diameter tube of brine heater with scale.

(6) Capability of vacuum ejectors.

5.1.2 Scale Control As described above effective alkaline scale control was achieved through the use of

optimized antiscalant dose rates in conjunction with the use of on-line ball cleaning

system.

SWCC has been actively involved in the control of scale formation on heat transfer

surfaces. A number of optimization tests, which have been conducted in Al-Jubail,

Jeddah and Al-Khobar plants were reported [4-10].

The Research and Development Center has also been involved since its inception in the

evaluation and dose rate optimization of commercial scale inhibitors in collaboration

with Al-Jubail and Jeddah plants [11-15]. Antiscalant dose rate has been successfully

reduced to as low as 0.8 ppm for low top brine temperatures (90-98oC) and 1.8 ppm for

TBT of 110oC. Summary of antiscalant dose rate currently used in SWCC MSF plants

is shown in Table 2. There is still more scope to optimize antiscalant dose rates in

some of SWCC MSF plants. Plants, which are currently operating at low TBT of 90oC

and with dose rate of 1.0 ppm such as Al-Khafji, Shugaig and Al-Khobar II have to

consider the possibility of reducing the dose rate to 0.8 ppm as is currently partially

employed in Al-Jubail plant. Plants which are operating at high TBT such as Al-

Khobar III, Jeddah III and Yanbu I, also need to be further optimized.

All SWCC’s MSF plants are currently using on-load sponge ball cleaning systems.

The combined use of chemical additives and on-load sponge ball cleaning system has

proved to be the most cost effective procedure to avoid tube scaling. There is a wide

diversity in the ball to tube ratio and number of cycles of ball cleaning operations as

shown in Table 3. The ball to tube ratio for plants using chemical additive treatment

varies from as low as 0.22 (in Al-Shugaig plant) up to about 0.45 (in Al-Jubail Phase-I

and Al-Khobar plants). The number of cycles per operation varies between as low as 3

in Al-Shoaiba-I and up to 12 in Yanbu-I. The sponge ball cleaning frequency required

to maintain the evaporator cleanliness depends on several factors, such as brine

chemistry, type of inhibitor, ball type, MSF design parameters (temperature, number of

stages, tube length, flow pattern) [16&17]. There is a need to establish a standard

procedure for ball cleaning operation, which shall be followed by all SWCC’s MSF

plants.

5.1.3 Materials of Construction The excellent thermal performance of these plants can also be attributed to the good

selection of construction material and better understanding of the corrosion associated

with the MSF process. The most commonly used materials of construction in SWCC

MSF plants are carbon steel, stainless steel, copper-nickel alloy and titanium. Shell of

brine heaters of all plants is made of carbon steel and the tubes are either Cu-Ni 70/30

or 90/10 or modified alloy containing 66/30/2/2 Cu/Ni/Fe/Mn except Al-Jubail Phase-I

which is having titanium tubes. The material of construction of flash chambers is

carbon steel with and without cladding. In some plants such as Al-Jubail, Al-Khafji II

and Jeddah III, the first high temperature stages are cladded with stainless steel.

Module 1 of Jeddah II and the first two modules of Jeddah IV are also cladded with

stainless steel. Al-Khobar II flash chambers are completely cladded with 90/10 Cu-Ni

and Al-Shugaig I is also totally cladded with stainless steel. The materials of

construction of the heat rejection tubes of all plants are made of titanium except Jeddah

plants, which are having 90/10 Cu-Ni. Projects which are recently built use the

following material of construction for the major components: 1 Flash chambers of both

recovery and heat

rejection sections

Carbon steel lined with stainless steel (floor lined

with 317L, walls with 316L and roof with either

316L or 304.

2 Water boxes Carbon steel lined with 90/10 Copper-Nickel

3 Tubes Brine heater tubes modified alloy 66/30/2/2

Cu/Ni/Fe/Mn; heat recovery tubes: Copper/Nickel

(first four stages 70/30 and remaining stages 90/10)

4 Heat rejection tubes Titanium & modified alloy 66/30/2/2 Cu/Ni/Fe/Mn

5.2 Micro-Thermal Analysis

A comprehensive analysis based on the second law of thermodynamic was carried out

for each of the seven examined MSF plants using a simulation program. For each MSF

distiller the total exergy losses as well as their breakdown among the major subsystems

were determined. As an example, Figure 9 shows the variation of the total and

subsystem exergy losses of unit 18 of Al-Khobar Phase-III plant. The total exergy

losses varies between 76 and 95 kJ/kg. Break down of the exergy losses among the

major subsystems demonstrated that the major losses occurred in the heat recovery

section which accounts to around 60 % of the total exergy losses.

Figure 10 shows that there is a wide variation in the magnitude of the average specific

exergy losses incurred by the seven different MSF distillers and it ranged between 60 to

85 kJ/kg. The design and operating parameters which are mostly responsible for

specific exergy losses of an MSF distiller include the number of stages, TBT and

specific condensing area. Increase of number of stages and/or top brine temperature

will result in a low temperature drop per stage which in turn reduce the exergy losses.

Increase of specific condensing area also has a positive impact in the reduction of the

exergy losses.

As shown in Figure 10, the distillers which have been thermally analyzed are

characterized by a wide range of number of stages, top brine temperature and specific

condensing area. In order to facilitate a direct comparison between these seven

distillers and based on the distiller operating top brine temperatures, they have been

grouped into three categories.

The first group includes Al-Jubail Phase 1 and Al-Shugaig Plant which were both

operating at the average top brine temperature 90oC. Although, Al-Jubail distiller is

having more number of stages, it was experiencing more exergy losses compared to Al-

Shugaig and this can be attributed to its relative low specific condensing area.

In the second group, Shoaiba Phase-I and Al-Khobar Phase-III distillers were operating

at a top brine temperature of around 100oC, it has been observed that Al-Khobar III was

exhibiting much higher exergy losses due to its low number of stages and low specific

condensing area compared to Shoaiba distiller. Yanbu Phase-I distiller which forms the

third category and although it was operating at high TBT of 115 oC and with relative

low specific condensing area, yet, the relatively high number of stages resulted in low

exergy losses.

Break-up of total distiller exergy losses among the major sub-systems as shown in Fig.

10 indicates that the total specific exergy losses of the heat recovery section losses

account to 57 to 71 per cent of, heat rejection section losses 9 to 16 per cent, brine

heater 8 to 14 per cent, leaving streams 7 to 10 per cent and ejector losses 1 to 5

percent.

Comparison of the exergy destruction flux (exergy losses per unit condensing area) of

the major subsystems of condensing surfaces, as exhibited in Fig. 11 shows that the

brine heater is consistently experiencing the largest exergy destruction flux, followed

by the heat rejection and heat recovery sections, respectively.

The second low analysis revealed the exergy losses of the MSF distiller range between

60 to 85 kJ/Kg (16.7 to 23.6 kWh/m3) which was ten folds that of an ideal reversible

process and two to three folds of a SWRO membrane process. Exergy utilization is

only part of the techno-economic story. High plant availability, load factor and

reliability are one of the most important elements of water production cost. SWCC

MSF plants are characterized by very high availability. For example, the average

availability of Al-Jubail Phase-II plant during a ten year period (1991-2000) was found

to be 90%, planned shut down for major overhaul and maintenance accounted for 7%

and forced shut down accounted for only 3%. Consequently, the water production

during this periods amounts to about 91% of the design production. The statistical data

obtained from a plant of the size of Al-Jubail Phase-II which incorporates forty MSF

distillers greatly enforces the very high availability and reliability of the MSF

processes.

High availability coupled with the vast amount of acquired operating and design

experience by SWCC and the increase of life expectancy are the major factors which

contributed in the success and popularity of the MSF process. Plants of that type are

continuously being built and are in steadily increasing sizes. Recently commissioned

plants (Al-Khobar Phase III, Shoaiba Phase-II and Yanbu PhaseII) are having water

production capacities equal to higher than 7.5 MIGD. With fewer units of higher

production capacity, the need for interconnective and control piping and valves will be

much reduced. The single unit is simpler to operate and the number of operators

required is smaller. Overall, major gains in initial cost, reliability, operation and

maintenance can be expected with large unit sizes.

6. PARAMATERIC ANALYSIS

A comprehensive parametric analysis was carried out to examine the interrelationships

between the top brine temperature, number of stages, performance ratio, exergy losses

and the terminal temperature approach. The terminal temperature approach is the

difference in temperature between the condensing vapor and the recirculating brine

stream leaving the condenser. This is an important design parameter dependent on the

heat transfer area and overall heat transfer coefficient. The interrelationships between

the different parameters are presented in the form of grid network which are shown in

Figures 12 and 13 for TTD of 2 and 4 oC, respectively. The top brine temperature was

varied between 90oC (the lowest TBT for additive scale operation) and 120oC (the

highest TBT for acid operation plants).The number of stages were varied between 16

and 40. The maximum number of stages is limited by the pressure drop required to

move the flashing brine from one stage to another, especially at the cold end. In

addition, for flashing to occur the minimum inter-stage temperature drop must be

greater than the boiling point elevation. Figures 12 and 13 show that for the same

number of stages and same top brine temperature, increase of TTD from 2 to 4 oC

results in the reduction of the heat transfer area due the large driving force for heat

transfer. Both figures reveal that increasing the number of stages while keeping top

brine temperature constant such as lines AC & BD, results in an increase of both

performance ratio and specific condensing area and a decrease of exergy losses.

Increase of number of stages will decrease temperature drop per stage which will

reduce the irreversibility of the system due to the reduction in condenser and flash

exergy losses and results in an improved thermal performance. Increase of

performance ratio will reduce steam consumption and consequently minimize

operating cost while increase in surface area results in an increase of capital

expenditure. Therefore, optimum performance ratio (i.e. that of lowest production cost

of water which is site specific) has to be determined by making cost tradeoffs between

the cost of process energy and capital cost of the process. Conversely, increasing top

brine temperature while keeping number of stages constant (lines such as AB and CD)

will increase the performance ratio and decrease the specific condensing area but its net

effect on exergy destruction is not significant. Within the selected range of number of

stages and top brine temperatures, the simulation analysis revealed that the dependence

of exergy losses on the number of stages is more appreciable than that due variation of

top brine temperature.

The conditions which represent an improved MSF distiller with the least amount of

energy consumption are presented by point D in Figures 12 and 13. In both cases, the

maximum achievable performance ratio was obtained at the highest TBT of 120oC and

the maximum possible number of stages (40 stages). At these conditions, the maximum

performances ratios obtained were 13 kg/2326 kJ and 9.8 kg/2326kJ at TTD of 2 and

4oC, respectively. However, the improved energy performance obtained with a TTD of

2oC is at the expense of high requirements of specific condensing area. The specific

condensing areas are 330 and 205 m2/(kg/s) for TTD of 2 and 4oC, respectively.

The parametric analysis also demonstrated that the MSF distiller is inheriting certain

thermodynamic constraints which limit the value of maximum achievable PR. In the

present analysis the maximum PR obtained is 13 kg/2326 kJ at TBT of 120 oC and forty

stages.

The major design and operating features of SWCC MSF plants are super-imposed on

the simulated envelops shown in Figures 12 and 13. Eight of the MSF plants are

located on Figure 12 with the TTD of 2 oC. The remaining three MSF plants which are

Al-Khobar Phase-II and III and Jeddah Phase-II are located on Figure 13 with the TTD

of 4 oC. The majority of the plants superimposed on Figure 12 lied within a narrow

shadowed band. The boundaries of this band are:

1. Number of stages = 19 to 24

2. Top brine temperature = 90 to 112 oC

3. Performance ratio = 7.8 to 8.6

4. Specific condensing area = 260 to 300 m3/(kg/s)

The only exception which is not lying inside the scope of this band is Yanbu Phase-I

plant which was designed to operate on acid treatment with top brine temperature up to

121 oC.

The three plants which are not lying within the boundaries of Figure 12 and are

compatible with the Figure 13, are Al-Khobar Phase-II and III and Jeddah Phase-II.

These three plants represent the two extreme boundaries of the design and operating

features of SWCC MSF plant. Al-Khobar Phase-II & III plants uniqueness stemmed

from the relatively very low number of stages (16 stages). Conversely, Jeddah Phase-

II with its very high number of 34 stages represents the other end of the spectrum.

Selection of the most appropriate design and operating parameters for an improved

MSF distiller to minimize the overall cost of water production can only be based on a

detailed optimization study. Figure 14 shows the variation of the cost of water

production vs performance ratio for a 10 MIGD distiller. It also shows the influence of

energy cost on water cost. With the current international oil price of $24/bbl, the

optimum performance ratio which corresponds to the minimum water cost is 11.

For lower fuel costs of $ 12 and 18 per bbl, the optimum performance ratios are 8 and 9

respectively. Most of SWCC current MSF plants are having performance ratios as

mentioned above between 7.8 and 8.6 and therefore, based on low fuel cost. Due to

effective alkaline scale control and good selection of material of construction and as

mentioned in Section 5.1.2 and 5.1.3, respectively, MSF distillers can be designed to

operate at TBT up to 115oC.

7. CONCLUSIONS

(1) Monitoring the thermal performance of seven MSF distillers representing

SWCC MSF plants on the East and West coasts revealed that the distillers

instead of being deteriorated due to ageing, the production capacities and

performance ratios are still maintained within or in most cases higher than the

design values.

(2) The main developments which were introduced to SWCC MSF during the last

three decades are the successful and economic implementation of scale control

techniques, the use of appropriate corrosion resistant materials and increase of

distiller size from 2.5 to 10 MIGD. Plants which were built more than twenty

years back are still operating with high availability and load factors which

approach or even sometimes exceed the original design figures.

(3) Excellent scale control capability of currently used high temperature scale

inhibitors and on-load sponge balls cleaning allowed successful operation at low

and optimized antiscalant dose rates.

(4) Second law simulation studies revealed that there was a wide variation in the

magnitude of the average specific exergy losses of the examined MSF distillers

and it ranged between 16.7 to 23.6 kWh/m3 compared to the expenditure of only

2 kWh/m3 exergy of an ideal reversible process and around 5 kWh/m3 for an

RO process.

(5) Break down of the total distiller exergy losses among the major subsystems

revealed that 57 to 71 % were lost in the heat recovery section, 9 to 16 % in the

heat rejection section, 8 to 14% in the brine heater, 7 to 10 % in the leaving

streams and 1 to 5% in the ejector.

(6) The parametric analysis indicated that MSF processes can be improved towards

high performance ratios by increasing the top brine temperature, number of

stages and the specific heat transfer area.

(7) Currently used MSF plants are inheriting thermodynamic constraints which limit the

maximum achievable PR to less than 13 kg/2326 kJ.

8. RECOMMENDATIONS

(1) Due to excellent thermal performance and over-sizing of most of SWCC

existing MSF plants, prospects for increasing water production by increasing

TBT above design values, have to be explored.

(2) As the result of the development of effective alkaline scale control methods and

the good selection of materials of construction, the design fouling factors of

0.15 m2K/kW can be safely selected for future design of MSF distillers.

(3) Plants, which are currently operating at low TBT of 90oC and with dose rate of

1.0 ppm such as Al-Khafji, Shugaig and Al-Khobar, could consider the

possibility of reducing the dose rate to 0.8 ppm.

(4) Plants which are operating at relatively high TBT such as Al-Khobar III, Jeddah

III and Yanbu II, need further antiscalant dose rate optimization.

(5) Due to the wide diversity in the ball cleaning procedures, which are currently

used in SWCC MSF plants, it is necessary to establish a standard procedure for

ball cleaning operation, which shall then be followed by all SWCC MSF plants.

(6) Vast amount of operating and design experience of SWCC can be collected,

analyzed and used in preparing a “code of practice”.

(7) Research efforts have to be focused to lower the primary energy consumption of

the MSF distiller without sacrificing the simplicity, robustness and reliability of

the standard MSF process.

Table 1 : Design overall heat transfer coefficients, fouling factors and performance ratios of SWCC MSF distillers

1 2 3

HTC (Design) W/m2K FF (Design) m2K/W x 10-3 Design PR S.

No. PLANT BH

H Rec.

H Rej.

BH

H Rec.

H Rej

Kg./2326 kJ

1. Al-Jubail Phase I 2139 2650 2207 0.264 0.1761 0.1761 8.02

2. Al-Jubail Phase II C2/C3 2182 2650 2350 0.176 0.176 0.2 9.51

3. Jeddah Phase II 4035 3468 2743 0.086 0.0861 0.132

9.26

4.Jeddah Phase III

1999

2800

1886

0.325

0.178

0.344

7.09

5. Jeddah Phase IV 1994 2770 1839 0.325 0.176 0.299 7.02

6. Al-Khobar Phase II 2568 2868 2037 0.160 0.12 0.20 6.5

7. Al-Khobar Phase III 2128 2955 2955 0.264 0.1056 0.1056 6.5

8. Shoaiba Phase I 1884-2004 2702-2717 2298-2279 0.3 0.17 0.20 8.00

9. Shoaiba Phase II 2434 2769 2325 0.211 0.15 0.176 9.00

10. Yanbu Phase I 3253 3390 23421 0.14 0.105 0.205 10.63

11. Yanbu Phase II 1960 2657 2300 0.30 0.17 0.20 8.61

12. Shugaig Phase I 1885 2539 2219 0.30 0.17 0.2 8.26

13. Al-Khafji Plant 2049 2800 2300 0.279 0.279 0.279 8.19

Table 2. Current antiscalant dose rate of SWCC MSF plants

S. No. Plants Top brine Temp. (oC) Dose rate (ppm)

1 Al-Jubail Phase I 90 0.8

2 Al-Jubail Phase II 98 0.8

3 Al-Khafji 90 1.0

4 Al-Khobar Phase II 90 1.0

5 Al-Khobar Phase III 105 1.5

6 Jeddah –III 108 3.0

7 Jeddah IV 110 1.8

8 Shoaiba I 101 1.1

9 Yanbu I 115 3.0

10 Yanbu II 108 2

11 Shugaig I 90 1.0

Table 3. On-load sponge ball cleaning

Ball/ Tube Ratio S. No. Plants Chemical

Treatment BH HRC Frequency of Ball Cleaning

Operation

No. of cycles per operation

1 Al-Jubail Ph. I Antiscalant 0.450 0.427 3 Oper. / Day 8 Cycles / Oper.

2 Al-Jubail Ph. II

C2 & C3 Antiscalant 0.342 0.324 3 Oper. / Day 8 Cycles / Oper.

C4 Antiscalant 0.270 0.257 3 Oper. / Day 8 Cycles / Oper.

C5 Antiscalant 0.300 0.302 3 Oper. / Day 8 Cycles / Oper.

3 Jeddah Ph II Acid 0.296 0.236 One/week 3 cycle/oper

4 Jeddah Ph III Antiscalant 0.29 0.665 3 Oper. / Day 4 Cycles / Oper.

5 Jeddah Ph IV Antiscalant 0.251 0.370 2 Oper./ Week 10 Cycles / Oper.

6 Al-Khobar II Antiscalant 0.453 0.458 3 Oper. / Day 9 Cycles / Oper.

Antiscalant 0.243 0.249 3 Oper. / Day 12 Cycles / Oper. 7 Yanbu I

Acid 0.243 0.249 One Oper./ Week 12 Cycles / Oper.

8 Al-Shugaig Antiscalant 0.22 0.22 3 Oper/ Day 8 Cycles / Oper.

( 16 for high TBT )

9 Al-Shoaiba I Antiscalant 0.251 0.253 3 Oper/ Day 3 Cycles / Oper.

10 Al-Khafji Antiscalant 0.351 0.351 One Oper/ Day 9 Cycles / Oper.

Perf

orm

ance

Rat

io

Kg/

2326

kJ

Perf

orm

ance

Rat

io

Kg/

2326

kJ

PWd = 979

PRd = 8.2

HTCd = 1885

FFd = 0.3

Figure 4: Thermal Performance of Unit 2 of Al-Jubail Plant Phase I

6.0

7.0

8.0

9.0

10.0

0 30 60 90 120 150 180 210 240 270 300 330 360

500

700

900

1100

1300

1500

0 30 60 90 120 150 180 210 240 270 300 330 360

0

2

4

6

8

0 30 60 90 120 150 180 210 240 270 300 330 360

0

0.1

0.2

0.3

0.4

0.5

0 30 60 90 120 150 180 210 240 270 300 330 360Days

40

60

80

100

120

140

0 30 60 90 120 150 180 210 240 270 300 330 360

TBT (oC) Flash Range (oC)

HTCd= 2.65

PRd= 8.0

PRd= 8.0

Tem

pera

ture

(o C)

Pr

oduc

tion

(m3 /h

)

HT

C o

f B

rine

Hea

ter

(kW

/m2 K

)

Perf

orm

ance

Rat

io

(kg/

2326

kJ)

FF

of B

rine

Hea

ter

(m2 K

/kW

)

PWd = 960

FFd = 0.264

0

40

80

120

160

0 30 60 90 120 150 180 210 240 270 300 330 360

Flash range TBT

600

800

1000

1200

0 30 60 90 120 150 180 210 240 270 300 330 360

6

8

10

12

14

0 30 60 90 120 150 180 210 240 270 300 330 360

2

2.5

3

3.5

4

4.5

5

0 30 60 90 120 150 180 210 240 270 300 330 360

0

0.05

0.1

0.15

0 30 60 90 120 150 180 210 240 270 300 330 360

Days

Production Design = 800 m3/h

PR Design= 9.9

FF Design= 0.1204

HTC Design= 3.253

Figure 5: Thermal Performance of Yanbu Plant Phase I

Tem

pera

ture

(o C)

Pr

oduc

tion

(m3 /h

)

HT

C o

f B

rine

Hea

ter

(kW

/m2 K

)

Perf

orm

ance

Rat

io

(kg/

2326

kJ)

FF

of B

rine

Hea

ter

(m2 K

/kW

)

60

70

80

90

100

110

120

0 50 100 150 200 250 300 350 400

TBT(°c) Flash Range(°c)

1300

1350

1400

1450

1500

1550

0 50 100 150 200 250 300 350 400

456789

10

0 50 100 150 200 250 300 350 400

0

0.2

0.4

0.6

0.8

0 50 100 150 200 250 300 350 400

Days

0123456

0 50 100 150 200 250 300 350 400

PR=6.499

PW Design =1416

FF=0.264

HTC=2.128

Figure 6: Thermal Performance of Unit 17 of Al-Khobar Plant Phase III

Tem

pera

ture

(o C)

Pr

oduc

tion

(m3 /h

)

HT

C o

f B

rine

Hea

ter

(kW

/m2 K

)

Perf

orm

ance

Rat

io

(kg/

2326

kJ)

FF

of B

rine

Hea

ter

(m2 K

/kW

)

PRd = 6.499

HTCd =2.128

FFd = 0.264

Figure 7: Thermal Performance of Unit 18 of Al-Khobar Plant Phase III

20406080

100120140

0 30 60 90 120 150 180 210 240 270 300 330 360

TBT(°c) Flash Range(°c)

0

1

2

3

4

5

0 30 60 90 120 150 180 210 240 270 300 330 360

1200

1300

1400

1500

1600

0 30 60 90 120 150 180 210 240 270 300 330 360

HTCd=2.1281

0

0.2

0.4

0.6

0 30 60 90 120 150 180 210 240 270 300 330 360

5

6

7

8

9

10

0 30 60 90 120 150 180 210 240 270 300 330 360

PRd=6.499

Days

Production Design = 1416

Tem

pera

ture

(o C)

Pr

oduc

tion

(m3 /h

)

HT

C o

f B

rine

Hea

ter

(kW

/m2 K

)

Perf

orm

ance

Rat

io

(kg/

2326

kJ)

FF

of B

rine

Hea

ter

(m2 K

/kW

)

FFd = 0.264

6.4

6.9

7.4

7.9

8.4

0 40 80 120 160 200 240 280 320 360 400

7.5

8.5

9.5

10.5

0 40 80 120 160 200 240 280 320 360 400

6

6.5

7

7.5

8

0 40 80 120 160 200 240 280 320 360 400

9

10

11

12

0 40 80 120 160 200 240 280 320 360 400

7

7.5

8

8.5

0 40 80 120 160 200 240 280 320 360 400

5

6

7

8

9

0 40 80 120 160 200 240 280 320 360 400

7.0

8.0

9.0

10.0

11.0

0 50 100 150 200 250 300 350 400

6.0

7.0

8.0

9.0

0 40 80 120 160 200 240 280 320 360 400

6

8

10

12

0 40 80 120 160 200 240 280 320 360 400

7

8

9

10

0 40 80 120 160 200 240 280 320 360 400

6.0

7.0

8.0

9.0

10.0

0 40 80 120 160 200 240 280 320 360 400

PRd=9.0

Figure 8. Operational Performance of MSF Distillers

TBT=106 – 110 oC Production 872 – 923 m3/h

Jeddah IV :-

TBT=100 – 110 oC Production 516 – 592 m3/h

Jeddah III :-

TBT=103 – 116 oC Production 270 – 488 m3/h

Jeddah II :-

TBT=84 – 90 oC Production 900 – 1040 m3/h

Jubail I :-

TBT=90 – 98 oC Production 923 – 1250 m3/h

Jubail II :-

TBT=72.7 – 88.7 oC Production 312 – 488 m3/h

Khafji :-

TBT=113 – 115 oC Production 826 – 905 m3/h

Yanbu I :-

TBT=89– 95oC Production 1008 – 1159 m3/h

Shugaig:-

TBT=88 – 98 oC Production 875 – 1120 m3/h

Shoiaba I:-

TBT=100.5 – 105 oC Production 1304 – 1525 m3/h

Khobar III:-

TBT=82 – 92 oC Production 865 – 975 m3/h

Khobar II:-

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR

(kg/

2326

kJ)

PR

(k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PR (k

g/23

26kJ

)

PRd=9.26

PRd= 7.09

PR= 7.02

PRd= 8.02

PRd= 8.19

PRd= 6.82

PRd= 6.39

PRd= 8.0

PRd= 8.00

B

TBT oC

50

60

70

80

90

100

110

6 7 8 9 10 11 12 13 14Performance Ratio, kg/2326 kJ

Exe

rgy

Loss

es, k

J/kg

0

50

100

150

200

250

300

350

400

Con

dens

ing

Are

a, m

2 /(kg

/s)

90100112120

Number of Stages 16

A

B

C D

16

20

24

28

3640

32

Number of Stages 16

20

A

2428

32 36 40 C

2024

28

3236

D40

24

28

32

36

40

Figure 12: Dependence of Performance Ratio, Exergy Losses and Specific Condensing Area on TBT and Number of Stages (Terminal Temperature Difference TTD =2oC )

20

TBT, oC

1

2 3 4

6

8

1

3

2

4

5

5

7 8

6

7

Legend 1. Al-Jubail I 2. Al-Shoqaq 3. Shuaiba-5 4. Shuaiba-6 5. Yanbu-I-5 6. Al-Khafji 7. Jed-III-10 8. Jed-IV-19

40

C

70

80

90

100

110

120

130

3 4 5 6 7 8 9 10 11 12 13 14

Perfo rman ce R a tio , kg /2326 kJ

Exe

rgy

Loss

es, k

J/kg

90

110

130

150

170

190

210

230

Con

dens

ing

Are

a, m

2 /(kg/

s)

90100112120

16

Num ber o f S tages

3 2

20

24

32

36 40

28

Number of S tages 16

24

36

20

D

C

A 24

28

36

20

3240

40

28

DC

BA

S tages 16

20

24

32 36

40

28

Figure 13: Dependence of Performance Ratio, Exergy Losses and Specific Condensing Area on TBT and Number of Stages (Terminal Temperature Difference TTD =4oC )

TBT, oC 1

3

2

2

1

Legend: 1. Al-khobar-III 2. Al-Khobar-II 3. Jed-II-5

3

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2

6.5 7 8 9 10 11 12PR

Fule cost = 24 $/bbl

Fule cost = 18 $/bbl

Fule cost = 12 $/bbl

Figure 14. Variation of total water production with performance for 10MIGD MSF distiller.

Tot

al W

ater

Pro

duct

ion

Cos

t $/m

3

10. REFERENCES

1. Hamed, A.O., et al, (1999), Modeling and Simulation of Multistage Flkash Distillation Process, Report No. TR 3808/APP 97002, December.

2. Hamed, A.O., et al, (2000), Thermal Performance of Multistage Flash Distillation Plants in Saudi Arabia, Desalination 128, 281-292.

3. Al-Mudaiheem, A.M.A., and Szostack, R.M., (1986), Evaluation of Chemical Additives for Scale Control in MSF Plants, Topics in Desalination (13), SWCC, Saudi Arabia.

4. Al-Sofi, M.AK., Mustafa, G.M., Hassan, A.M., Dalvi, A.G.I and Kither, N.M., (1998), Nanofiltration as a means of achieving higher TBT of ≥ 120oC in MSF, J. Desalination 118, pp. 123-129.

5. Al-Sofi, M.A., (1999), Fouling Phenomena in Multistage Flash (MSF) distillers, Desalination 126, 61-76.

6. Al-Sofi, M.AK., Al-Husain, M.A., Al-Omaran, A.A. and Farran, K.M., (1994), A Full Decade of Operating Experience on Al-Khobar-II Multistage Flash (MSF) Evaporators (1982-1992), Desalination 96,.313-323.

7. Al-Sofi, M.AK., Al-Hussain, M.A. and Saleh, G.A., (1987), Additive Scale Optimization and Operation Modes, Desalination 66, 11-32.

8. Al-Sofi, M.AK, Khalaf, S. and Al-Omaran, A.A., (1989), Practical Experience in Scale Control, Desalination 73, 313-325.

9. Al-Zahrani, G. S., Al-Ajlan, A.M.and Al-Jardan, A.M., (1993), Using Different Type of Antiscalants at the Al-Jubail Power and Desalination Plant in Saudi Arabia, Proceedings of IDA and WRPC World Conference on Desalination and Water Treatment, Yokohama, Japan, Vol. 1, 421-431.

10. Hamed, O. A., Al-Sofi, M.AK., Imam, M., Ba-Mardouf, K., Al-Mobayed, A.S. and Ehsan, A., (1996), DSB(M) Antiscalant Testing at low dose rate”, Research & Development Center, Report No. TSR-3808/97026 Part I.

11. Hamed, O. A., Al-Sofi, M.AK., Ba-Mardouf, K, (2001), Successful Operation of Multi-Stage Flash distiller at Low Antiscalant Dose Rate, Proc. of the 4th Annual Workshop on Water Conservation in the Kingdom, KFUPM, D-2-1.23 - 25 April, Dhahran, Saudi Arabia,

12. Hamed, O. A., Al-Sofi, M.AK., Imam, M, Ba-Mardouf, K., Al-Mobayed, and Ehsan, A, (2000), Evaluation of Polyphosphenate Anti-scalant at a low dose rate in Al-Jubail Phase II MSF Plant, Saudi Arabia, J. Desalination 128, 275-280.

13 Hamed, O. A., Al-Sofi, M.AK., Imam, M, Ba-Mardouf, K., Al-Mobayed, and Ehsan, A, (1998), DSB(M) Antiscalant Testing at low dose rate in Al-Jubail Plant, Research & Development Center, Report No. TSR-3808/97026 Part II.

14. Hamed, O. A., Al-Sofi, M.AK., Mustafa, G.M, and Dalvi, A.G.I., (1999), The performance of different antiscalants in multistage flash distillers, Desalination 123, 185-194.

15. Böhmer, H, (1993), On-load tube cleaning systems and debris filters for avoidance of microfouling in MSF desalination systems, Desalination 93, 171.

16. Böhmer, H., (1998), On-load sponge ball cleaning system”, Encyclopedia of Desalination and Water Resource, Chapter II.

Summary of the technical report of Part-I of this project was presented at International Association World Congress on Desalination and Water Reuse which was held in San Diego, USA, 29th August – 3 Sept. 1999.

THERMAL PERFORMANCE OF MULTISTAGE FLASH DISTILLATION PLANTS IN SAUDI ARABIA

Osman A. Hamed, Mohammad AK. Al-Sofi, Monazir Imam,

Ghulam M. Mustafa, Khalid Ba-Mardouf and Hamad Al-Washmi

Research & Development Center, Saline Water Conversion Corporation P. O. Box- 8328, Al-Jubail, Kingdom of Saudi Arabia

ABSTRACT

The Saline Water Conversion Corporation (SWCC) of Saudi Arabia is currently producing around 16% of the total worldwide capacity of desalted water [1]. The majority of SWCC desalination plants employ the multistage flash (MSF) distillation process which produce 94% of SWCC's total desalinated water. SWCC various MSF distillers are characterized by a wide range of operating and design conditions. The capacity of operating distillers is ranging between 2.5 to 7.5 MIGD. Some large distillers of rated capacity up to 10 MIGD are under construction in Shoibah and Yanbu. The number of stages of these distillers varies between 16 and 34 while the operating top brine temperature varies between 90oC and 115oC. Design and operating parameters of various SWCC MSF distillers have been collected and effectively utilized to simulate and analyze the thermal performance of these distillers. The thermal performance of each distiller is quantitatively assessed using a computer program which is based on the first and the second law of thermodynamics. A comparative study on energy and exergy analyses is conducted for different MSF plants. This paper also presents a comprehensive micro-thermal analysis to identify the potential for improving plant efficiency. The exergy losses due to irreversibility in different subsystems of each MSF distiller is evaluated. Exergy destruction in brine heater, heat recovery and heat rejection stages and friction in liquid paths are determined. The impact of different design and operating parameters such as top brine temperature and number of stages on the thermal performance of the MSF system is evaluated. Keywords: Multistage flash distillation, Thermodynamics, Exergy and Exergy flux

INTRODUCTION Accumulated experience obtained from the operation of SWCC plants can be effectively utilized to simulate and analyze the thermal performance of MSF distillers. Steady and unsteady state simulation models are normally used to determine the performance of an operating MSF plant under wide range of process parameters. They also give relevant guidance for process improvement and simulate for short-term changes in the operating conditions. Furthermore, they provide design parameters for new projects of desalination plants. Most of the thermodynamic analyses performed on MSF desalination plants are based on the first law of thermodynamics [2-1٢]. Although the first law is an important tool in evaluating the overall performance of desalting plants, such analysis seldom takes into account the quality of energy which is transferred. Thus, the differentiation between high and low grade energy are not clearly evident in the majority of such research works. On the other hand, the second law analysis places all the energy interactions on the same basis thus giving relevant guidance on process improvement. In this approach, all losses are calculated in terms of exergy (available energy) which would be a true measure of these irreversible processes. Exergy method gives information on the process details which are mainly responsible for the energy losses and thus can identify locations where losses of useful energy occur within the process [13]. Reported research work on exergy analysis of the MSF process is limited. An exergy analysis of a simplified once-through multistage flash evaporator was reported [14]. In this analysis, the exergy input / output ratio of the 10 stage MSF system was found to be around 88 which revealed that the distiller experienced a very low exergy efficiency. A comprehensive study on the second law analysis of recirculation multistage flash desalting plant was reported [15]. The study showed that the major exergy losses occurred in the vapor condensation process, vapor flash process and feed heating process of brine heater. Exergy losses can be reduced by increasing number of stages and gain output ratio or by reducing top brine temperature. A study based on energy and exergy considerations was conducted to obtain the optimum design parameters for an MSF seawater distillation plant to be coupled to an existing steam power plant in Abu Dhabi [16]. Sulaiman and Ismail reported a simple scheme to evaluate overall exergy losses in Al-Khobar II, Al-Jubail II and Shoibah-I [17]. The study was limited to the design conditions and no actual test data was used in the thermal analysis. The study showed that the three MSF desalting plants were highly irreversible and the exergy losses were directly proportional to the top brine temperature. In this paper, it is intended to evaluate and compare the thermal performance of SWCC commercial MSF plants under a wide range of actual operating conditions using the concepts of the first and second law of thermodynamics. METHOD OF ANALYSIS A commercial computer program for MSF process simulation which was developed by El-Sayed [18] is used to analyze the thermal performance of six distillers of varying features representing Al-Jubail, Al-Khobar, Al-Khafji and Jeddah MSF desalination plants. Design characteristics of these distillers are shown in Table 1. Detailed design specifications of SWCC MSF plants were published [19-21]. The program solution

flowchart is shown in Figure 1. The program has the capability to predict the physical and the thermodynamic properties of all liquid and vapor streams involved in the process and to simulate the MSF process under various conditions. Concepts of the first and the second law of thermodynamics are used in this simulation study. Performance ratio was used as the first law evaluation criterion while specific exergy losses due to process irreversibility, exergy destruction flux and exergy rational efficiency were used as the second law performance criteria. Field visits were arranged to these plants to collect operating data. For each distiller, the operational data collected include temperature, pressure, flow-rate and salinity of all streams. Frequency of data collection ranged between 1 and 3 weeks.

Parameter No. of stages

Plant Heat Recovery

Heat Rejection Total Production

m3/day TBT , oC GOR

Jeddah Phase II 31 3 34 10,800 115 9.5 Phase III 14 2 16 22,000 108 6.9 Phase IV 21 3 24 22,000 108 7.0

Al-Jubail Phase II 19 3 22 24,000 90 7.8 Al-Khobar Phase II 13 3 16 21,500 90 6.5 Al-Khafji Phase II 19 3 22 11,600 87.8 9.5

Table 1. Design Parameters of the Examined MSF Distillers RESULTS AND DISCUSSION Macro Thermal Analysis For each distiller the variation of performance ratio, specific exergy losses and exergy efficiencies with time were evaluated. The impact of short-term changes in the operating conditions such as TBT, seawater inlet temperature and steam temperature entering the brine heater on the distiller performance was examined. Jeddah MSF Plants Three distillers representing phases I, II, III of Jeddah plants were selected to analyze their thermal performances. Variation in operating conditions and their effects on system exergy with time for these three distillers are shown in Figure 2. During the period of performance analysis, seawater salinity was within the range of 40,500 to 41,000 ppm while its temperature ranged between 28 to 30 oC. Figure 2 shows that Jeddah II distiller, working at a TBT of 115 oC, yields the highest performance ratio (PR) of around 11.5. This high PR value is expected because of the higher number of stages as well as the operation with a relatively higher TBT compared to other groups. The specific exergy losses of the distiller are relatively low and range between 54 to 60 kJ per kg product, which is reflected in the relatively high rational exergy efficiencies which range between 5.8 and 6.4 percent. Although Jeddah IV distiller has higher number of stages compared to Jeddah III distiller and is operating with a higher TBT, it yields a lower thermal performance. The unit performance ratio and rational exergy efficiencies are lower than those of Jeddah-

III distiller. Decrease in thermal performance is attributed to its low specific condensing area which is 20 % lower than that of the phase III unit. However, both Jeddah III and IV are generating higher exergy losses and hence higher irreversibility compared to Jeddah II which incorporates high number of stages.

Al-Jubail Plant Figures 3 shows thermal performance of Al-Jubail II plant. The unit was operating mostly at a TBT of 90 oC and after 200 days of operation the unit was shut down for acid cleaning and thereafter TBT was increased to 95 oC. The unit performance ratio during the first 200 days was ranging between 7.7 and 8.4 which is within design performance ratio of 8.01 at 90 oC. During this period, the unit specific exergy losses varied between 57 and 61 kJ/kg and the rational exergy efficiency varied between 5.8 and 6.2 %. After tubes acid cleaning, the thermal performance of the unit improved remarkably. The performance ratio increased to above 9 and the specific exergy losses dropped below 53 kJ/kg which is reflected in an increase of exergy efficiency to 6.6 percent. This can be attributed to the combined effect of acid cleaning and high TBT operation.

Al-Khobar Plant Figure 4 shows the variation in the thermal performance of Al-Khobar Phase II distiller. The TBT of the monitored distiller was maintained between 84 and 92 oC while the steam temperature varied between 92 and 108 oC. The unit which is a 16 stage distiller was yielding low performance ratio ranging between 6.7 and 7.6 and relatively higher specific exergy losses which varied between 64 to 75 kJ/kg while the rational exergy efficiency varied between 6.7 and 5.6 percent. It is observed that the increase of steam temperature resulted in an increase of the specific exergy losses. This is because the thermal energy supplied to the brine heater has a high exergy value which is eventually dissipated due to phase change. Although the distillers experienced a relatively high specific exergy losses, it was not reflected in the magnitude of rational exergy efficiency. This is due to the fact that the unit was subjected to a make up of relatively high salinity exceeding 50,000 ppm. In other words, the unit was producing a useful product output with a relatively elevated chemical exergy related to the seawater make up .

Al-Khafji Plant Figure 5 shows the thermal performance of Al-Khafji distiller. Although the distiller is operating with a low TBT around 87 oC, it was operating at relatively high performance ratio. The performance ratio ranged between 8.8 and 9.4 which is 10 to 20 % higher than the design value. This could be attributed to the high specific condensing area of the distiller which is 3.84 m2/m3/day. The distiller specific exergy losses ranged between 51 and 61 kJ/kg. During the first 80 days of testing, the specific exergy losses was decreasing despite the constant values of TBT and steam temperature. This could be attributed to the increase of the seawater temperature during that period and which is directly reflected into a decrease of the flash range. The increase of specific exergy losses after 80 days is mainly due to increase of the steam temperature which led to an increase of exergy losses in the brine heater.

Micro-Thermal Analysis It is essential to determine the distribution of the overall exergy losses among the various subsystems of the MSF distiller and identify locations where losses of useful exergy occur within the process. Subsystems which are responsible for exergy losses include brine heater, heat recovery section, heat rejection section, leaving streams and the ejector system. The magnitude of the exergy losses in each subsystem is calculated for each investigated distiller and shown in Figure 6. It shows that specific exergy losses are mainly dependent upon steam and top brine temperatures, number of stages and specific condensing area. Al-Khobar distiller was presented by two bar charts where in both cases all the operating parameters are kept constant except the temperature of steam entering the brine heater. Increase of steam temperature from 95oC to 105oC resulted in 30% increase of exergy losses in the brine heater while exergy losses in the other subsystems were largely unchanged. The influence of steam temperature on the overall exergy destruction is thus very significant. Al-Jubail and Al-Khafji distillers are characterized by relatively low specific exergy losses and this can be attributed to their large specific condensing areas. Although Jeddah II is operating at a high TBT, it is having low exergy destruction because of its large number of flash stages which resulted in moderate temperature drop in each stage. Conversely, Jeddah III and IV are both showing high specific exergy losses and this is due to the high TBT operation and low condensing areas. For all distillers major exergy destruction has occurred in heat recovery section which accounts for more than 50 percent. The specific exergy losses in the brine heater, heat rejection and losses through leaving streams are much lower than those of the recovery section. The recovery section represents the largest part of the distiller and its condensing area is several times higher than that of both the rejection section and brine heater. Thus the high exergy losses in the recovery section is primarily due to its large condensing area. To make a rational comparison between the irreversibility associated with the recovery section to that associated with the brine heater and rejection section, it was essential to determine the exergy destruction flux (exergy per unit condensing area) for each subsystem. Figure 7 shows that the recovery section in most cases is exhibiting the lowest exergy destruction flux. The only exceptions are Jeddah III and IV which are having comparable values of exergy flux in the recovery and brine heater. This is because both units are operating at high TBT and are having less number of stages compared to Jeddah II which is also operating at a high TBT. Increase of TBT; while number of stages are not proportionally increased; causes an increase of both condenser and flashing exergy losses due to increase of temperature drop per stage.

CONCLUSIONS

1. Thermodynamic analysis of MSF distillers representing Jeddah, Al-Jubail, Al-Khobar and Al-Khafji desalination plant showed that all the examined distillers are having performance ratios equal to or higher than the design values.

2. Specific exergy losses of distillers are found varying between 54 and 77 kJ/kg distillate. These losses are much higher than that necessary for an ideal reversible

process which is only 7.2 kJ/kg distillate[22]. The rational exergy efficiency of the examined distillers ranged between 4.3 to 6.7 percent.

3. Distillers which were generating high exergy losses are Jeddah III and IV and both are operating at high TBT and of relatively low number of stages. Al-Khobar distiller which is also of limited number of stages and subjected to seawater of high salinity exhibited high exergy destruction.

4. Al-Jubail, Al-Khafji and Jeddah II distillers demonstrated comparable exergy destruction. The first two plants are characterized by large specific condensing area while the third is having more number of stages.

5. Subsystems’ exergy analysis revealed that the brine heater in most cases is responsible for the highest exergy destruction flux. Brine heater exergy losses are highly influenced by steam temperature, and its associated exergy contents.

6. Overall exergy losses due to irreversibilities in different subsystems can be reduced by increasing the number of stages and specific condensing area and by decreasing heating steam temperature.

REFERENCES

1. Wangnick Consulting & GMBH, (1998). IDA worldwide desalting plants inventory report 15.

2. Burley, M.J.(1967). Analytical comparison of the multistage flash and long tube vertical distillation, Desalination, 2, 81-88.

3. Darwish, M. A. and El-Hadik, A. A.(1986). The multieffect boiling desalting system and its comparison with the multistage flash system, Desalination, 60, 251-265.

4. El-Dessoukey, H., shaban, H. I. and Al-Ramadan, H. (1995). Steady state analysis of multistage flash desalination process, Desalination, 103, 271-287.

5. Hamed, O. A. and Aly, S., (1991). Simulation and design of MSF desalination processes, Desalination, 80, 1-14.

6. Hussain, A., Woldai, A., Al-Radif, A., Kesou, A., Borsani, R., Sultan, H. and Deshpandey, P.B. (1994), Modelling and simulation of a multistage flash (MSF) desalination plant, Desalination,97, 555-586.

7. Honburg, C.D. and Walson, B.M.(1993), Operational Optimization of MSF Systems, Desalination, 92, 331-351.

8. Tanios, B.Z.(1984). Marginal operation field of existing MSF distillation plants, Desalination, 51, 201-212.

9. Darwish, M. A.,(1991). Thermal analysis of multistage flash desalting systems, Desalination, 59-79.

10. Helal, A.M., Medani, M.S. and Soliman, M.A.(1986). A Tridiagonal matrix model for multistage flash desalination plants, Computers and Chemical Engineering, 10, (4), 327-342.

11. Rasso, M., Beltramini, A., Mazzotti, M and Morbidelli, M. (1996). Modelling multistage flash desalination plants, Desalination,108, 335-364.

12. El-Dessoukey, H., Shaban, H. I. and Al-Ramadan, H. (1995). Steady state analysis of multistage flash desalination process, Desalination,103, 253-279.

13. El-Sayed, Y.M. (1992). Advances in the methodologies of optimal thermal design, Proceedings of Desal 92 Arabian Gulf Regional Water Desalination Symposium, Al-Ain, Unites Arab Emirates, 2, 555-595.

14. Koot, L.W. (1968). Exergy losses in a flash evaporator, Desalination, 5, 331-348.

15. Darwish, M. A., Al-Najem, N. M. and Al-Ahmed, M. S.(1993). Second law analysis of recirculating multistage flash desalting system, Desalination, 89, 289-309.

16. El-Nasher, A.M.(1994). An MSF evaporator for the UANW 9 and 10 power station. Design consideration based on energy and exergy, Desalination, 107, 253-279.

17. Sulaiman, F.A. and Ismail, B. (1995). Exergy analysis of major recirculating multistage flash desalting plants in Saudi Arabia, Desalination, 103, 265-270.

18. El-Sayed, Y.M. (1998). Design and simulation software, Advanced Energy Systems, Fremont, CA, USA.

19. Al-Sofi, M. AK., Al-Hussain M. A. and Al-Zahrani S. G. (1987). Additive scale control optimization and operation modes, Desalination, 66, 11-32.

20. Al-Mudaiheen, A. M., Al-Sofi, M. AK., Al-Omran, A. A. and Al-Jardan, A. A. (1986). Practical experience in operating multistage flash MSF evaporators, Topics in Desalination, Saline Water Conversion Corporation.

21. Nada, N. (1986). Operating experience of MSF units in Saudi Arabia, Topics in Desalination, Saline Water Conversion Corporation.

22. Spiegler, K. S. (1983). Thermal analysis , Desalination, 44, 3-16.

.

Figure 1: Mathematical Models and Computation Flow Chart

INPUT DATA Number of stages, temperature, flow and concentration of boundary streams, top brine temperature (TBT), terminal

temperature difference (TTD), pressure rise in pumps and pressure drop in heat input, recovery and rejection stages.

Initialize stage brine temperatures Initialize makeup and brine recycle flow rates.

STAGE TO STAGE CALCULATION Computation for mass and energy balances

Calculate temperature, flow rate and salinity of brine and vapor in each stage.

Calculate total distillate, recycle and makeup flows Calculate the difference between the calculated and targeted

blowdown concentrations (DIFF CONC)

Calculate number of recovery and rejection stages Determine pressure drop per stage for recovery and rejection

ABS(DIFF CONC) ≤ 0.02

THERMAL CALCULATION Perform thermal calculations for determination of

HTC, GOR and PR.

EXERGY CALCULATION Calculate the thermo-mechanical and chemical exergy for flashing brine, distillate vapor, condensate and heated brine in each stage.

Determine exergy input for recycle, makeup, distillate product, blowdown, and cooling water pumps

EXERGY ACCOUNTING Calculate exergy destruction in brine heater, recovery and

rejection section, wasted in leaving streams, friction in liquid paths and useful chemical exergy of the product.

Specific exergy losses, ε = Total exergy lossesDistillate production

, and

Exergetic Efficiency, η = Useful chemical exergy outputThermomechanical exergy input

STOP

START

Seawater Properties

Seawater Properties

NO

YES

Adjustment of makeup and brine recycle flows

100

105

110

115

120

TBT(

°c)

TBT(°c)-2 TBT(°c)-3 TBT(°c)-4

60

65

70

75

80

Flas

h R

ange

(°c)

Flash Range(°c)-2 Flash Range(°c)-3 F. R. Temp.(°c)-4

6

8

10

12

14

P.R

. (kg

/232

6kJ)

PR. Plant-2 PR. Plant-3 PR. Plant-4

50

60

70

80

90

Sp. E

xerg

y Lo

sses

(KJ/

Kg) Sp. Exergy losses-2 Sp. Exergy Losses-3 Sp. Exergy Losses-4

4

4.6

5.2

5.8

6.4

7

0 10 20 30 40 50 60 70 80 90 100

Days

Exer

gy E

ffici

ency

Exergy Eff.-2 Exergy Eff.-3 Exergy Eff.-4

Figure 2: Operational Performance of Jeddah II, III & IV Plants

0

20

40

60

80

100

120

Tem

pera

ture

(o C )

SW Temp TBT

Steam Temp Flash Range

7

7.5

8

8.5

9

9.5

P.R

. (k

g/23

26kJ

)

45

50

55

60

65

70

Spec

ific

Exer

gy L

osse

s, (k

J/kg

)

5.5

6

6.5

7

7.5

8

0 20 40 60 80 100 120 140 160 180 200 220 240

Days

Exer

gy E

ffici

ency

Acid Cleaning

Acid Cleaning

Figure 3: Operation Performance of Al-Jubail Plant Unit # 8

60

64

68

72

76

80

Sp. E

xerg

y Lo

sses

, (kJ

/kg)

0

20

40

60

80

100

120

Tem

pera

ture

(o C)

SW Temp. TBTSteam Temp. Flash Range

6.6

6.8

7

7.2

7.4

7.6

P.R

. (kg

/232

6 kJ

)

5

5.4

5.8

6.2

6.6

7

0 20 40 60 80 100 120 140 160 180 200

Days

Exer

gy E

ffici

ency

Figure 4: Operation Performance of Al-Khobar Phase II Plant

8.2

8.6

9

9.4

9.8

10.2

P. R

. (kg

/232

6kJ)

44

48

52

56

60

64

Spec

ific

Exer

gy lo

sses

(kJ/

kg)

5.5

6

6.5

7

7.5

8

0 20 40 60 80 100 120 140 160

Days

Exer

gy E

ffici

ency

0

20

40

60

80

100

120

Tem

pera

ture

(o c)

SW Temp(°c) TBT(°c)Steam Temp(°c) Flash Range(°c)

Figure 5: Operational Performance of Al-Khafji Plant

0

10

20

30

40

50

60

70

80Sp

ecifi

c Ex

ergy

Los

ses

(KJ/

Kg

)Destruction in Brine Heater Wasted in EjectorWasted in Leaving StreamsDestruction In RejectionDestruction in Recovery

Plant Khobar Khobar Jubail Khafji Jeddah 2 Jeddah 3 Jeddah 4Unit No. 2 2 8 1 5 10 19 Steam Temp.(oC) 105 95 95 94 120 112oC 115oC

TBT, oC 90 90 90.6 87 115 108 110 No. of Stages 16 16 22 22 34 16 24 Ad/Md, (m2/m3/day)

2.85 2.85 3.56 3.84 1.646 2.25 1.7

Figure 6: Comparison of Breakdown of Exergy Destruction

0

0.1

0.2

0.3

0.4

0.5

0.6

Al-Khobar Al-Jubail II Al-Khafji Jeddah II Jeddah III Jeddah IV

Exer

gy D

estr

uctio

n Fl

ux (

KW

/m2 )

Recovery Section Rejection SectionBrine Heater

Figure 7: Breakdown of Exergy Flux Among Major Subsystems