Fouling phenomena in multi stage flash (MSF) distillers

Fouling in multi stage flash (MSF) distillers has been occupying researchers for many years. A lot of work has been done and more is yet to come in order to fully understand the role of various components and their interaction including the effectiveness of scale control techniques. In this paper an attempt is made based primarily upon visual and reported observations of fouling in various parts along the flow path of brine solutions in MSF distillers. This analysis is aimed at proposing certain sequence of scale forming reaction steps and to suggest certain experiments that could verify the validity of the proposed reaction mechanism. The proposed reaction steps are shown in alphabetic order to denote the sequence that could prevail inside heat transfer tubes of recovery and heat input (brine heater) sections and water boxes and thereafter in flash chambers including demister pads of MSF distillers. To predict the fate of various species and pH values one must identify reaction steps. The common practice is that: pH values are measured of solutions drawn out of heat exchanger tubes of recovery section or brine heater. This leads to the evolution of carbon dioxide. In view of the above, variation in pH values which could support the proposed mechanism were never addressed. Certain experiments should, therefore, be devised where pH values of recirculating brine inside heat exchanger tubes could be measured on-line while still under pressure. It would then be probable (especially if particular species detection electrodes are installed) to measure variation in pH values due to hydronium ion generation and its consumption. Meanwhile, the presence of hydroxyl ions will be short lived primarily due to Mg(OH)2 precipitation. The proposed steps support CaCO3 precipitation ahead of Mg(OH)2. The abundance of magnesium ions and the extreme low solubility of magnesium hydroxide will rapidly then lead to its formation. However, scale precipitation inside tubes are not only from initial scale formation under pressure inside the tubes but also due to nucleates recirculation from flash chambers back into heat gain exchanger tubes because of brine recycling. Recent analysis of variation in coloration of water boxes came as a strong support to this hypothesis. Reaction steps suggested by other workers will be shown for the sake of comparison. It is worth to note that the end results of various reaction mechanisms are almost the same in the cited ones also the same to the proposed overall reaction.

Scale and sludge formation have negative impacts on heat transfer, pressure drops, as well as flow starvation and above all deterioration of distillate product water yield and quality. Such formation can by no means be stopped or totally eliminated. Even though, safe prolonged operation can be achieved by bringing deposition of sludge and scale under control [9]. For such control to be effective deposition mechanisms, causes and pasevation techniques are to be clearly understood. This understanding would be the primary initial step in any full sludge and scale deposition control program. The proposed mechanism suggests that the formation of low temperature instead of the so called alkaline scale inside heat gain exchanger tubes of heat recovery section and the brine heater could proceed without the need for carbon dioxide generation step to take place. By this, one can explain scaling inside heat gain tubes even where the re-circulating brine is under pressure and the CO2 generation step is selflimiting in the absence of gaseous CO2 release off the re-circulating brine solution. Should the proposed mechanism is proven to be correct then the so called alkaline scales are to be renamed as suggested above. Furthermore, sludge formation is primarily due to • salt growth around antiscalant molecular chain, • hydrolysis of antiscalant molecular chains • silt and corrosion product polarized species accumulation on the molecular chain of antiscalant material. All these types of build-ups on antiscalant molecular chains would render the additive less effective. It is, therefore, important to monitor residual antiscalant potential (RAP) [1]. Taking into consideration that the suggested mechanism and the deliberated upon causes will lower production cost through reduced use of consumables, e.g. antiscalants, cleaning balls and acids; plus better prolonged operation thus shorter down time hence minimized production losses and longer plant productive life. Moreover, improper approach to ball cleaning may also give rise to sludge and/or scale deposition inside heat transfer tubes, which would also add to the above negative impacts. One of the primary observations, which this analysis would put forward, is that the carbon dioxide evolution step (E) could have very little effect on scale formation inside heat exchanger tubes. This proposition is put forward in order to give way to explain scale formation even if CO2 gas is not released where its reaction is suppressed due to its concentration. Reaction steps suggested in this paper differ from other proposed mechanisms in the sense that scale formation inside heat exchanger tubes can proceed in an acidic media (see reaction step A) without the need for CO2 generation step to be occurring. That is to say, the sequence is pre- dominantly ending with step (D) before any appreciable CO2 is released off the flashing brine, i.e. in flash chambers where carbon dioxide generation hence evolution is controlled by reaction step (E). In order to verify the validity of the proposed mechanism some well thought-out experiments are to be conducted. The most detrimental checks on this validity verification are felt to be by the results of pH measurements and detection hence the fate of various species. The commonly introduced concept referring to ‘positive LTA as scaling yard stick’ is to be replaced with ‘positive LTA is not a confirmation sign of scale formation but it is only an indication of increased scale formation potentials’. 4. Recommendations 4.1. Process control Based on this review certain process parameters are to be placed under scrutiny, hence closely monitored and controlled within clearly specified ranges. These parameters are outlined hereafter. 4.1.1. Top brine temperature (TBT) Reference is to be made to the range given in Fig. 1 of 90 to 115 ºC. 4.1.2. Antiscalant Dose Rate Again reference is made to Fig. 1. In this respect, it is recommended that the proposed optimum dose rates are to be targets that could be reached through careful investigation at each site based on seawater and plant design specifics. It is also recommended to maintain a sufficiently healthy safety margin of antiscalant, above minimum requirements, i.e. to apply an optimum rather than the minimum required dose rate. In order to coupe up with variations in conditions of the plant and the environment, some of the due most crucial changes, which deserve close attention, are: 1. Antiscalant preparation and dosing system malfunction 2. Make-up flow rate 3. Brine recycle flow rate hence brine tube velocity 4. Malfunction in ball cleaning system 5. Sea roughness hence seawater turbidity 6. Steam de-superheating, flow and temperature hence the TBT 4.1.3.Residual antiscalant potential (RAP) RAP has not become a common controlling parameter because it would require elaborate and tedious chemical analysis, yet it is felt to be quite detrimental. Its adaptation as a monitoring control parameter is strongly recommended. 4.1.4. Ball Cleaning Philosophy It is strongly recommended to upgrade the understanding of ball cleaning aspects. Misconceptions, e.g. high ball to tube ratios, and ball per tube per day are to be abolished. 4.2. Future work As a result of this analysis a number of tests are recommended. These tests are to be aimed at verifying the proposed reaction steps and their sequence. The proposed tests can be divided into three categories as shown below. 4.2.1 Laboratory testing 1. Bench top tests to verify the validity of the proposed mechanism by detection of various species and more importantly pH value measurements. 2. Pressurized plug flow device is to be designed with proper selective ion electrode detection and pH value measurements to study reaction constants, temperatures, pressures and time dependence in order to verify the proposed mechanism at various ionic species and antiscalant concentrations. 4.2.2.Pilot plant testing Some tests to verify points raised in 4.2.1 above are to be devised on a pilot plant MSF distiller. In Fig. 2a locations of additional instrumentation are identified by crosses. 4.2.3. Commercial plant testing The third phase of the work is to be carried out on some commercial MSF distillers. They are to be equipped with essential components to carry out tests of similar nature to those proposed in a pilot plant (in 4.2.2 above). Fig. 2b shows monitoring locations for the proposed tests as marked again by crosses.