PTQ Q1 2024 Issue

results warranted a change in the microbiological control programme. To address the biofouling issue, Solenis recommended implementation of its biofilm detection and control pro- gramme on one of the refinery’s cooling towers on a trial basis. The recommended chemistry for the trial was the patented, in situ stabilised active chlorine solution. After the six-month trial, the general corrosion rate was cut by a factor of three and the pitting rates were cut by a factor of two. The refinery’s leadership decided to adopt the biofilm detection and control programme for all of its cooling towers. General corrosion rate and corrosion pitting, measured by metal coupon testing, and average weighted wall loss, measured during heat exchanger inspections by non-de- structive testing, all showed dramatic improvement. After converting to the in situ stabilised active chlorine solution, corrosion rates of 0.2-0.3 mpy were achieved without pit- ting. Eddy current testing data collected before and after the implementation of the in situ stabilised active chlorine solution showed a decrease in heat exchanger wall loss of 45%. This loss corresponded to a 50-80% increase in bundle life. Solenis continued to monitor heat exchanger flows and cooling tower efficiency. Monitoring of the heat exchangers revealed that exchangers that historically had lost flow rapidly shortly after a turnaround now maintained their start-up flows. This improvement was validated dur- ing annual flow studies, as shown in Figure 5 . Furthermore, the biofilm detection and control pro- gramme effectively eliminated algae on the cooling water return hot decks. Prior to implementing the programme, even with aggressive doses of conventional biocide, algae covered the hot decks, short-circuited the cooling tower fill, and drove up supply water temperatures, resulting in pro - duction rate reductions, until the hot decks were cleaned. After the algae build-up was removed by the new in situ stabilised active chlorine solution, the hot decks remained clean and the supply side approach to wet bulb tempera - tures immediately improved by -17.2°C to -16.7°C (1-2°F). Towers with high-performance fill experienced the great- est gains. The approach to wet bulb readings were moni - tored closely for three years. In the first year, the approach to wet bulb temperatures decreased by -16.1°C (3°F) and in the third year by almost -13.9°C (7°F). The colder water flow to the process improved vacuum on overhead exchangers, resulting in production gains with only a neg - ligible increase in operational expense. Additional data analysis would corroborate the evidence of improved per - formance and profitability. Next, Solenis analysed the data and determined how many heat exchangers required cleaning outside of turna - rounds and how many experienced failures before and after the implementation of the biofilm detection and control programme. If biofilm caused the fouling, then the imple- mentation of the programme should result in fewer heat exchanger cleanings during production runs. As shown in Figure 6 , the number of heat exchanger cleanings outside of turnaround decreased by 89% with the programme. If biofilm causes corrosion, fewer heat exchanger fail- ures should result from improved biofilm control. The data

Biolm detection & control program me

Turnarounds

Measured velocity (ft/s) Design velocity (ft/s)

used the modelling capabilities of the HexEval performance monitoring programme to categorise the heat exchangers at risk of developing biofouling, scale or both. Prior to the implementation of the performance monitoring programme, American Petroleum Institute (API) guidelines were in use. The guidelines identified at-risk exchangers as having a water velocity less than 0.91 m/sec (3 ft/sec), a cool - ing water outlet temperature greater than 48.9°C (120°F), and a process inlet temperature greater than 60.0°C (140°F). According to these guidelines, 233 of the 400 exchangers in the refinery were at risk of developing deposition. Managing the risk to 233 heat exchangers would have been a daunting task. However, the Marathon engineers used the heat exchanger performance monitoring pro - gramme to calculate each exchanger’s HSC value. The HSC assesses the reliability of each heat exchanger and identi - fies factors threatening their performance. The higher the HSC value, the greater the risk of deposition. An HSC value less than 2.0 identifies a low risk, and a value greater than 2.0 identifies an increasing risk of biofouling or scale. The calculated HSC values reduced the number of at-risk bun- dles from 233 to 94. After identifying the 94 at-risk exchangers in the plant, the engineers concentrated on improving the mechanical aspects of the cooling system to reduce the overall risk of biofouling and scale formation. Better transient debris mitigation using improved tower screens combined with other mechanical modifications helped to reduce the risk of fouling in the at-risk exchangers and improved the overall performance of the cooling system. These modifications included flow balancing across the exchanger network, using a hot process bypass instead of throttling the cool - ing water flow, adding supply side jumpers for back wash assistance, introducing metallurgical changes, and making exchanger design changes. The number of at-risk exchang- ers was reduced to 37. Reducing the number of ‘bad actors’ from 233 exchang - ers to 37 exchangers brought focus to the problem. Recalculating the HSC values revealed biofouling risk fac - tors for 32 of the 37 problem exchangers. Clearly, these Figure 5 Heat exchanger water velocity before and after implementation of the biofilm detection and control programme

PTQ Q1 2024