Protecting process CHILLERS

Chilled circuits are found in many processes in the chemical and pharmaceutical industries. For example, in the operation of reactor vessels they are used to control the speed of exothermic reactions, or to recover and purify the reaction product.

Reactor systems using chiller circuits include single-fluid systems with jackets, multi-fluid systems with half-pipe coils, and multi-fluid systems with jackets. And within these different types of systems, different types of thermal fluids may be used as the heat transfer medium.

Typical thermal fluids in use in the process industries include hydraulic oils, methanol or ethylene glycol/water mixtures, and brine systems based on the chlorides of calcium, magnesium or sodium.

The choice of thermal fluid is based on the range of system temperatures and the economics of operating the system - and, on both these bases, calcium chloride brine has proved particularly attractive in many systems. But there can be a drawback. The major disadvantage with brine systems is the severe corrosion that can be experienced in the chiller circuit.

Traditionally, this problem has been addressed by the use of corrosion inhibitors such as chromate-, silicate- and nitrite-based formulations, with the chromate-based treatments generally considered to give the best results. In most cases, buffered chromate inhibitors in the range of 500 to 1000ppm are satisfactory unless bimetallic influences exist, in which case higher chromate levels are needed to control corrosion.

The nitrite-based inhibitor is another widely accepted closed cooling water treatment. Concentrations in the range of 2000 to 5000ppm suitably inhibit iron and steel corrosion when the pH is maintained above 7.0. However, to protect mixed metallurgy systems combinations of nitrite/borate/silica are required.

The major drawback of chromate treatments is the unacceptable toxicity. In closed systems using calcium chloride brine, the only traditional treatment is 1250ppm chromate with the pH adjusted to 7.0-8.5 with caustic. Some successes have been recorded with nitrite-based treatments of closed brine systems, but at levels in excess of 2000ppm nitrite.

While technically the use of chromate can control corrosion in calcium chloride brine systems, from an environmental view its use is really unacceptable. Its toxicity at high levels precludes its use even in closed systems that are only drained infrequently.

Until now, there has been no environmentally acceptable inhibitor for such systems.

But, as the case studies in the panels on this and the preceding page indicate, BetzDearborn has now successfully addressed the problem with its Brine Control System.

This new control system for chiller circuits is a custom blend of environmentally acceptable organic materials that is specifically designed to control corrosion and scale formation in calcium chloride brine systems.

Electrochemical tests have shown that the brine control system provides both anodic and cathodic inhibition. This is clear from the polarisation curves shown in Figure 1, which show a reduction in both anodic and cathodic currents for mild steel in the presence of the brine control system.

Application of the system provides excellent corrosion inhibition and deposit control. Corrosion rates of <1.0 mils per year (mpy) for mild steel, 0.0mpy for stainless steel and <0.1mpy for copper/nickel are routinely achieved.

Comparisons with the more traditional treatment programmes are shown in Figure 2, where it can be seen that the brine control system out-performs most of the less environmentally acceptable alternatives.

The treatment programme is all-organic and is environmentally acceptable - unlike those containing nitrite or heavy metals such as chromium, molybdenum or zinc.

Dr Alan Marshall is Corporate Market Manager of Chemical and Pharmaceutical Industries for BetzDearborn, a division of Hercules Inc, Widnes, Cheshire.

Case 1 - Nitrite problems

This site manufactured fine chemicals and pharmaceutical intermediates. It operated several calcium chloride brine systems treated with a nitrite-based inhibitor. Some problems were experienced with nitrite degradation and the maintaining of correct treatment reserves. Electrochemical studies indicated that insufficient nitrite had a negative impact on the mild steel corrosion rates in the brine systems. After a laboratory study using electrochemical and coupon testing (see opposite), the BetzDearborn brine control system was recommended to replace the nitrite system. Initially, a trial was carried out in one of the closed circuits. The programme was dosed to give a phosphate concentration in the range 240 to 350ppm. A copper passivator was dowsed at 5 to 10ppm to provide protection for copper and copper-based alloys and to complex the 6-12ppm copper present in the brine from the supplier.

The plant trial was a great success, reducing mild steel corrosion rates to <1.0mpy. There was also a significant reduction in the localised corrosion that had been noted when the nitrite treatment was in use. Corrosion rates on 316L stainless steel and copper were <0.1mpy - and the brine control system did not degrade as readily as the nitrite.

Case 2 - Chromate problems

A chemical plant had previously treated its CaCl2 brine system with a chromate-based treatment. For environmental reasons the plant decided to stop using the chromate treatment and operated the brine system untreated for a year. This proved to be a costly decision because four heat exchangers had to be replaced at a cost of $80,000. And there were significant other costs associated with labour, reduced efficiency and lost production.

Corrosion monitoring equipment installed on the brine systems showed mild steel corrosion rates to be about 80 to 90mpy, with evidence of localised corrosion in the form of pitting and crevice corrosion. Soluble iron and copper levels in the recirculating brine were also very high.

It was recommended that the brine control system be introduced for ferrous metal corrosion control, supplemented with a copper passivator for the control of copper deposition and the control of copper and copper-based alloy corrosion. Furthermore, a small continuous bleed from the system was introduced to reduce high levels of iron and copper in the circuit.

With this approach, it was possible to reduce the corrosion rates of mild steel down to <0.1mpy as shown opposite. There was also no significant evidence of localised corrosion after the new regime was established.

The significant benefits to the plant were: mild steel corrosion rates reduced 98 per cent; reduction in localised corrosion; equipment life extended; labour costs reduced; and production rates restored.