Features Equipment & Systems
CONTROLLING BLACK LIQUOR SPILLS

A detailed study of the mill is required to assess the significance of unplanned liquor dischargesBy Neil McCubbin

Asignificant proportion of the effects of effluents discharged by kraft mills has always been due to unplanned discharges of black liquor, usually in concentrations that are low relative to the values normal in the pulp mill or recovery areas.

Today, discharges of organochlorine have been reduced to below the level of scientific significance in most Canadian and virtually all overseas mills, and all Canadian mills except one have effective biological treatment of all wastes. However, the effluent from many mills is still toxic to fish at sub-lethal levels, although the extent and significance are open to some debate. The effluents are also highly colored, which causes offence to users of some waterways, as well as leading to concern on the part of the public as to whether the effects of the effluents are minor.

WHAT ARE “UNPLANNED DISCHARGES”?

Mills discharge a certain quantity of pollutants continuously, by design, including bleach plant filtrates, cooling tower blowdown, screen rejects and several other process streams that cannot be recovered and reused with the equipment and operating skills available at the mill. Mills also deliberately discharge various process fluids when storage tanks or equipment have to be drained for safe access during repairs.

All the above are “planned discharges”. All other discharges of effluents are “unplanned”. In the case of black liquor, these may include foam, leaks, failures in pump seals, tank overflows, unauthorized draining of equipment for repair, etc. These are often called “spills”, or “accidental discharges.” Generally, there are no direct regulations limiting unplanned discharges of water-based fluids in pulp mills, although the US EPA is attempting to exert some control with its 1997 BMP regulations, and impending COD discharge guidelines. Discharges of oils, and of strongly toxic substances, are generally prohibited.

The pulp industry has reduced both planned and unplanned discharges substantially over the 30 years I have been involved in it. The planned discharges are well known, and in most mills, the technology available to control the remaining discharges is quite expensive. There is relatively little published information on the unplanned discharges, and the level of control varies widely from mill to mill.

A detailed study of a mill is required to assess the significance of unplanned liquor discharges in any one mill, but the COD of the treated effluent provides a good first approximation for kraft mills. This is because the biological treatment system removes most of the COD from all sources (including paper machines) except those from the bleach plant and black liquor. Bleach plant losses, and planned black liquor discharges (such as washer losses) are constant, and fairly predictable.

The bleach plant will contribute about 2.5 kg COD per tonne (t) of pulp for each unit kappa number of the pulp entering the bleach plant to the untreated mill effluent. The biological treatment plant will reduce this by about 40 to 70%.

Most of the remaining COD in the treated effluent is due to black liquor discharges. The planned discharge from the brownstock washers and screening will correspond to roughly 1kg COD for each kg saltcake lost. This will be reduced to about half this value by biological treatment.

All remaining COD will be due to unplanned discharges. Actual values in mills range from about 1 kg/t to 40 kg/ton pulp. If unplanned discharges are not already minimal, by far the most cost-effective way of improving effluent characteristics is to reduce the frequency and size of spills. This will also reduce the cost of operating waste treatment systems.

HOW TO REDUCE THEM

In most cases, black liquor is the essential unplanned discharge of concern. The most important single factor in controlling spills is the knowledge of mill personnel at all levels. Once staff understands what spills are they will readily develop techniques for minimizing them, if management encourages such activities. To do this, the extent of spills must be measured daily, and discussed at the routine meetings on production, product quality, etc. In the past few years, I have noticed that several companies are introducing criteria for spill control into staff bonus schemes that also include elements of productivity, product quality and safety.

One physical requirement for spill control is that there must be a means of recovering black liquor that has to be removed from equipment for maintenance. This requires appropriate pumps and piping. Such systems are now quite common.

Recovery pumps must be installed in all sewers that may receive unplanned discharges of black liquor. These must activate automatically. All of the successful systems I have seen use conductivity sensors for this. Most systems activate recovery pumps when conductivity exceeds 5000 micro-mhos/cm. Almost all mills now have such systems for some parts of the mill, but few work really well because there is so much uncontaminated water entering the systems. The spills are diluted. This dilution causes significant black liquor spills to pass undetected, and often overloads the black liquor evaporators to the point that operators have to deactivate the spill recovery system. One of the hardest parts in designing a spill recovery system is keeping clean water out of it. This requires careful analysis but no space age technology.

I have seen spill recovery sumps not much larger than a garbage can, and others as big as a home swimming pool. I have seen recovery pumps up to 50 hp, with fully installed standby, as well as small pumps without a standby. Mills with very effective spill control often have small, simple, low-cost recovery pumps and sumps, but these are located to catch all potential spills, without recovering significant quantities of clean water.

It is normal practice to pump recovered material to a “spill tank”, for eventual routing to the evaporators or the blow tank. It is widely assumed, particularly by engineers designing systems to comply with the new US EPA regulations, that large spill tanks are desirable. Some are as large as 4000 m3. However, the best spill control systems I have seen are in old mills and have small spill tanks, typically less than 500 m3. Large tanks are liable to be used as a way of ‘A shift’ leaving an operating problem to ‘B shift’, instead of rectifying the problem.

ED. NOTE: This month good friends Neil McCubbin and Phil Riebel have switched jobs with Neil writing Life Cycles and Phil trying his hand at Input/Output (p. 79).