Jan-2011

Drivers for zero discharge in refineries

Wastewater treatment utilising zero-discharge techniques has reduced the impact of effluent discharges

Gary Howard
Foster Wheeler

Article Summary

Recent years have seen an increasing focus on the environmental impacts of refineries and petrochemical plants, not only on reducing concentrations of contaminants in effluent discharges but also on overall pollution load and discharge volumes. Increasing demands on water consumption around the globe mean that existing facilities need to reduce water use. New facilities have to be designed to minimise water use and some to even achieve zero effluent discharge. This article looks at common drivers for reduced water usage. It reviews the technologies that are typically used and key challenges that have to be addressed if this strategy is to be successful. It will also look at two recent examples of how designs utilising zero discharge techniques have minimised the impact of effluent discharges.

Drivers
Refineries and petrochemical facilities use a considerable amount of water. Table 1 shows some typical values for a refinery and an ethylene cracker.

Typical plant water usage
Whole life costs are currently favouring evaporative cooling over seawater usage, increasing the typical demand of potable water. However, recovering water for reuse is an expensive undertaking, and this is driving the increased adoption of a range of site-specific water-recovery techniques.

The petrochemical industry is only one of many users competing for the existing water supplies.  Many industrial developments have to rely on water supplies from remote locations, or desalination of seawater, to provide reliable supplies. The increasing level of investment required to maintain reliable supplies is promoting a serious review of water usage in an attempt to reduce the relative cost of recovery.

At the same time as water scarcity increases, best practice with regards to environmental discharges is becoming more onerous. In some instances, political rather than sound environmental science can result in consents that are difficult to achieve with conventional 
treatment systems. Table 2 is a summary of various discharge consent criteria and the typical capability of a basic wastewater treatment process using a conventional activated sludge plant without tertiary treatments.

Consents and limits
It is important to understand the implication of the compliance level required. A maximum allowable concentration (MAC) needs to be met at all times. A 95 percentile basis allows one failure in every 20 samples. Often a 95 percentile basis will also have an upper tier requirement that is a MAC. In rough terms, the MAC equates to about four times the average, while a 95 percentile is about double the average. In Table 2, this means that the IFC and Saudi standards can be reasonably met by a conventional plant, while the Russian standards would need significant tertiary treatment or even zero discharge.

Uses of recovered water
Across an industrial site there are many operations and processes that require water. Good waste minimisation techniques look at elimination or direct reuse of water as the first step. An example of direct reuse in a refinery is the now common use of stripped sour water in the desalters. Once the water has undergone conventional treatment, it can be used in washing down dirty areas and for dust suppression in partially paved sites. Use for irrigation is possible if the sodium absorption ratio is low enough, although chloride is also an issue for some plants. If the total dissolved solids (TDS) level is low enough, use as cooling water feed is another option. Typically, these low-grade uses are insufficient to use up all of the wastewater produced on site and further treatment is then required. Once at a higher purity, use in the demineralisation system and unlimited use in cooling towers is common. One novel use that is being applied in Australia is irrigation of salt-resistant Eucalyptus. This salt-controlling plant uses brackish water and directs the salt to below the root system.

Technologies used
The processes used in water recovery up to a full zero-discharge plant are now established processes. A typical plant scheme is shown in Figure 1. This shows a typical industrial plant with pretreatment to protect a conventional effluent treatment plant. Water recovery is usually employed after the majority of the contaminants have been removed. A water upgrade step is used, which recovers high-quality water that is then reused in the cooling towers or to feed into the boiler system. The remaining fluid is reduced to slurry suitable for landfilling after further, more intense water recovery.

When recovery to high-grade water is required, the conventional treatment scheme is often enhanced to remove more contaminants. The processes that might be considered are:
•    Nutrient removal (eg, AO process)
•    Tertiary treatment (eg, DAF, sand filtration)
•    MBR (membrane bioreactor)
•    MBBR (moving bed bireactor)
•    PACT (powdered activated carbon treatment).

Following secondary treatment, the TDS need to be removed and this is usually achieved by either reverse osmosis (RO) or multiple effect distillation (MED). This step results in a concentrate that can be further concentrated by solar evaporation in arid climates, or by using evaporator/crystalliser systems. Both of these processes produce a slurry or solid that should be suitable for controlled landfill.

Challenges
The challenges with a zero or partial-zero-discharge system are essentially operational issues. The contaminated streams will become more concentrated as good-quality water is recovered. If we take a typical industrial effluent with a refractory chemical oxygen demand (COD) of around 50 mg/l, this is concentrated to an average 250 mg/l in a RO system recovering 80% of the feed. At this concentration, the effluent is no longer compliant with any of the discharge standards shown in Table 2. This means the concentrated reject stream needs further treatment.