The legislation being drafted by the Commission is still some way from being issued as a formal proposal. An initial draft was circulated to Member States at the end of last year (ENDS Report 202, pp 36-8). A revised version was discussed at meetings with industrial representatives and national officials early in April.
The revised version is similar in structure to its predecessor. The main text contains a series of general provisions, the most important of which is a duty on businesses using organic solvents above a specified quantitative threshold to prepare solvent management plans. There then follows a series of annexes which lay down specific emission control rules for particular industrial sectors.
Solvents contribute over 30% of the total man-made VOC emissions within the EC, which amount to 10 million tonnes per year. The Commission's first draft contained annexes on three sectors which between them account for more than 20% of the total solvent emissions. These are metal degreasing (445,000 tonnes per year), printing (188,000 tonnes) and the coating of new cars (136,000 tonnes).
Another three annexes have been added to the revised draft. These are dry cleaning (77,000 tonnes), impregnation of wood (47,000 tonnes), and the coating of metals, wood and plastics. According to a study carried out for the Commission by Warren Spring Laboratory (WSL), annual solvent consumption in metal coating amounts to 1.12 million tonnes across the EC. This includes 242,000 tonnes of cleaning solvents. Excluding these, the main consumers are as listed in the table.
The number of sites in each sector across the EC range from 25 producing wire enamels to 60,000 or more vehicle refinishing businesses. A number of sites in some sectors such as coil coating and drum manufacture already practise VOC abatement, but the vast majority do not. In some sectors, such as the coating of ships and structures, the only practical abatement option is reformulation of the coatings because most of the processes involved are carried out in the open air.
According to another WSL report, solvent consumption in wood coating amounts to 203,000 tonnes annually across the EC. To this can be added another 56,000 tonnes of cleaning solvent. Industrial coating of wood is carried out at 9,000 sites within the Community, the majority of which are in the furniture industry.
Overall, the new draft would cover industries responsible for some two-thirds of the EC's total solvent emissions. However, it would reduce these emissions by much less than this figure, at least in the short term. This is because many smaller users would be exempted from most provisions of the legislation, while businesses using low-solvent coatings would not be required to comply with the emission limits laid down.
The main text of the revised draft now contains a specific requirement for prior authorisation or "notification" of installations which use more than the threshold quantity of solvent prescribed in each annex (see below). "Substantial alterations" which increased the organic solvent consumption of existing processes by 25% or more would also require prior authorisation or notification. "Notification" is not defined.
A number of conditions would have to be fulfilled before a prior authorisation or notification could be issued. The competent authority would have to be satisfied that substances less harmful than organic solvents could not be used, and that all appropriate measures to minimise solvent consumption had been taken. Operators would have to submit drafts of their solvent management plans when applying for an authorisation.
The authority would also have to be satisfied that all emission limits would be complied with and that all relevant air quality standards had been "taken into account". Measures to avoid soil pollution, and to ensure that control of emissions to atmosphere did not result in additional water pollution - a distinct possibility if operators switched to water-based coatings - would have to be taken.
Finally, the authorities would have to be satisfied that all wastes contaminated with organic solvents would be collected separately, stored in closed containers, and delivered to waste treatment facilities.
The draft goes on to lay down provisions on solvent management plans. These would have to be submitted annually to the authorities. As in the initial draft, their objective would be to provide operators with the information needed to reduce their VOC emissions as much as possible, and to replace solvents harmful to health or the environment or which cause nuisance with less damaging materials.
A detailed specification for the solvent management plans is set out in an annex. This says that the plan should be based on two types of solvent balance - one for the complete installation, and one for part of an installation - but then appears to contradict itself by providing that the authorities should decide which of these balances is more appropriate in a specific case.
The plan would be based on an audit of the solvents purchased, stored in products, released to air, water or waste, recovered and/or reused, and leaving the installation in products. "As far as possible" the audit would have to be carried out for each solvent individually, but where this was not feasible the authority would have to order an audit of selected solvents "constituting a major part of the original solvent mixture and identified as representative for the usage and fluxes of organic solvents in the installation."
The amounts of solvent entered in the inventory could be "estimated by controls and measurements," whatever that means. The only exception would be for fugitive releases to air, which could be estimated by deducting all other outputs from the solvent inputs.
At this month's meetings with the Commission, several delegations appear to have expressed concern about the complexity of the solvent management plan, in particular for small businesses. The details may be changed once the Commission has received a further report on how these plans may be compiled from WSL and Coopers & Lybrand.
However, it should be noted that solvent management plans and the need to obtain authorisation are the only short-term requirements to be imposed on smaller businesses - generally those using 1-5 tonnes of solvent annually. Because these operations may account for a sizeable fraction of the VOC emissions from each sector, the Commission may be reluctant to water down its current proposal too far.
Last year's initial draft contained an annex which set out a hazard classification for more than 80 solvents. In turn, this formed the basis for VOC emission limits which varied according to the ranking of the solvents concerned.
The classification is used in a different way in the revised draft. In complying with the duty to switch to less harmful solvents, operators would now have to use the hazard classification simply as a guide. Thus a business using a class I solvent - the most hazardous to health or most liable to contribute to photochemical pollution - should switch to a class II or III substance, while a business using a class II solvent should switch to a class III material. Where a solvent did not feature in the list given in the annex, it would be for the authorities to assign it to the relevant class on the basis of its behaviour and effects in the environment.
The only exceptions to this approach are 16 solvents considered to have a high potential to harm human health. Operators using any of these would have to switch to less harmful substances "with the shortest possible delay." In addition, tight emission limits would apply to the 16 substances, as follows:
Several other general provisions are laid down in the main text. Existing installations would have five years after the Directive comes into force to comply with the VOC emission limits laid down, although in some cases different periods are laid down in the sector-specific annexes. Fugitive emissions are to be kept "as small as technically and economically feasible," and guide values - ranging from 10% to 15% of total solvent inputs - are laid down for most of the sectors.
Another general clause provides that emissions should be collected and discharged "as far as possible" via a stack. The authorities would have to ensure that the stack height was sufficient to safeguard health and the environment and to "prevent" nuisance at ground level. These provisions could have major repercussions for many businesses discharging solvents through a multiplicity of low-level vents.
On monitoring, the draft would require continuous measurement of emissions of more than 100 kilograms of solvent per day, and periodic measurement - at a frequency to be determined by the authorities - of emissions of more than 10 kilo-grammes per day. The results would have to be kept for five years, and be available to both the authorities and the public.
The compliance conditions are different from those in the original draft. Where emissions are subject to continuous measurement, compliance with the emission limits laid down in the annexes would be achieved if none of the daily means exceeded the limit, if 97% of all half-hourly means did not exceed the limit by more than 20%, and if none of the half-hourly means exceeded the limit by more than a factor of two. For periodic measurements, the limits laid down are absolute values.
Operators would be obliged to notify the authorities if any limit was exceeded, and they could not continue to operate a process as long as the breach was likely to continue.
Similarly, if an authority concluded from the annual solvent management plan that the limits laid down for fugitive emissions had been exceeded by up to 50%, the operator would have a maximum of three months to come into compliance before being forced to shut down. The draft does not say what should be done if the limits were exceeded by more than 50%.
Another significant change from the initial text is that the VOC emission limits are no longer expressed in terms of milligrams of carbon per cubic metre. Instead, they are expressed as a mass emission of solvents, to be calculated by summing the mass of each individual solvent emitted. This is important because under the former approach - which has been used in the UK in setting limits on VOCs for processes regulated by local authorities under the Environmental Protection Act 1990 - the mass concentration of individual solvents can be way above, say, 50mg/m3 yet still comply with a limit of 50mg/m3 expressed as carbon.
A number of general amendments have been made to the annexes. One of these introduces a quantitative threshold in terms of annual solvent consumption at which installations in a particular sector will be subject to the Directive. The UK regards this as more satisfactory than the initial draft, which contained other thresholds such as annual solvent emissions or, in one printing sector, ink consumption.
For most sectors, the threshold is one tonne per year. As noted above, installations using 1-5 tonnes of solvent annually will be subject only to the requirements to compile a solvent management plan and obtain prior authorisation or "notification". These provisions will apply for 10 years after the Directive enters into force in the case of existing installations, and for five years for new installations. Seven years after it enters into force, however, the Commission will report on specific measures to reduce VOC emissions from these smaller installations as well.
For most sectors, the draft also introduces the idea of "compliant coatings" or cleaning solvents. This provides that where a coating or cleaning agent contains less than a specified quantity of organic solvent, the authorities may give a derogation from the emission limits laid down, although in most cases a more relaxed limit is prescribed as a safeguard.
The thinking behind this approach, which has been used in most of the guidance notes dealing with VOC emissions from processes regulated under the Environmental Protection Act, is that it will encourage innovation in low-solvent products as an alternative to potentially costly abatement devices which, in the case of incinerators in particular, have the added disadvantage of consuming energy and increasing emissions of carbon dioxide.
However, the Commission has taken much of the gloss off these derogations by stipulating that they may in general last no more than 2-3 years. This clause is likely to be strongly opposed by the UK, which regards it as almost certain to discourage the development of low-solvent coatings and, indeed, encourage solvent users to invest in incinerators and other VOC abatement devices rather than adapt their processes to low-solvent materials.
The Commission's view, however, may be that short-term derogations are essential to promote the development of water-based coatings and other products which contain genuinely minimal levels of organic solvents. Water-based products currently on the market can contain 20% or more of organic solvents, and are only described as such in relation to conventional products which contain 50% or more of organic solvents. This is reflected in the temporary higher emission limits proposed for them, which range up to 700mg/m3. The arguments over this issue are likely to be crucial to many industries.
The emission limits and other requirements for the sectors dealt with in the annexes are too detailed to summarise here. However, the basic VOC emission standards are as follows:
In some sectors other requirements will be at least as significant. For example, in the car manufacturing sector, the revised draft now has an overall VOC emission limit of 35g/m2 of the car body - almost half the level recently introduced in the UK - or 2.3 kilograms per vehicle as an alternative. Sharp exchanges can be expected on this issue.
Likewise, operators in the wood, metal and plastic coating and wood impregnation sectors using more than five tonnes per year of material containing more than 10% of organic solvent by weight will have to demonstrate to the authorities that the coating and drying processes cannot be carried out under contained conditions in order to facilitate solvent abatement. These provisions may well require businesses to invest heavily in covering or containment of operations traditionally carried out in the open.