Genvinding af afvaskningsmidler i den grafiske branche

Summary and conclusions

The background for this project is the widespread desire of the graphic trade to minimise their consumption of resources and the costs for the cleaning of offset printing machines.

The cleaning process includes cleaning of ink rollers, rubber cloth, ink box, etc.

Detergents used for the cleaning are typically mineral-based or vegetable detergents. After the cleaning process, remnants of detergent have to be removed.

The removal of detergent remnants is done either manually or automatically in a dry-cleaning process or by rinsing with water. Manual cleaning is mainly done as a dry-cleaning method in small printing units.

Cleaning of large machines is done automatically. The technique applied is usually rinsing with water. The water quantities applied are 5 to 10 times as big as the detergent quantity. During this rinsing process a waste emulsion is generated, which is collected and transferred to Kommunekemi. The waste emulsion generated consists of water, detergent and "dirt".

The development tends towards an increase in the number of printing machines equipped with automatic washing systems.

Project objective
The aim of this project is to recover the detergent contained in this waste emulsion and re-process it. Simultaneously an attempt was made to minimise the liquid fraction by cleaning the water from the waste and reusing it in the washing process.

It has been attempted to develop a regeneration method, which is widely applicable for most detergent waste in Danish offset printing works.

Course of the project
The first phase of the project should determine the market potential (number of companies and quantities used) and the types of waste (detergent types) generated. This investigation revealed that the project target group includes some 200 printing works.

A review of literature and data sheets collected by GA was carried through during the introductory project phase. This review showed that the detergents could be divided into three different kinds, i.e. mineral-based detergents, vegetable detergents and mixed products. This is quite in line with the results obtained through the Graku project "Detergents for Offset, 1997-98".

Two detergents - a vegetable and a mineral-based detergent - were selected in order to find the most applicable regeneration method.

The product "Solren" was selected as a representative of the traditional mineral-based detergents. Solren was selected because it has been available on the marked for many years and still is prevailing.

Wash & Clean 2000 was selected to represent vegetable detergents. Being a rather recently developed product with major content of surface active substances, Wash & Clean 2000’s mode of operation differs greatly from that of Solren. If its content of surface active substances is not taken into account, its basic composition is almost identical with several others, somewhat older detergents. Thereby, this selection of detergents covers a wide range of available vegetable detergents.

Two printing works participated in the project in order to allow assessment of and choosing the "correct" cleaning method. The printing works were Phønix and Levison & Johnsen & Johnsen. The tests were made with waste emulsions from these two companies, and the pilot tests as well were performed at these printing works.

These two printing works do not only use the wanted detergents; they also both produce "swan-marked" printed matters, and both have implemented environmental management according to ISO 14001. Besides, both are verified by EMAS. This means that both companies are precursors in the field of environment and see the possibilities to improve their production by installing plants for the processing of waste emulsion.

Introductory laboratory examinations were made in order to assess the potential for recovery of detergents. The examinations showed a recovery potential of up to 20% of the waste emulsion as detergent and 80% as water.

Knowing this potential, the technical possibilities of splitting the waste into at least two phases were examined.

In order to break the emulsion a number of simple physical operations such as filtering, centrifugation and heating were tested.

Solren was split into two phases right from the beginning; therefore efforts only had to be made with Wash & Clean 2000. Heating was the final choice, because the method is uncomplicated, inexpensive and doesn’t require any chemical additives.

Once method and processing time for splitting the emulsion had been decided, various methods for cleaning of water and oil phases were tested.

By adding an appropriate amount of the clay mineral sepiolite, the two water phases could be cleaned till they appeared in a visually good quality.

If this treated water shall be discharged into the sewer before it is mixed with other wastewater, the authorities would only approve its quality after it had passed through an activated carbon filter.

The oil phase is cleaned by adsorption on activated alumina, as this material proved the most suited for cleaning of both types of oil phase (detergent).

Based on the examinations and assessments made, a pilot plant was designed and built.

This plant was built at MILJØ-KEMI, where the first tests, adjustments and changes were made. The tests resulted in an "optimised" process, and results and theoretical values obtained from laboratory analyses were confirmed in pilot scale.

The pilot plant was transferred to Phønix. All regenerated detergent was reused without problems. The purified water phase as well was reused without problems.

The conclusion arrived at in the pilot tests at Phønix was that Wash & Clean 2000 could be reused without problems in a ratio of 75/25 with fresh goods, and that the water could be reused in a ratio of 50/50. At the same time the waste generated during the cleaning process is reduced by 70%.

After the pilot tests at Phønix the plant was moved to Levison & Johnsen & Johnsen.

At LJJ as well all regenerated Solren could be reused without problems. No tests with reuse of purified water phase were performed at LJJ.

However, it may be expected that the same degree of reuse as obtained at Phønix can also be achieved at LJJ, as no special requirements are made for the water.

The conclusion of the pilot tests at LJJ was that Solren can be reused without problems in a ratio of 75/25 with new goods, and that the amount of liquid waste that is passed on to Kommunekemi is reduced by 85%.

A full-scale plant was dimensioned and prices for it were fixed on the basis of examinations and experience gained at MILJØ-KEMI and the two printing works.

Economy
A semi-automatic plant with a daily capacity of 250-750 l could be delivered at a price of approximately 160,000.- DKK.

This price and the operating costs stated by the printing works and the product suppliers made it possible to calculate a simple payback time for both types of detergent.

A simple payback time of 1.3 to 3.6 years is found for printing works of Phønix’ size and with Phønix’ consumption of Wash & Clean. The reason for this short payback time is the high price and the large quantities of vegetable detergents consumed.

Printing works of LJJ’s size and with their consumption of Solren will get a simple payback time of more than 10 years. The reason for this long payback time is the small consumption compared with the low price.

Conclusion
It may be concluded that a relatively simple method for the re-processing of the waste emulsions generated when cleaning offset printing machines was developed during the project. Thereby the project objective has been achieved.

With the developed method it is possible to handle and re-process both mineral-based and vegetable detergents.

Through the project it has been verified that recovered detergent and water can be directly used for offset printing machines without causing problems for operation and printing quality.

Pilot tests have shown that use of the developed method will not only reduce the environmental strain; it is also financially attractive for printing works that use vegetable detergents.

None of the two printing works wanted to install the final full-scale plant. Both had their reasons why they didn’t want it, but the result is that it has not been possible to build and install a final full-scale plant within the framework of the project.

Two other technologies were developed during the project and are now commercially available. These technologies are assessed as applicable, but both are only applicable for certain types of detergents. The technology developed through this project can be used for all types of detergents. This makes it a technology that can be used much more widespread than the existing alternatives.

We estimate that plants for regeneration of detergent waste will be installed in the majority of large-scale offset printing works during the coming 5-10 years and that the technology developed through this project will be the natural choice for a number of these printing works - not least when the printing works replace mineral-based by vegetable detergents.