Waterbased cleaning of mixing vessels
6. Results of Tests at Manufacturers' Plants
6.1 Results of test program carried out in 1997
In 1997 laboratory tests and full-scale tests for water-based cleaning were carried out by Teknos Technology A/S in co-operation with a supplier company.
The aim of these tests was to develop a suitable technique for water-based cleaning and, based on the test results, to show the impact on the environment and establish the economics.
The detailed results of the test program (screening of cleaning techniques, chemistry and full scale testing) are presented in Appendix B.
Laboratory and full-scale tests
The results can be summarized as follows:
 | In principle, water-based cleaning of equipment contaminated with paints and coatings is possible |
| Hot alkaline cleaning in combination with surface-active agents is applicable for a very wide range of products and raw materials |
| Removal efficiency can be promoted by employing, for example, mechanical cleaning equipment like brushes |
| Alternatives to hot alkaline cleaning are possible (e.g. acidic paint remover, neutral paint remover), but these as a rule will only work for a narrow range of products and feedstocks |
| Removal of paint and coatings by mechanical and thermal processes is ineffective in most cases and is uneconomic |
| Solvent-based removal is in general effective, but emission abatement has to be considered |
Because it was not possible to reproduce the positive results of laboratory tests during full-scale testing it was decided to repeat full-scale testing under different conditions.
6.2 Preliminary tests
The basic initial trials were conducted with ten problematic coatings from Teknos Technology A/S. The results of these trials at Renzmann and Riobeer are described in the following.
6.2.1 Results of preliminary tests at Renzmann
Preparations for trials
Several stainless steel plates and a mild steel vessel (800 l) had a coating applied to them from each paint system and these were allowed to dry under shop working temperatures. From this it is assumed that hardening of the coatings will be at the maximum possible.
The tests were conducted making exclusive use of a special cleaning agent (Stripper 303 G/E) with high-pressure cleaning nozzles.
For all paints with the exception of the isocyanate product, the cleaning results were satisfactory. The isocyanate specimen could be cleaned down to a small residue of paint (reaction between isocyanate and water) on the floor of the vessel.
Cleaning of the remaining nine products was successful for both mild and stainless steel.
The time for cleaning amounted to about 15 minutes.
Photos from the preliminary tests are shown in Appendix C.
6.2.2 Results of preliminary tests at Riobeer
Preparations for testing
Several stainless and mild steel plates were coated by each paint system, and allowed to dry for between 15 and 90 minutes under the working temperatures prevailing in the shop.
Test phase I soaking process
First the soaking process of each of the media was tested as described in the following. For reasons of environmental protection, the test using special cleaning agent was omitted; the action of this cleaning medium is in any case widely known from its use in practice.
| Solvents: Three different regenerates are employed with admixture of approximately 5% NMP (n-methyl-pyrrolidone). Although solution behaviors varied between them, for all three it was favorable and rapid. |
| Caustic soda (NaOH 60°C, 8% solution): The universal and favorable, but relatively slow, action of caustic soda was confirmed in the tested products. |
| Alkaline cleaning additive: For the trials, an alkaline cleaning additive produced in Switzerland was tested at a concentration of around 15% at 75°C. If cleaning is carried out within some 15 to 20 minutes, the results are acceptable, if not altogether complete. After 30 to 60 minutes, the action of this medium progressively drops off. |
For all tests, a known soaking behavior was observed.
Alkaline cleaning media were not employed for isocyanate containing substances, since firstly isocyanate reacts strongly with water and caustic solutions and, secondly, it is not permissible for moisture to be introduced into isocyanate production.
The solvent mixtures also tested demonstrated in all respects better solution results in comparison to the alkaline cleaning agents.
Test phase II jetting and brushing
During Test Phase II, hydrodynamic jetting and brushing were included in small-scale trials.
Despite high pressures, with solvents greater than 50 bar and alkaline solutions greater than 350 bar, with none of the media it was possible to attain a complete and satisfactory result within an acceptable timeframe. The results did not differ essentially from those of the tests conducted in 1997.
With mechanical support by the use of brushes, it transpired that by continuous rubbing of the metal surface, it is possible to achieve the required removal of residues within the specified time. It was possible to attain complete removal of dried on paint and coating residues in combination with both solvents and caustic soda.
It proved impossible to completely remove these residues with just the usual mechanically applied brush pressures combined with an alkaline cleaning agent.
With both solvents and caustic soda, it is possible to achieve rapid and thorough cleaning when supported by appropriately designed brushes.
6.3 Results of full-scale tests at Renzmann
During 2 days in December 1999 a combination of washing equipment and washing media was tested at Renzmann in Germany. The test includes equipment being traded by Renzmann and suitable for the special requirements of the customers.
6.3.1 Equipment used for the tests
The tests are conducted with 3 types of equipment:
Cleaning plant of the type "SKM"
It is a closed system for cleaning of vessels of different diameters. The cleaning is performed by means of a rotating spraying system on both the inside and outside of the vessel. Photos of "SKM" can be seen in Appendix A.2. Possible washing media are water, alkaline media (for instance sodium hydroxide) and solvents.
The plant has an integrated tank for cleaning agent (800 l) and exhaust of the residual cleaning agent, so that the vessel is left dry. The cleaning takes place during pressure (8-80 bar) and increased temperature (approx. 60-80oC) for water and caustic cleaning agents.
The system requires subsequent treatment of waste. By use of solvents distillation equipment is required. Wastewater cleaning/precipitation will be required by use of water and caustic cleaning agents. See chapter 7.6.2 concerning wastewater treatment.
"Robus" washing equipment with brushes
The system is designed for cleaning of the inside of the vessel and tanks with valve. The cleaning takes places with rotating brushes variable depending of type of vessel and tank. Photos of the "Robus" system is shown in Appendix A.2.
A tight adapter works as lid during the cleaning process. The system is designed for non-pressure cleaning with solvent (or water) and usually takes place at ambient temperature. A subsequent drying of the vessel is possible. The cleaning media recirculates and at some point it will require distillation equipment for treatment of waste (solvents) or wastewater cleaning/precipitation (water-based).
Pressure washer equipment - type "SP50 / SP80"
The system is for cleaning of the inside of containers and mobile tanks with valve. Photos can be seen in Appendix A.2.
The cleaning takes place by high-pressure washing at 50 bar by use of solvent and 80 bar by use of water as cleaning media. A tight adapter is required as lid adjustable for each type of tank/container. The cleaning media recirculates and at some point it will require distillation equipment for treatment of waste (solvents) wastewater cleaning/precipitation (water-based).
6.3.2 Cleaning media
For tests in full-scale the following 3 cleaning media are used:
| Stripper 303 G/E (from Foster Chemicals GmbH) Alkaline mixture of glycol ether and caustic potash/potassium hydroxide in concentration 5-6% |
| Sodium hydroxide solvent, 12% |
| Teknosolv 601 (from Teknos technology A/S) Mixed/multilevel thinner for cleaning |
The cleaning media are further described and assessed in chapter 7.2.
6.3.3 Tests and test results
Based on Renzmann’s experiences fixed parameters have been chosen for each plant. Moreover, the tests have been conducted with combinations of vessels in stainless/plain black steel, lacquer film dry/wet, washing media and washing time. Dry lacquer film is after drying/hardening for 5 days. The numerous combinations can be grouped according to type of plant in series tests a, b and c.
Each combination has been tested on a number of lacquer and paints chosen by Teknos Technology A/S. The chosen types are stated together with the test results in table 6.2.
Water-based cleaning (Stripper 303 G/E and NaOH-solvent) is carried out in cleaning plant SKM. For comparison the same lacquer and paints are cleaned by means of equipment with brushes or pressure washing by use of mixed/multilevel thinner.
Besides use as standard of comparison for water-based cleaning the tests with multilevel thinner can inform something about efficiency of different pressure and cleaning times.
The test combinations are shown in table 6.1.
Table 6.1 Look here!
Test parameters and variants
Table 6.2 Look here!
Test results
6.3.4 Assessment of test results
Water-based cleaning with alkaline special product (Stipper 303 G/E) is generally suitable. For isocyanate it is not absolutely suitable, however it should be noted that isocyanate generally is difficult to clean which is also the fact with multilevel thinner, cf. the test results.
As for a large number of the investigated types of lacquer the NaOH-solution is suitable. However, there is a demand for longer washing time than with Stripper 303 G/E. For the types of lacquer with limited efficiency, similar or other problems appear also by use of multilevel thinner. For some of these types of lacquer solution-based cleaning demands at the same time a mechanical impact.
The NaOH-solution is somewhat cheaper than Stripper 303 G/E both as to buying and as to subsequent treatment/removal of waste.
In case complete or partly cleaning with solvents is chosen tests show that the best efficiency is reached with a mechanical impact with brushes instead of pressure washing. A plant like "Robus" will not reduce the VOC emission considerably compared to known and more manual cleaning methods, however depending on the construction of the plant an improvement of the working environment may take place.
Each company has to assess relevant cleaning methods and cleaning media based on types of lacquer. It might be necessary to choose more methods and media, in case the product mix of the company is very different and varied. After choice of method the cleaning time can be optimized to each type of lacquer.
It would also be relevant to let existing types of vessels, materials and sizes be part of an overall assessment. Valves may demand additional cleaning. A general replacement of vessels will be very costly.
Chapter 8 shows a calculating example of the economy by using water-based cleaning in the "SKM" plant.
Subsequently advantages and disadvantages for the tested plants are stated.
A – Cleaning plant of the type "SKM"
Advantages:
| Efficient cleaning of the vessels |
| Can be used with vessels of different sizes. However, it will be relatively costly to clean very small vessels |
| Closed system, considering the working environment |
| The vessel is completely dry after the process |
Limitations/disadvantages:
| Possible problems with cleaning of the valves |
| The process leaves invisible, basic film on the vessel – requires increased amount of rinsing water |
B – "Robus" washing equipment with brushes
Advantages:
| Satisfactory cleaning with the most solvent–born lacquers |
| Cleaning of the valve |
| The process runs at ambient temperature and atmospheric pressure |
| Can be extended with equipment for drying after cleaning |
Limitations/disadvantages:
| Problematic as for air-drying lacquers and partly as for isocyanate reactor |
| The system is not completely tight |
| Can only be used for vessels/containers with valve |
| Cleaning at the top edge is unsatisfactory |
| Loose residues of lacquer at the bottom of the vessel is difficult to flush out |
| Brushes shall be chosen/replaced according to the size of the vessel |
C – Pressure washer equipment – type "SP50/80"
Advantages:
| The process runs at ambient temperature |
| Cleaning of valve |
| Can be extended with equipment for emptying after cleaning |
Limitations/disadvantages:
| Not suitable for all types of lacquers |
| Can only be used for vessels/containers with valve or manhole |
| Requires adapter adjustable for each type of vessel |
| Cleaning at the top edge is unsatisfactory |
| Vessels and containers of plastic may be problematic due to static electricity (when using solvent as washing medium) |
| Loose residues of lacquer at the bottom of the vessel is difficult to flush out |