At its Dingolfing production site, BMW has established a nickel- and nitrite-free phosphating process that allows it to reduce the nickel concentration in the wastewater to less than 0.2 ppm...

For almost ten years, trication phosphating has been the state-of-the-art technology for car-body pretreatment. Applying a thin phosphate layer to the cleaned metal surface improves adhesion and creepage resistance of automotive paint systems. Automotive paint systems consist of a cathodic electro-paint, primer surfacer and topcoat.

A characteristic of tri-cation phosphating processes is that the phosphating solution and, consequently, the deposited phosphate layers, contain zinc, nickel and manganese. As a rule, nickel concentrations in the phosphating bath vary from about 400 to 1,000 ppm. The deposited layers contain approximately 0.5 to 1.2 pct nickel, depending on the substrate to be treated.

The 40th AvW (German Administrative Regulation on Waste Water), annex 40, prescribes an admissible maximum nickel concentration in wastewater of 0.5 ppm. In individual cases, an even lower nickel concentration may be specified by the local water management authorities.

For BMW's Dingolfing production site, the maximum limit was 0.2 ppm. To comply with this specification, either a selective ion exchange or a nickel-free phosphating process had to be installed. BMW and Oakite Products, Inc., a division of Chemetall Group, Berkeley Heights, New Jersey, developed a nickel-free, low-zinc phosphating process that was implemented in January 1996 in the two car-body plants at Dingolfing.

Preparation of the plants for the changeover.

The two plants to be modified were 11-zone, fixed-cycle, vertical-immersion (VERTAK) plants (Table I, below).

All pipes, pumps and treatment tanks in the phosphating section had to be thoroughly cleaned to prevent the nickel-free phosphating solution from being contaminated by nickel from back-dissolution of sludge deposits and incrustations. All relevant pipes were dismantled and cleaned at high pressure (800-2,500 bar) and rinsed with acid. The same procedure was applied to the treatment tanks, settling tanks and desludging system. The thorough cleaning of the phosphating section contributed to the successful changeover. Immediately after the process modification, the residual nickel concentration of the phosphating bath was less than five ppm and has not changed significantly since then.

Nickel- and nitrite-free phosphating.

The nickel- and nitrite-free phosphating process is based on layer-forming cations of zinc, manganese and copper, with copper used only in traces and hydroxylamine serving as an accelerator. To prevent the formation of "white spots" on galvanized steel, the phosphating solution contains complex fluorides.

The nickel-free process is used in both VERTAK plants with very similar parameters. It is designed for layer-forming phosphating of aluminum and makes use of free fluorides.

The nickel-free phosphating process features application properties that are similar to those of the nitrite-accelerated and nickel-based process. This is reflected in that it was possible to maintain almost the whole pretreating process. Some of the characteristic data of the phosphating process are given in Table II (below).

Copper ions as nickel substitute.

Traces of copper (II) ions are used as a substitute for nickel. As for copper, the 40th AvW, annex 40, specifies a low limit value, less than 0.5 ppm. Due to the very low application concentrations (approximately five ppm) in the phosphating bath, this level was already met in the rinsing bath after phosphating, without any additional measures.

The positive influence of copper ions has been known for some time in phosphating technology. The innovation here lies in close concentration monitoring and precise copper metering. Even though it is not yet clear how the added copper works, it can be assumed that, depending on the material, part of the copper is integrated in the phosphate layer and another part cements on the free metal surface as copper oxide.

The process is suited for phosphating aluminum surfaces, too. Modifying the phosphating solution by adding copper ions does not lead to the formation of a galvanic element and, therefore, to accelerated corrosion of the aluminum.

The copper contained in the phosphating solution is determined by photometry as a bathocuproin disulfonic acid complex. Measuring takes only a few minutes and has eight pct relative accuracy. Adding copper via the replenisher leads to a constant bath concentration.

Hydroxylamine as a nitrite substitute.

In the nickel-free phosphating process, hydroxylamine is used as an accelerator. It is added directly with the make-up solution. The process works on the iron (II) side, the iron (II) content being limited to less than 100 ppm for practical reasons.

Under certain conditions, especially with little bath movement (cavity), hydroxylamine is somewhat slower than sodium nitrite as an accelerator. In this process, the lower reaction velocity is compensated for by adjusting a somewhat higher zinc content (1.9 plus or minus 0.1 g/liter), without having to accept any loss in quality (adhesion and creepage resistance).

Due to the reaction with the metal surface during phosphating, ammonium forms as a hydroxylamine decomposition product according to the following equation:

NH₃OH⁺ + H₂ --> NH₄⁺ + H₂O

In the application described, the concentration of ammonium ions in the phosphating bath in equilibrium is approximately 1.5 g/liter.

Stability of the accelerator.

Practical application has shown that hydroxylamine has a measurable self-decomposition rate that is higher than similarly accelerated nickel-based processes, but clearly lower than that of sodium nitrite.

The copper, iron (II), temperature, bath movement, sludge and free-acid content have been found to be factors that influence the hydroxylamine decomposition rate in the system. In the Dingolfing plants, the accelerator self-decomposition rate is approximately 10 pct relative/day. For example, at constant temperature with no throughput, the accelerator content decreases by approximately 10 pct of its initial value per day. At lower bath temperatures, the self-decomposition rate is lower. In plants with a high throughput, hydroxylamine self-decomposition leads to a moderate accelerator consumption. The situation is different in plants with a low throughput and those just commissioned with a flat starting curve.

Process control and bath monitoring.

The bath is monitored in part by automatic titration (zinc, free acid and total acid) and in part by manual titration (hydroxylamine content) or photometric determination (copper (II) content).

The resulting phosphate sludge has a crystalline consistency and must be converted, by adding a sludge conditioner, into a more easily removable state with no formation of solid crusts.

The sludge content in both plants is about five ml/liter phosphating bath, despite small tilting-plate separation volumes (approximately three pct of the phosphating bath volume).

Composition of the phosphate layers.

The compositions of the phosphate layers formed on steel, electrogalvanized steel or aluminum are given in Table III (below).

The layers' copper contents amount to 0.4 - 0.8 pct. They are of the same order as the nickel contents in nickel-containing phosphate layers. On steel, electrogalvanized steel and aluminum, fine crystalline layers are obtained, with crystal size varying from five to 15 mm.

Analysis of the phosphate sludge.

The analysis of the phosphate sludge generated in the nickel-free system is given in Table IV (below).

The sludge sample has a comparatively low zinc content, which points at the high efficiency of the process. The copper content (0.03 pct) is very low and allows a more favorable classification of the phosphate sludge under the waste management legislation.

Corrosion resistance and paint adhesion.

The lab test carried out before switching to the nickel-free process and the test plates taken after the changeover show that the nickel- and nitrite-free phosphating process applied to steel, electrogalvanized steel and aluminum complies with all corrosion resistance and paint adhesion requirements.

As for hot-dip galvanized steel body panels, which at present are not being used in series production at BMW's Dingolfing site, a satisfactory paint adhesion cannot yet be achieved for all qualities with the nickel-free process.

Performance when electro-painted.

Layers produced by means of the nickel- and nitrite-free phosphating process can be coated with the cathodic electropaint used in both car-body plants without any change to their parameters. No impact on the baking performance or paint layer (roughness) has been observed.

The new nickel- and nitrite-free process at BMW's Dingolfing production site has proved its effectiveness in the pretreatment of steel, electrogalvanized steel and aluminum. PF