HIGHLIGHTS
- The owners of Tamanend Winery found what appeared to be rust on the welds of stainless steel cooling coils in plastic wine tanks that had been filled with water for many months.
- The stainless steel manufacturer eventually identified the problem as microbiologically influenced corrosion.
- The authors concluded that stainless steel equipment and tanks should not have contact with standing water for long periods. Another area of concern is concrete flooring.
After more than 30 years in the wine business, we thought we had seen just about everything that could happen in a winery. In March 2009, however, we at Tamanend Winery experienced an event that totally surprised us. This is a cautionary tale that all winemakers should take very seriously and apply to their own winemaking facilities.
We had moved our winery into a 15,000-square-foot building late in 2007. The cellar area did not have HVAC to keep the ambient temperature low enough for wine storage. It’s a large space, and when the weather started to warm up in spring 2008, we hadn’t filled the space with tanks full of wine, which would have helped to lower the cellar temperature. What we did have were three 120-hectoliter (3,200-gallon) plastic tanks with stainless steel tubing used as cooling coils. We filled the tanks with water and turned on the glycol system to keep the water at 40ºF.
The tanks then provided cooling for the winery. This system worked flawlessly for 10 to 11 months.
In late March 2009, when we needed to use the tanks for wine, we drained the water and began to clean up the tanks and stainless steel cooling coils. The first tank seemed to be normal, with only the anticipated surface contamination from whatever was left residually in the tank. The second tank had what appeared to be rust spots on some of the welds on the cooling coils. This was a crusty, granular, rust-colored material that partially brushed away. The remainder, however, looked embedded in the welds.
The cooling coil in the third tank was more broadly covered with the rust-colored material—not just on the welds, but also on the smooth surfaces. We were able to brush the material off the smooth surfaces without too much difficulty.
At that point, we contacted the welding company that had manufactured the cooling coils and asked them to send a welder to clean up the welds and the welding splatter that appeared to be present. Closer examination by the welder revealed that the problem was not with the welds, but with something else.
The next step was to determine if there was a manufacturing problem with the stainless steel that was used. The welding company traced the heat number of the lot of stainless steel back to the foundry to see if there was a problem with that particular lot.
It was then that we found out what had been lurking surreptitiously in our tanks for 11 months.
According to the manufacturer, a series of conditions exist in which either a combination of electro-chemical reactions and/or oxidative/reductive species of bacteria can aggressively attack and live off the iron in metals such as stainless steel. In essence, these bacteria can “eat” stainless steel at the rate of as much as one-eighth-inch per month.
Microbiologically influenced corrosion (MIC) is not a familiar condition in the wine industry. It is better known in the energy industry, where cooling towers and heat exchangers use water for cooling in both open and closed systems. In one case, a company ordered a $200,000 cooling system for its plant. The system was set in place and hydro-tested to be sure that it was delivered in ready-to-go condition. The manufacturer sealed up the system and left it for two months. At that time, when the plant started the system, so many of the stainless steel tubes in the water transfer system burst open due to MIC that the entire $200,000 system was scrapped.
Gradually, more information has been found about these organisms, and scientists now know that they are found everywhere from the soil to the deepest trenches of the oceans around the thermal vents spewing hot, iron-saturated liquid into the ocean.
There are many species of these oxidative/reductive bacteria, including Gallionella, Leptothrix and Geobacter metallireducens. Microbiologists have known for 150 years that the Gallionella species will oxidize and fix iron. The mechanisms used by the bacteria to accomplish the task are less understood, and scientists do not know if this ability provides some competitive advantage to the bacteria.
Niche ecosystem
It is now understood that one of the primary mechanisms of action of these bacteria has to do with the formation of biofilms. This is a condition in which the bacteria set up a niche ecosystem on the surface of metal (see illustration on page 134). In this niche, various species provide habitats where they work together to provide the energy sources for growth. These three bacteria and others can live in oxygen regimes that range from aerobic through facultative (both aerobic and anaerobic) to anaerobic metabolic systems.
The implications of microbiologically influenced corrosion should be taken very seriously by the wine industry. Obviously, equipment or tanks made of stainless steel should not have contact with standing water for long periods of time. However, another area of concern should be concrete floors, and this is also based on our experience at Tamanend Winery’s facility.
Our facility was built many years ago as a 7-Up bottling plant. The sloped concrete floors were in reasonably good condition, but they were not sealed. The building hadn’t previously been used as a winery, and most recently it had been a forklift repair shop. Even after much cleaning of the floors, the concrete would not accept epoxy coatings, and we made the decision to wait and coat the floors at a later date.
We installed our 120-hectoliter tanks and filled them with water. Because the tanks were below the dewpoint, there was significant condensation on the sides of the tanks that then dripped on the floor—a common and not surprising sight at many wineries. During the 11-month timeframe that our tanks were filled with water, the floors were cleaned regularly. At no time was there any wine on the floor in the area of these tanks.
Three months into the summer, one particular area of concrete started pitting and eroding. In another several months, this area looked much like floors in wineries where the concrete has been eaten down to the aggregate. Many winemakers believe this pitting is due to the acids in the wine and the sulfites from either barrels or other sources.
Once we learned that the corrosion problems on our cooling coils were caused by bacteria, we took another look at the concrete floors under and near the tanks. Even after routine cleaning of this area, there was a significant film on the floor that was as slippery as slime. We think that the slime-forming bacteria were present on our concrete floor, and a biofilm ecosystem was eating our concrete (as well as the stainless steel in the coils).
We now use quaternary ammonia disinfectants in cleaning our floors. The slimes have disappeared, and the rate of erosion has virtually stopped. Our tanks are used for holding juice or wine, not water. We hope that other winemakers will learn from our experience and eliminate all possible sources for bacteria to live in water and eat stainless steel or concrete.
Dr. Richard Carey and Linda Jones McKee are owners of Tamanend Winery in Lancaster, Pa. Carey is president of Vitis Wine Center, and Jones McKee is co-editor of Wine East.
CURRENT MONTH'S FEATURES INDEX ยป
