Stop Loss

Not all ethanol leaving the fermenters goes to the tank.
By Susanne Retka Schill | August 06, 2012

Ethanol yield is often judged by what ends up in the tank farm but there’s plenty of places in the corn-to-ethanol process where small amounts of ethanol can be lost. In a tight margin environment that’s worth going after.

For some time, ICM Inc. has fielded questions about whether the ethanol found in the CO2 scrubber bottom waters is actually recycled and recovered. “ICM and part of the industry had the belief the ethanol in the CO2 scrubber recycles. So every plant has this continuous recycled flow that isn’t necessarily lost,” says Adam Anderson, product manager. The questions persisted, though. “About a year and a half ago we took on the task to come up with a way to get a study done to quantify whether ethanol truly is lost.”

Indeed, they found some ethanol is lost in the recycling of process waters. The question was also asked: just how much ethanol is being lost in venting process tanks? ICM process engineer Mareus Lopez helped design the tests to answer those questions. “We did a mass energy balance before and after the cook water tank,” she explains. They measured the ethanol flow going into the cook water tank and the ethanol in the water coming out, and also analyzed other relevant process streams.

Cook water is stored in a tank that receives all sorts of process water being recycled through the plant. Fresh water is added and mixed with ground corn in the slurry tank and enzymes are added for liquefaction and saccharification—turning corn starch to glucose, the food for yeast. As living cells, yeast consume glucose to make ethanol as well as a lot of CO2. Roughly one-third of the corn becomes ethanol, one-third CO2 and one-third unconverted solids, the coproduct distillers grains. Ideally, all of the ethanol leaves the fermenter as a liquid and is purified in the distillation process, but in reality, ethanol remains in the other process streams, which is what ICM sought to quantify.

Fermentation gases are passed through the CO2 scrubber, a large, columnar vessel where water washes through a filter medium as the gas moves through, scrubbing the ethanol, and a small amount of other compounds, out of the CO2. The scrubber bottom waters are then sent to the cook watertank to be recycled through the process. Other sources of ethanol come from the backset (water removed from the solids in the evaporators) and sidestripper bottoms from distillation. Six ethanol plants, ranging in size from 40 MMgy to 60 MMgy, cooperated with the ICM team to test the actual ethanol in the various process streams.

By far, the largest source of ethanol being recycled through the process comes from the CO2 scrubber bottoms. The average ethanol content in the scrubber bottom waters at the six plants was 2.7 percent by weight/volume. With every ethanol plant running a slightly different flow rate of water through the scrubbers, the accompanying chart shows the calculated ethanol flow in gallons per hour (gph) on the vertical axis in relationship to various CO2 scrubber water flow rates in gallons per minute (gpm) on the horizontal axis. “The chart is showing for a plant running 60 gpm water, a good estimate would be they would have about 90 gph of ethanol flow out of their CO2 scrubber,” Anderson points out.

“This is not ethanol loss,” Anderson adds. “That water is going to flow to the cook water tank and on to slurry tank.” In theory, that ethanol would just move through the process and have another chance of being distilled off. But, the ICM mass energy balance tests of the ethanol going into and out of the cook water tank showed an ethanol loss of as little as 21.3 gph or as high as 82.2 gph, with an average of 49.1 gph loss in the six plants. “If not all the ethanol is coming out, where is it going?” Lopez asks, adding that it could be through the slurry vent or consumed by bacteria. In the end the conclusion was that the main cause of ethanol loss is due to bacteria.     

Loss in the cook water tank was the easier one to calculate, Anderson adds. Far more complicated was devising a method to measure the ethanol content and possible losses from vents on the slurry and centrate tanks as well as the distillation vent.  “Those were the much more difficult ones to verify,” he says.  Both the slurry and distillation vents are routed to the top of the centrate tank where all three vapor flows move on to emissions control. It is in the thermal oxidizer or regenerative thermal oxidizer (RTO) that volatile organic compounds are burned off to eliminate dust and smelly emissions. With a bit of experimentation, the ICM team devised a method to pull vapor samples, and quantify the sample and the flow in the pipe. “We used a vacuum to pull a sample from pipes, running the samples through a water bath,” he says.  Most of the ethanol loss from venting came from the distillation vent, with an average of 14.2 gph. The slurry vent was much smaller at 4 gph, putting total average centrate vent loss at 18.2 gph.  The range for the six plants was from a 10.8 gph minimum to 26.3 gph maximum of ethanol in the vapor flows moving through the centrate vent.

To do the math, adding the 49.1 gph from the cook water tank loss to the 18.2 gph centrate loss results in about 1,600 gallons lost per day for every 50 MMgy produced at the six plants—about a 1 percent loss. “If you hear 1 percent, it doesn’t sound really bad,” Lopez says, “but if you take into account the revenues you get for a gallon of ethanol and talk about 67 gph, it is significant. If you figure per month or per year, it is alarming,” Lopez says.

Once the losses were quantified, the challenge for the ICM team was to see if they could devise a cost-effective solution. A rather simple piping change and rerouting of the CO2 scrubber bottoms water to enter into the cook water tank discharge eliminates the stream’s exposure to ethanol-eating bacteria. That still recycles the ethanol, though, so the team decided to see if they could find a way to economically recover it. With a two-year payback in mind, Anderson says they set $150,000 as the price point. “The potential loss of the ethanol in the CO2 scrubber bottoms and the slurry vent are areas where plants are limited on operational or simple piping project options,” he explains. “We wanted to see what could be done for that amount of money and that eliminated some of the high price tag items such as an additional scrubber or something inappropriately more extravagant.” Instead, they route the scrubber bottoms water through a heat exchanger for preheating and introduce it into one of the distillation columns capable of handling the extra flow. The changes bring the cook water tank loss to zero and the slurry vent average from 4 gph to less than 0.5, thus dropping the centrate vent average loss down to around 14.5 gph.

Kansas Ethanol LLC is one of the six plants where the testing was done, and the first to install ICM’s recovery solution commercially. Down the road a little over an hour from ICM’s Colwich, Kan., headquarters, the 55 MMgy ethanol plant has frequently worked with ICM’s team on various projects, says operations manager, Thane Combs. “The goal here has always been to be as efficient as possible in every aspect of our plant,” he says, adding that the company consulted ICM about this issue, suspecting ethanol losses due to the tank environment.

It was calculated that the scrubber bottom discharge was roughly 2 to 2.5 percent ethanol, Combs says. With 50 gpm of water being sent to the CO2 scrubber, there is about 1 to 1.5 gpm of ethanol being washed out. “Most plants are designed where that went to the cook water tank and then into the slurry at 185 degrees,” he points out. “So you are most likely flashing a pretty good portion of that off. Of course, that goes to the centrate blower and you’re just burning that up in the RTO.”

There’s no doubt Kansas Ethanol’s RTO runs cooler since the system was installed. Using natural gas to add to the heat generated from the burning emissions, the plant’s RTO is supposed to maintain 1,600 degrees Fahrenheit. “Before we put this system online, there were times that the gas would shut off to the RTO, when it was burning in excess of 1,600 degrees,” Combs says.

Anderson adds that when ICM reviewed the cost analysis for possible solutions, one of the factors considered was the energy contributed by ethanol vapor being oxidized in the RTO. “But ethanol is worth 10 times more than natural gas,” he says.

For Combs, the ethanol recovery project has been worth it. “It’s hard to measure and say [the loss] is zero,” Combs says. “We are seeing 700 gallons per day additional ethanol when it’s running compared to when it’s off. If you run the numbers, there’s approximately 1,400 gallons that are in the scrubber bottom waters.” He adds that at first they saw more ethanol recovery, “but the longer you run it, the system balances out and you’ve got less recycled ethanol throughout the plant.”

Author: Susanne Retka Schill
Contributions Editor, Ethanol Producer Magazine
(701) 738-4922