Agitation Challenges in Cellulosic Ethanol Production

By Gregory T. Benz | February 05, 2008
  • WARNING: Resizehelper couldn't find requeted file: /datadrive/websites/
Key fluid agitation problems face the emerging cellulosic ethanol industry. Of particular importance is the conflict between the need to use high solids fibrous slurries, which require expensive solids mixing equipment, versus the desire to use inexpensive fluid agitation equipment, which cannot handle high-solids fibrous slurries. Despite the challenges, possible solutions and areas where further research would be beneficial are available.

One of the challenges facing the industry is the variety of different feedstocks, each having different rheological properties that affect mixing difficulty. Examples of feedstock include corn stover, switchgrass and recycled newspaper. Some facilities must be designed to handle multiple feedstocks. The key parameter for mixing is how much water can be absorbed inside the fibrous cells before there is any "free" water available, marking the transition from a damp solid to a liquid slurry. When the slurry behaves as a damp solid, it must be mixed with expensive solids-mixing equipment, which is not available with large volumetric capacity. When the slurry behaves as a liquid, it can be handled with fluid-mixing equipment, which is much less expensive and readily available for very large tank volumes.

Another challenge is that many different processes are still being developed, and each has its own requirements. However, the wide variety of processes can be simplified in order to focus on their common elements and identify agitation issues. In general, all cellulosic ethanol processes have four major process steps: pretreatment, hydrolysis, fermentation and ethanol concentration. Sometimes some of these steps occur concurrently. For example, fermentation may begin in the hydrolysis step. Though there can be many differences in each of these steps from one process to another, the mixing challenges are qualitatively the same.

Pretreatment involves making the individual cellulosic fibers more accessible to the agents of hydrolysis. Pretreatment may involve combinations of mechanical steps such as chopping the fibers, thermal processing, and chemical processing, such as treatment with acids, alkalis and/or enzymes. Generally, liquid agitation is not possible, due to low free moisture. A solids processor or a modified rotary dryer which tumbles the damp solids without the addition of heat might be useful in this step.

For an economically viable process, the solids content feeding the hydrolysis step should be as high as possible, typically 20 percent to 30 percent or more. Any excess water not only dilutes the enzymes added for hydrolysis, reducing the reaction rate, but also dilutes the sugars produced. This slows fermentation, lowers ethanol concentrations and adds energy cost for concentrating the ethanol.

Water in fibrous slurries can be found both inside and outside the cells. However, only the water outside the cells, which we will call "free" water, contributes to product fluidity. The practical significance of this is that cellulosic slurries above approximately 12 percent to 15 percent dry solids content are not liquids, but damp solids, having little or no free water.

They cannot be handled by turbine agitators. Instead, they must be mixed by expensive, low volumetric capacity solids mixers. Turbine agitators are much less expensive and have no real size limitations, but generally are not practical above approximately 8 percent to 10 percent solids because the required torque of a turbine agitator is roughly proportional to percent solids cubed. Therefore, from a turbine viewpoint, the solids should be more dilute. How can we keep the feedstock at high percent solids and yet use inexpensive turbines? In other words, can we somehow use the water inside the cells to fluidize the slurry? The answer lies in what is happening in the hydrolysis reactor.

In this reactor, with time and reaction progress, the cellulosic structure is broken down, and the cellulose is hydrolyzed into various soluble sugars. This means the end product of a hydrolysis reactor is basically low viscosity liquid, not fibrous slurry. The key to using a turbine, therefore, is to avoid exposing it directly to the high-solids, fibrous feedstock from the pretreatment step, allowing it to run in a mixture of feedstock and the hydrolyzate reaction product. The strategies are slightly different for batch and continuous reactors.

Batch Reactor
A pure batch reactor would involve filling the tank with the pretreated feed, say at 30 percent solids, adding the necessary reagents and letting the reaction proceed. Such a reactor would then have to be able to mix the 30 percent fibrous material. This is not practical for large scales of operation, as it would require a solids mixer. A better way is to operate as fed batch, starting out with a small heel of water. The pretreated feed plus enzymes would be added until the tank was full, allowing hydrolysis to occur, breaking down the cellular structure and releasing internal water so as to increase the "free" water content. Filling would be done at a rate so as to keep the tank contents fluid enough to be agitated by a turbine, then the tank would continue to be agitated until hydrolysis was essentially complete. The research needed for such a system would likely employ empirical kinetics studies, plus agitation studies to determine what the maximum effective percent solids after partial hydrolysis might be, and the minimum agitation at various solids concentrations. This would be highly dependent on the pretreatment process used as well as the kind of feedstock.

Continuous Flow Reactor
In a continuous flow scheme with a backmixed reactor, the bulk contents of the tank have the same composition and fluid properties as the effluent. Thus, it is necessary to assure that the retention time in the first stage of a continuous flow system is sufficient to liquefy the feed. The minimum mixing torque/volume to create full motion is roughly proportional to the cube of the percent solids. However, mixing clumpy, high solids material into a lower percent solids bulk liquid will require substantially more torque than that required to maintain full liquid motion. Thus, a better scheme than using a single large tank or equal-sized tanks in series is to start with a small first stage, which will have an equivalent discharge consistency as high as possible for a turbine (likely 8 percent to 12 percent). This scheme avoids the need to create high agitation intensity in a large tank size. The small first stage is followed by one or more tanks in series, with the final tank discharging an almost fiber-free material. It minimizes agitator cost and power. Research needed for this scheme includes at least an empirical study of kinetics plus agitation studies as a function of percent solids. Feedstock will heavily influence this approach.

Although for the purposes of this article fermentation is treated as a separate step, many processes under consideration actually begin the fermentation in the hydrolysis reactor. Aside from pure economics, such processes may be used to deal with the fact that some sugars impede the very hydrolysis reactions that make them, but this does not affect the minimum mixing requirements.

Ethanol fermentation has been used for thousands of years. Even fuel ethanol has been in production around the world for at least 50 years on a large commercial scale. Yet the influence of agitation has not been well studied. Most ethanol processes based on sucrose and yeast just use agitators to keep the tank bottom reasonably clean, and no additional research is needed to design for that function. However, work done by Professor Enrique Galindo of the Institute of Biotechnology at National University of Mexico showed that even for simple yeast/sugar ethanol fermentations, the process results were directly affected by agitation. Specifically, he showed that rate of production, yield and maximum percent ethanol all increased up to an agitator-specific power input of 1.6 watts per kilogram (8 horsepower per 1,000 gallons), which is much higher than anything used on a commercial scale. Others have anecdotally reported similar results. A mechanistic study needs to be done to explain the reason(s) for such strong influence and the implications for proper scale-up.

Ethanol Concentration
Agitators in the ethanol concentration area of the plant are needed to keep solids from settling. This function is well understood by agitation consultants familiar with the industry, and no additional research is needed.

The fledgling cellulosic ethanol industry has major agitation challenges in the hydrolysis reactor area, due to the need to handle high solids feed. Research into agitation issues is essential for viable process operation as well as minimizing agitation capital and energy costs. Less critical, but worthy of study to allow process optimization, is the influence of agitation on fermentation. If the industry gets this right, productivity could be greatly enhanced.

Gregory T. Benz is president of Clarksville, Ohio-based Benz Technology International Inc. Reach him at or (937) 289-4504.