In the previous essay, I discussed the National Advanced Biofuels Consortium (NABC) and the recently announced success of Virent Energy Systems (Virent), a member of the consortium. In this essay I try to dig into the process details a bit more with a series of technical exchanges with Virent Founder and CTO Randy Cortright.
My questions are denoted as RR and his responses are in blue as RC. I began by asking Dr. Cortright to confirm my understanding of Virent’s process.
RR: Let me make sure I am clear on the process. You take cellulosic biomass and hydrolyze that to sugars. You hydrogenate those sugars and put them through your aqueous reformer. The reformer output consists of ketones, acids, aldehydes, and alcohols, and those are then run through various processes to produce gasoline, jet, or diesel. If I have missed on the basics, please correct me.
RC: Our technology has tremendous optionality. In some configurations the sugar streams are hydrogenated and then fed to the APR system that generates volatile oxygenate compounds that are subsequently reacted (condensed) to generate the “drop-in” hydrocarbons. A few years ago, we discovered reactor configurations and catalysts that allow us to feed directly aqueous-solutions of sugars to the APR system. We have processed a broad range of oxygenated compounds derived from biomass, including C5/C6 sugars, polysaccharides, organic acids, furfurals, and other degradation products generated from the deconstruction of biomass.
As you have read in the NABC/Virent press releases, we have been able to process the hydrolysates generated from both loblolly pine and corn stover that were solubilized using NREL and Washington State deconstruction technologies, we have also been able to generate hydrocarbons from hydrolysates generated by Virent’s in-house catalytic deconstruction technology (this was developed under a program funded by the NIST Advanced Technology Program). Additionally, we are waiting for hydrolysates produced by HCl Cleantech from pine. Chemical analysis suggests that we will be able to generate hydrocarbons from source as well.
Virent’s technology strength is that we have broad feedstock flexibility with our process, we use catalysts and catalytic reactor technologies that are very similar what you find in today’s oil refineries, and we generate a range of hydrocarbon molecules, that when fractionated can be used as “drop-in” fuel components for gasoline, jet fuel, or diesel. Furthermore, in our gasoline process, we generate an aromatic-rich stream that is similar to a naphtha reforming reformate stream, and it is possible to process this reformate stream to generate a biomass derived paraxylene that can be used to generate PET for bottling or fiber production.
RR: What is the history of aqueous phase reforming?
RC: Aqueous Phase Reforming was discovered at the University of Wisconsin in 2001. We discovered that it was possible to react oxygenates such as sorbitol and sugars with water over a catalyst to generate hydrogen with carbon dioxide as a byproduct. Later we found that it was possible to use this in-situ generated hydrogen and generate hydrocarbons from the oxygenated compounds. Virent was started in 2002 and has an exclusive license for the technology. At the time, we expected a robust market to develop for hydrogen due to the advent of fuel cells cars. We focused on making hydrogen for 3 years. However, during that time oil went up to $60/barrel, and we decided to use the APR technology platform to see if we could produce hydrocarbon liquid fuels via this aqueous catalytic process- it worked.
RR: Is the process used commercially today for other applications?
RC: While some of catalytic technologies employed in Virent’s BioForming® is currently being used in today’s oil refineries, Virent has not commercially deployed the APR and BioForming® technology. While we haven’t commercialized it yet, we have scaled sugars-to-gasoline in our full-scale demo plant (called Eagle) which produces 100 liters per day in our facilities in Madison, Wisconsin.
RR: What is your ideal feedstock for the process? Are there feedstocks that are troublesome for you?
RC: The ideal feedstock is the one that we can have delivered to the plant gate at the lowest cost, and is available at the largest volume. Furthermore, a feedstock that can be made water soluble at the lowest cost is also desirable. Again, we can use a broad range of oxygenated compounds derived from biomass including polysaccharides, organic acids, C5/C6 sugars as well as tolerate solubilized lignin. We are truly feedstock agnostic, so we prefer using the cheapest one taking into account abundance, proximity, and the amount of processing necessary. Some components that are troublesome over our catalytic process are sulfur and nitrogen compounds. We have developed pretreatment strategies for removal of these compounds as well as we are working to develop catalysts that are more tolerant to these potential poisons.
RR: Given the sorts of compound that you are converting in your reformer, have you ever considered or tested pyrolysis oil? It seems like it could be ideal, and possibly a cheaper first step option than hydrolysis for breaking down biomass.
RC: Yes we are investigating using pyrolysis products as a feedstock. Materials that are still water soluble tend to work better.
RR: Your website described the process as “water positive.” What does that mean?
RC: A primary method to remove water from the soluble oxygenated compounds and generate a non-oxygenated hydrocarbon is the hydrodeoxygenation of the oxygenate compound with hydrogen (either generated in-situ or using external hydrogen). Water is generated from this hydrodeoxygenation step and, therefore, water is a byproduct of the process.
RR: What is your cost of production today?
RC: At commercial scale, the BioForming® technology can generate gasoline, jet fuel, or diesel from carbohydrates that on an energy basis are competitive with the production of ethanol. Importantly, the costs of our carbohydrate-derived distillates (jet fuel and diesel) are significantly lower than distillates generated from lipids and seed oils by either esterification or hydrogenation
RR: What sort of piloting have you done? At what scale and for how long?
RC: We have built a world-class R&D facility for catalytic processing of biomass-derived feedstock. We have 25 pilot plants in operation which are capable of generating ~1 liter per day of a fuel or chemical. We are able to configure these smaller scale pilot plants such that we combine the different catalytic reactor systems that allows us to generate gasoline, jet fuel, and diesel from sugars. We have been operating versions of these systems to generate liquid fuels since 2006. We have scaled sugars-to-gasoline in our full-scale demo plant (called Eagle) which produces 100 liters per day. We would be pleased to show you our facility at some point, based on your questions, I think you would appreciate it.
RR: What are some major challenges you still need to overcome?
RC: Technical Challenges: While we are now demonstrating commercially viable yields to liquid fuels, there is still optimization required. Furthermore, there is a need to demonstrate catalyst lifetimes with lower cost (but dirty) feedstocks as well as the need to demonstrate process durability.
Market Challenges: Regulatory risk due to the heavy influence of government regulations on the biofuels market. Margin risk between feedstocks and products.
Commercialization Challenges: Raising the capital to build either a demo plant or the first commercial plant.
RR: The hydrolysis process results in 5 and 6 carbon sugars. Does that mean that your APR process generates hydrocarbons that are 6 carbon or shorter?
RC: The APR reactor system can generate volatile oxygenate intermediates (alcohols, ketones, acids, cyclic ethers) from a broad range of oxygenates including polysaccharides, C5/C6 sugars, as well as fragments produced in the deconstruction processes. The volatile oxygenated intermediates generally have carbon numbers less than 6, but with some APR catalyst systems we do see some condensate to generate compounds with more than 6 carbons. The process steps downstream of the APR are utilized to condense these materials to greater carbon number hydrocarbons.
RR: Then to make molecules in the jet fuel range requires a bit more processing?
RC: Yes. We take the range of oxygenated compounds form APR (ketones, acids, aldehydes, alcohols) and process them, condensing them into the hydrocarbons needed for jet ranges. Then, subsequent hydrotreating is necessary, just like it is necessary in the treatment of similar crude-derived molecules.
RR: Your white paper says that hydrogen is generated in situ. Does that mean you don’t have to add extra hydrogen to the mix?
RC: Again this highlights the optionality of the technology. The technology has the ability to generate the necessary hydrogen in-situ (at the cost of loss of biomass-derived carbon through CO2 formation). We can enhance the overall yields of liquid fuels from biomass if we use hydrogen-derived from the steam reforming of low-cost natural gas currently being produced in the US.
RR: How large do you envision that a commercial plant will be?
RC: 50-150 million gallons per year, which is heavily driven by the feedstock chosen for the plant.
RR: What is the sensitivity of your feedstock costs to the bottom line? For instance, how much would the price be impacted if you had to pay $40 more per bone dry ton of feedstock?
RC: Feedstock costs are the largest cost driver for the generation of liquid fuels from biomass. Virent’s BioForming technology has the potential of generating the highest yields of liquid fuels from carbohydrates. Particularly, compared to biological routes, Virent’s catalytic technology has the potential of 50% higher yields (on a carbon basis) of liquid fuels from biomass-derived carbohydrates.
RR: The white paper also says that you can compete with $60/bbl oil. I am curious about the assumptions that went into that; in other words how much of that is forward looking, and what are the assumptions?
RC: Suffice it to say we need to write a new white paper! The white paper was written in 2008 when feedstock costs were significantly lower than what we are seeing today. As mentioned above, at scale, Virent’s technology can generate liquid fuels at a cost equal to ethanol production on an energy basis. Importantly, we can generate distillate range molecules from carbohydrates at costs significantly lower than distillate molecules generated from triglycerides by esterification or hydrogenation.
RR comment: Just a general comment here on feedstock costs. Note that in my essay Bad Assumptions, I flagged the assumption of low biomass costs as one of worst assumptions that many biofuel companies make. It is quite easy to plug an assumption of dirt cheap biomass into a model, and then to conclude that one can make cheap biofuels. In reality I believe that cheap biomass is going to be rare, and companies for the most part should assume prices on par with the price of hay (which as of this writing averages $75 to $143 per ton depending on the quality). This is why I always ask questions about sensitivity to biomass costs — because many companies underestimate their feedstock costs.
RR: Three-part question around the energy balance:
· For one BTU of biomass input, how many BTUs of liquid fuel output can you achieve?
· And what is the energy balance for the output?
· Are you counting on burning the lignin for process heat in the energy balance?
RC: With in-situ hydrogen generation, up to 94% of the heating value of the sugar can be incorporated in the liquid fuel. If we use external hydrogen generated from natural gas, then well over 100% of the heating value of sugar can be incorporated in the liquid fuel.
Overall, the Bioforming process is exothermic and with appropriate heat integration the overall process will be very thermally efficient. While we can use lignin as a source of process heat it is not required. More than likely the use of lignin will be used to provide excess energy that would be converted to electricity that can be exported. Furthermore, the BioForming process has the capacity to convert solubilized lignin to liquid fuels which would further improve the overall yields of biomass to liquid fuels.
RR: What is the ownership structure of Virent? Who are your major investors?
(This question was a follow-up, answered by Kelly Morgan, Marketing Manager at Virent): We are privately-held at this time. Our major investors are Cargill, Royal Dutch Shell and Honda. Other investors are Venture Investors, LLC and the Wisconsin Alumni Research Foundation.
With that, I would like to thank Virent for taking the time to share information with me. Just a reminder that when you are doing due diligence, you should pay particular attention to how specific questions are answered, and note which questions are not completely answered. In some cases, questions may not be fully answered for proprietary reasons, but in other cases information may be left out because there are some sticky issues around a particular area. (That isn’t pointed specifically at Virent; that is a general observation). That is the point of due diligence; ask questions, review answers, and then hone in on certain areas.
At this stage I see no obvious “knockout factors,” but there are always plenty of pitfalls to be navigated when scaling up a technology. Having said that, it does appear that they are separating themselves from many competitors with a unique spin on the conversion of biomass to fuels. At this point in their process, the technology appears quite promising to me.
If readers have additional questions that they don’t feel were covered, perhaps they can post them in the comments following the essay and Virent may be willing to provide additional answers.