Improving biofuel economics using upstream innovations for sample preparation
“Responding to these challenges effectively requires a ‘life cycle perspective,’”1 says Thomas McKone, a senior staff scientist at Lawrence Berkeley National Laboratory.
It takes energy to transform something into a usable form of energy (see Diagram 1). This life cycle perspective attempts to explain the net impact of energy production on the environment—from the source to its final destination with the end user.
For example, the U.S. uses almost 3 billion gallons of gas just to transport the fuel it consumes each year. This number may seem large; however, it is only 2 percent of the total 133 billion gallons consumed each year.2
First- and second-generation biofuels come from food sources such as edible corn, and the second generation from a variety of feedstocks such as liganose or municipal waste. Third-generation biofuels are typically microbial, using CO2 as their feedstock, and are much more carbon neutral.3
Depending on the type, biofuels can have a 40 percent more efficient energy balance than fossil fuels. Fuels from second-generation and third-generation fuels can actually have a positive balance.3 While biofuels make environmental sense, they are still more costly than fossil fuels.
Second-generation biofuels eliminate the foods versus fuels land-use controversies of the first generation because they are produced from agriculture and forestry residue or inedible by-products. Input materials cost considerably less than first-generation fuels. Still, these fuels cost about 70 percent more to produce because extra steps in the production process are inefficient.4
Searching for innovation
One critical way to lower production costs for secondgeneration fuels is to improve industrial microbes, enzymes, and batch chemistry.
Better industrial yeast strains are needed to transform cellulose to glucose through an amylase breakdown step. This step alone accounts for more than half of the total hydrolysis process cost.5 This step is inefficient and costly because the current strains of industrial yeast have limited success when breaking down the dominant sugar types in second-generation biomass. Typically, industrially engineered yeasts are ten to 100 times less effective at digesting xylose than when digesting glucose in first-generation crops.
Testing and optimizing yeast strains is a time-consuming and difficult business. Yeasts need to be carefully monitored through their initial growth phase, called “doubling time,” and then monitored as the xylose is produced. Cells are grown in high-density cultures to maximize productivity by volume, and the medium is composed of highly concentrated sugars and enzymes.
Cell density is monitored regularly using photometric analysis so it can be maintained at specific OD660 measurements. Doubling time for a yeast culture is typically two to three days, while xylose production can peak between days four and six of the culture and may need to be sustained for longer to get the appropriate yield.
Sampling rapidly happens on the yeast culture’s schedule and not within the lab workday constraints. Adding complexity, the reagents often need to be added or changed outside work hours. Experiments are conducted in triplicate to ensure reproducibility and statistical rigor; sampling and testing assays rapidly can ramp to hundreds of aliquots.
Process innovation can ease brute-force screening
Process innovation can improve this long protocol and enable labs to conduct enough assays to efficiently screen promising new strains. Because of the challenges of strain optimization, many second-generation biofuel projects have languished in the recesses of university campuses and federal laboratories—until recently.
A genomics lab at the State University of Campinas in Brazil chose to automate yeast strain optimization testing. Using automation, scientists working there more accurately and efficiently screened yeast strains that could be used in second-generation biofuel refineries. The automation greatly assists brute-force cloning and screening clonal populations.
“With automation I can get accurate and reproducible results that simply would not be possible with standard means,” says Pedro Tezei, a State University of Campinas researcher (see video).6
At the State University at Campinas, the engineered yeast strains were grown and screened to find the most productive organism. The experiments monitored yeast strain activity over four to eight days.
Flexible automation platform made it possible
Using automation, they could collect samples and measure at any time during the multiday protocol. The core lab developed an end-to-end workflow that starts with yeast inoculation, grows the cell population to the appropriate density, and measures the population’s productivity over time. They achieved this with the Hamilton Microlab® STAR liquid handler.
Automation can make difficult protocols easy. For each core lab working on strain optimization for biofuels, hundreds of smaller labs are optimizing strains for other synthetic biology applications. These labs can benefit from early automation to make their processes more robust and ready for scale-up.
Why a robust process matters for scaling up
As researchers scale up bioreactor production, small changes in nutrients, pH, and temperature can influence yield in subtle ways. Careful and frequent testing is needed to maintain consistency and quality. An automated protocol can save months of chasing down false leads and not knowing what parameters have changed and when.
Conclusion and future trends
Automation will be a key piece of the puzzle to improve the efficiency of microorganism screens, sample during cell growth, and analyze cell production output. Without this information taken at reliable intervals—with reliable chain-of-custody data, Technology understanding how these strains grow, and defining the optimal growth conditions—setting parameters for the bioreactors within biorefineries will be greatly hindered.
However, automation alone will not guide second- and third-generation biofuels to supplant fossil fuels or firstgeneration biofuels. A few trends are likely to occur.
Biofuels will be niche contributors to worldwide energy consumption. Road maps for the U.S. and the EU both predict that renewables will contribute about 15 percent of our energy diet until 2030 and beyond.7, 8, 9 To incentivize a switch to more carbon-neutral renewables, policy will be informed by life cycle sustainability assessment metrics.
New companies and technologies will be commercialized to support the new biofuel ecosystem. Policymakers and entrepreneurs alike are conducting gap analyses to understand what technological building blocks are readily available and what is still needed.
Best practices will be pollinated from the agrochemical and petrochemical businesses. Early partners for biorefineries will come from the existing infrastructure of the agrochemical and petrochemical businesses. Already, states such as Minnesota are finding that existing grain co-ops and commercial refineries are extending their business by adding biofuel services.
Automation that scales from the research lab to the pilot lab will be in demand as production inputs are validated. To maximize the return on their automation investment, researchers would be wise to choose a flexible liquid-handling platform that can automate clone selection to microbial screening and reagent mix optimization to pilot scale-up.
1. Lawrence Berkeley National Laboratory, “Challenges for Biofuels—New Life Cycle Assessment Report from the Energy Biosciences Institute.” February 8, 2011. Accessed online February 2014. http://newscenter.lbl.gov/featurestories/2011/02/08/lca-challenges-for-biofuels/
2. U.S. Energy Information Administration website, Frequently Asked Questions. How much gasoline does the United States consume? Accessed online February 2014. http://www.eia.gov/tools/faqs/faq.cfm?id=23&t=10
3. U.S. Department of Energy website, Energy Efficiency & Renewable Energy, Ethanol Fuel Basics. Accessed online February 2014. http://www.afdc.energy.gov/fuels/ethanol_ fuel_basics.html
4. Roland Lee and J.M. Lavoie. “From first- to third-generation biofuels: Challenges of producing a commodity from a biomass of increasing complexity.” Animal Frontiers. 2013. Vol. 3 (2) 6-11. doi:10.2527/af.2013-0010.
5. Gennaro Agrimi, L. Palmieri, et al. “Process development and metabolic engineering for bioethanol production from lignocellulosic biomass.” In “Biorefinery: from biomass to chemicals and fuels.” 2012. ISBN 978-3-11- 026023-6.
6. Select Science Product Review Video. Accessed online February 2014. http://www.selectscience.net/SelectScience-TV/Videos/new-yeast-strains-forsecond-generation-ethanol--pedro-tizei,-university-ofcampinas/?&videoID=1760
7. U.S. Energy Information Administration website, Today in Energy. Accessed online February 2014. http://www.eia.gov/todayinenergy/detail.cfm?id=4950
8. European Renewable Energy Council website, Hatrick 2030, an integrated climate and energy framework. Accessed online February 2014. http://www.erec.org/media/publications/hat-trick-2030.html
9. Vision 2020 – Europe Biorefinery Joint Strategic Research Roadmap. Star Colibri Report 2011. Page 30. Accessed online February 2014. http://beaconwales.org/uploads/resources/Vision