## PARC building cleantech portfolio; co-extrusion printing of novel battery electrodes and carbon-neutral renewable liquid fuels from atmospheric CO2

##### 09 March 2011
 Schematic of the HP-BPMED device used in the renewable fuels research. (a) Atmospheric CO2 separation using a continuous-flow electrodialysis system. (b) Detailed view of the stack. Source: PARC, portions courtesy of Ameridia Corp. Click to enlarge.

PARC (Palo Alto Research Center), the home of a long line of technology innovations, including laser printing, object-oriented programming, and personal workstations with graphical user interfaces (e.g., the Alto), is building a portfolio of research in the cleantech area. PARC was founded in 1970 at Xerox; in 2002, it was incorporated as a wholly owned independent research and development subsidiary of Xerox Corporation, providing services to external customers as well as to Xerox.

Dr. Scott Elrod, VP and Director of PARC’s Hardware Systems Laboratory (HSL) research organization also directs the Cleantech Innovation Program at PARC, which develops solutions for delivering affordable solar energy, increasing solar cell efficiency, purifying water, managing energy utilization, and producing renewable fuels. Two of the projects Elrod and PARC were discussing at last week’s ARPA-E Energy Innovation Summit in Washington DC were a technology for the co-extrusion printing of novel battery electrodes, enabling higher energy and/or power densities; and an approach to producing carbon-neutral renewable hydrocarbon fuels using air, water and CO2 captured from the atmosphere.

Electrodes. PARC has developed a technology for co-extrusion printing—i.e., for depositing thick films of pastes of densely interdigitated functional materials. For batteries, these functional material pastes would be the electrode active materials. A post deposition processing step dries and sinters the deposit into the final electrode structure. The stripes can be as narrow as several microns wide and as tall as several hundred microns.

The first application of this technique has been for silver gridlines on the front surface of solar cells. Compared to screen-printed gridlines, the narrower and taller front gridlines created by the technique cover less solar cell surface area, resulting in increased absolute cell efficiency. A production prototype machine is under test at a customer site, with an additional gain of 1% absolute cell efficiency and process speeds up to 200 mm/sec having been demonstrated.

While the solar cell application has a near-term sales opportunity, commercial application of the technology to battery electrodes is probably 2-3 years out, Elrod noted. There is further opportunity for the method in air cathodes. The current density in an air-breathing electrode is proportional to the amount of electro-catalytic surface area that is exposed to air. The PARC technology provides a directed-assembly printing method for producing a greater proportion of this “three-phase boundary” than conventional electrode manufacturing methods—up to 10x the air-breathing surface area of conventional electrodes.

Carbon-neutral liquid fuel. PARC is developing a non-biological approach for producing liquid fuels from renewable energy, air, CO2 and water. PARC’s proposed approach captures CO2 from the air and reacts it with hydrogen produced via the electrolysis of water to produce liquid hydrocarbon fuels. As long as the energy for the process is renewably generated, PARC notes, the overall process is carbon-neutral.

Current PARC research (funded by DARPA) focuses on the critical step of capture CO2 directly from the atmosphere. PARC is focusing on recovery using a novel electrochemical approach developed at PARC: high-pressure bipolar membrane electrodialysis (BPMED), and has designed, constructed and begun testing of a prototype for CO2 separation.

The technique uses a solvent such as sodium or potassium hydroxides, converted by reacting with CO2 to aqueous carbonates or bicarbonates. (One possible technique for this might be the use of spray towers, Elrod noted.) The spray solution is then pressurized to 5-10 atmospheres, Elrod said, and introduced to the BPMED unit. BPMED then regenerates the CO2 gas.

The bicarbonate is transferred across the anion exchange membrane to the CO2-rich acid stream that is held at a constant pH of 3 to 4 by a combination of acidic buffers and flow-rate control. The capture solution is buffered against excessive pH increases and held at a constant pH of 8 to 10 by the presence of significant concentrations of HCO3- and CO32- ions. The capture solution is regenerated by the OH--ion flux from the bipolar membrane and by partially depleting it of HCO3- and CO32- ions via electrodialysis.

The high-pressure acid stream is transferred to a gas evolution/separation tank where the pressure is reduced with concomitant release of pure CO2. The now CO2- depleted acid stream is returned to the electrodialysis unit via a repressurization pump while the regenerated capture solution is returned to the spray tower.

—Eisaman et al. (2009)

PARC is focusing on three key innovations to enable this approach: (1) active pH control for energy efficient CO2 concentration; (2) very high-pressure operation for suppression of in-stack gas evolution; and (3) very high current densities in a gas-evolving system.

In a paper published in the RSC journal Energy & Environmental Science, the PARC researchers present results indicating that the energy consumption required to regenerate CO2 gas from aqueous bicarbonate (carbonate) solutions using this method can be as low as 100 kJ (200 kJ) per mol of CO2 in the small-current-density limit.

Resources

A potentially interesting manufacturing process for lower cost automated production of future battery electrodes and solar cells components. When fully developed, similar process could enable USA's industries to compete with their lower wage Asian counterpart.

Higher automation levels, use of lower cost materials, increased mass production and world wide competition could reduce future batteries price in the same proportion as other electronic components.

Deniers will have a fit when they realize the major advancements in battery technologies and manufacturing process and specially the makor drops in price.

The CO2 extraction scheme is close to the holy grail of sustainable fuels, but I have to wonder: is the electrolytic method any more efficient than using e.g. waste heat from a higher-temperature electrolysis cell to dissociate KHCO3 directly? I had the figures at one time but don't remember if I saved the links, and Google has changed its formats again and broken the Scroogle scraper.

Something for both it seems - electric for the short range vehicle and renewable liquid fuel for the long range vehicle - can't go wrong with that.

As a denier I would say that the cost of electronics has dropped steadily because of Moore's Law ie the halving of feature sizes for integrated circuit manufacturing because of continual improving lithography on an 18 month cyle.

Battery manufacturing is not governed by Moore's Law.

Printed battery electrodes and solar cell components may become close parallels to integrated circuit manufacturing and will certain improve in significant steps every few years or so. The cycles length and size may not be identical or would probably peak sooner and will not be called Moore's Law but somebody will eventually give it a name and suitable progression formula.

Improvement in batteries energy density (could go from (20, 40, 80, 160, 320, 640, 1280 Wh/Km or at slower non-linear rate etc), batteries charging time, BEVs range between charges, BEVs body weight, BEVs energy consumption per Km, BEVs market penetration, etc could also advance in major non-linear steps. They will all move along in the coming years.

Extracting CO2 at 385ppm will take a lot of energy. More likely it's better to locate a liquid fuel refinery adjacent to a coal fired plant and use its CO2 stream.

How much CO2 is now discarded as a byproduct of LN2 and LO2 production? That may be a place to start, at least as a demonstration.

Mike, thanks for this detailed piece; these comments are great! So we (Scott Elrod, Matt Eisaman, and Karl Littau) took the time to respond here under our company Twitter account (@parcinc):

@Engineer-Poet -- Because the electrodialytic approach does not require heating the entire solvent mass to extract the CO2, relative to thermal regeneration, electrodialytic regeneration has an increasing advantage for solvents and gas streams with high binding energies and low CO2 loading (the amount of dissolved CO2 it is possible to "load" into the solution). Relative to flue-gas capture, the capture of CO2 directly fom the atmosphere necessitates the use of capture solvents with high binding energy, and the low concentration of CO2 in the atmosphere results in relatively low CO2 loading, giving electrodialytic separation more of an advantage in this case.

@Mannstein We completely agree that battery manufacture isn’t governed by Moore’s Law. The emphasis on making finer geometric structures in batteries is not meant to be similar to the “transistor shrink” associated with successive semiconductor technology nodes. Rather, it is a very specific geometrical approach to alleviate the limitations on Li ion transport in the cathode material. Unlike Moore’s Law (which benefits from ever finer features), the fine segmenting of battery electrodes will reach a point of diminishing returns. Still, we expect the optimally-sized structures to result in a battery with a higher energy capacity than a conventional Li-ion battery.

@HarveyD There will indeed be specific “learning curves” associated with different battery performance parameters. The learning curve for microelectronics -- Moore’s Law -- is enabled by two primary factors: wafer size increase, and photolithography feature size decrease. Moore’s Law is the fastest learning curve anywhere in manufacturing. While substrate size for batteries might increase somewhat, the substrates are alreasy quite large (i.e., roll-to-roll systems). And there is no analogue to the photolithography shrink associated with batteries (or solar, or fuel cells). So the learning curves for batteries will likely be much more gradual.

@Reel Although the concentration of CO2 in the atmosphere (about 386 ppm in 2009, rising at a rate of about 1.9 ppm) is about 260 times less than the concentration of CO2 in flue gas (about 10%), the logarithmic scaling of the thermodynamic minimum energy of separation means that the minimum energy required to separate 90% of the CO2 from air (21 kJ/molCO2) is only 2.9 times greater than the minimum energy required to separate 90% of the CO2 from flue gas (7.3 kJ/molCO2). Also, it is important to note that using CO2 separated from power-plant flue gas would not create a carbon-neutral fuel because the CO2 emitted upon combustion could not be re-separated for a subsequent fuel synthesis and combustion cycle.

@richard schumacher CO2 is indeed removed from air before making liquid air products. However, the quantity is quite small and the production locations are not centralized making it difficult to utilize. In the entire U.S., approximately 10,000-20,000 tons of CO2 is removed from the air in the manufacture of all liquid air products. Typically the CO2 is captured as carbonates and is discarded because the energy required to regenerate it is prohibitive.

PARCinc: That's interesting, but I question the reasoning. Heat energy can be regenerated with recuperators. The higher the binding energy, the more losses you sustain in the conversion from heat to electricity. I also question whether loading is such a big deal; potassium carbonate-bicarbonate dissociates slightly above 100 C (well-matched to flat-plate solar collectors and many waste-heat streams) and I've not heard of it failing to take up enough CO2.

I suppose the devil is in the details. If you have any comparison studies that are worth reading, I'd appreciate a link or three.

PARCinc: Thank you for the broad information on what you called specific learning curves. However, we should not forget that learning curves or design and manufacturing progress can be accelerated when needs grow.

We've all seen such acceleration during wartimes or the race to space and the moon etc.

The growing worldwide need for electrified vehicles will certainly support much greater battery development in the nest 10, 20 or 30 years. Progression may not be as fast as for microelectronics but it may be much faster than the last few decades. In other words it may not be linear nor gradual. Breakthroughs will come every other year or so. Many existing battery technologies will be improved and new ones will be found. Will Moore's law apply? Probably not, but some king of growing rate will be identified and could become the source of another learning curve law. Somebody will give it a name.

The comments to this entry are closed.