Industrial Scale Ethanol

After my recent posting on resurgent food vs. fuel competition from expanding corn ethanol production, one of my contacts called to ask if I was familiar with an industrial process developed by Celanese Corporation for producing ethanol from a variety of feedstocks, including natural gas, coal, and potentially cellulosic biomass. My initial reaction to him was based on my knowledge that such processes have been around for decades, and that until the policy-inspired growth of the corn ethanol industry, much of the ethanol for industrial use was produced in that fashion. However, I was unaware of plans to deploy this technology on a truly massive scale, in the form of a pair of 400,000 ton-per-year coal-to-ethanol plants in China. I consider this a really interesting development on several levels.

The attraction of producing ethanol for industrial or fuel use from indigenous non-food raw materials in China seems obvious. It enhances the country's food and energy security by avoiding imports of both. As I delved into the technology involved, I realized it starts with gasification, a process that my former employer, Texaco Inc., licensed to numerous facilities in China, going back to the 1980s. So China has deep experience with gasification as an effective and reliable way to turn feedstocks as diverse as waste oil, petroleum coke, low-value coal, and even natural gas into syngas, or synthesis gas, a mixture of carbon monoxide and hydrogen from which all sorts of useful organic chemicals can be produced. One of those is acetic acid (the acid in vinegar.) It turns out that Celanese's new ethanol process is an offshoot of the company's well-established "acetyl platform" for making acetic acid in plants like this one in Singapore.

It's noteworthy that the first ethanol plants Celanese is building are so large. 400,000 metric tons per year equates to 134 million gallons per year, larger than all but a couple of the corn-based ethanol plants in the US. I've also seen hints that these facilities could be expanded to 1 million tons/yr, which would put their output in the same league as the gasoline yield of the smallest oil refineries. That would be truly industrial scale fuel production that conventional or advanced biofuels can't yet match and may never do, because of their much more complex supply chain considerations. That also explains why Celanese could consider building a 40,000 ton ethanol plant in Texas based on natural gas. The supply chain isn't an issue when it's just an existing pipeline. In any case, large scale and low feedstock cost should result in ethanol output that's more than competitive with ethanol from biomass. US biofuel producers eyeing export markets ought to be concerned about the potential competition from Celanese, even if the federal Renewable Fuels Standard (RFS) guarantees them a market here.

My other instant reaction when I heard about this process focused on the potential environmental consequences of producing ethanol from coal. However, as I thought about it more carefully, it occurred to me that processing coal into ethanol using the extremely clean gasification process, which allows for sulfur and other contaminants to be easily and safely collected and disposed of, is probably a lot more benign than burning the same coal to produce electricity, particularly in power plants without state-of-the-art pollution equipment. Assessing the greenhouse gas impact of coal-to-ethanol requires a thorough lifecycle analysis that I have not yet found.

At the same time, it's clear that the environmental comparison to biofuels like corn-based ethanol isn't nearly as bad as suggested by an erroneous comment in a Business Week article on the subject last November, which stated that corn ethanol production "doesn't use a fossil fuel as a raw material." In fact, analysis by the Argonne National Laboratory of the US Department of Energy found that 78% of the energy in a typical gallon of corn ethanol comes from fossil fuels, including coal, diesel fuel, and natural gas. That's why the emissions from corn ethanol aren't much lower than from gasoline, after factoring in the natural-gas derived fertilizer used in growing the corn, the diesel fuel required for cultivation, harvesting and transportation, and the coal and natural gas used to generate electricity and process heat for the fermentation and distillations steps. Ethanol from coal might emit incrementally more greenhouse gases than food-crop based ethanol, but not orders of magnitude more. And I'd bet that a gas-to-ethanol plant would match or beat the emissions from a standard corn-based biorefinery, based on avoiding the need to separate the alcohol product from water. Distillation requires lots of energy.

It's getting harder to draw meaningful distinctions between conventional fuels and alternatives when we can make ethanol efficiently from fossil fuels and produce "drop-in" fuels--synthetic gasoline, diesel or jet fuel--from biomass like sugar cane or algae. I haven't seen how the detailed economics and energy balance of the Celanese ethanol process compare to traditional and advanced processes for producing ethanol from biomass, but I think we're going to be hearing a lot more about this option in the future. I was surprised to see that it even garnered a mention in the White House press release for the President's visit to China earlier this year.

Conference Highlights

I spent the last three days at the annual energy investment conference held by the sponsor of this blog, John S. Herold, Inc. Many of the panels I attended were overshadowed by the enormous uncertainty about the US financial system and pending bailout proposals, along with the Presidential election, the dynamics of which appear to have shifted again. However, the sessions provided some very interesting insights into an important unfolding natural resource play, along with showcasing some nifty applications of existing technology that could help to narrow the gap between growing global energy demand and the stagnating supply of conventional oil.

The two words that I heard most frequently this week were “shale gas”, the development of which just might facilitate achieving some of Mr. Pickens’s ideas about energy security. This is not the kind of shale that has been touted as a nearly unlimited source of unconventional oil, but rather a layer of natural gas-bearing rock that until recently was very difficult to tap. But as several panelists explained, companies have “cracked the code” for drilling into these deposits and producing flows that compete favorably with conventional gas fields in both output and cost. The result could be a modest gas bubble—a period of relatively abundant US natural gas supplies—though it comes with an inherent price floor not far below current levels. So while it is unlikely to rejuvenate struggling gas-based industries such as fertilizer production, for which $7/MMBTU is still quite dear, it could support expanded natural gas use in both transportation and power generation, where it could yield significant environmental and cost benefits.

One of the two technologies that impressed me was featured on the Alternative Energy panel I moderated. One of the founders of DKRW Advanced Fuels described a clever application of off-the-shelf technology that turns Wyoming coal into unleaded gasoline without releasing the vast quantities of CO2 that have made coal liquefaction look unpalatable. This trick is accomplished by marrying GE’s gasification technology (the old Texaco Coal Gasification Process on which I worked briefly as a young engineer) with ExxonMobil’s methanol-to-gasoline process that operated for 10 years in New Zealand, until the natural gas field feeding it was depleted. The output is 87 Octane unleaded gasoline and a pure CO2 stream that will supply the region’s extensive enhanced oil recovery projects, which will effectively sequester it. This scheme creates a double energy benefit: mainstream liquid fuel from America’s most abundant energy resource, and increased output at some of our aging oil fields. Even better, it looks like this can be accomplished with lifecycle greenhouse gas emissions no worse than from conventional oil.

The other technology that caught my attention was presented by an old friend and former Texaco colleague, who is now the CEO of Compact GTL. Instead of using proven gas-to-liquids technology to unlock “stranded” natural gas reserves—non-associated gas deposits far from infrastructure or markets—he aims to apply it to the problem of “distressed gas.” He defines that as natural gas produced in conjunction with oil in projects for which the cost and logistics of traditional methods for handling the gas have become an obstacle to developing the oil field. Previously, such gas would be flared, but that practice is being phased out on environmental grounds. Turning it into synthetic oil could prove cheaper than re-injecting it into the ground, while also shortening the development cycle of some large oil fields. Another double win, if it proves practical.

With the country still debating the merits of expanded oil drilling and looking to renewable energy sources that have not yet achieved the scale necessary to wean us off imported oil and slash our greenhouse gas emissions, the approaches described above can provide a valuable bridge. They could also be real money-spinners, at a time when other parts of the economy are looking pretty sick.

Ending Oil's Monopoly

In yesterday's Financial Times (subscription required for full text) Daniel Yergin suggested that the current oil price spike is creating a historical "break point" for petroleum that will result in the loss of oil's dominance in the global transportation fuels market. The commentary by Mr. Yergin, the Chairman of Cambridge Energy Research Associates articulated a shift that has become increasingly apparent to careful observers of the industry. His conclusion that oil will "share the transport market with other sources as never before" is almost certainly correct, even if oil prices were to revert to $60 per barrel next week. There is an important corollary to Mr. Yergin's analysis that he didn't explore in his FT op-ed: At the same time that gasoline and diesel will have to share the market with other fuels, the primary sources of transportation energy will also become much more diverse, as well. That has important implications for both national energy policy and corporate strategies.

Consider the supply chain for petroleum products. Oil is extracted from underground reservoirs and transported to refineries that separate it into its familiar product categories, while transforming low value portions of the barrel into high-quality fuels and removing sulfur and other impurities along the way. A modern refinery is a complex, expensive set of hardware, but its functions would still be recognizable to an oilman from the 1930s. Even ethanol has retained this model, with corn going in one end of an ethanol plant and ethanol and its byproducts coming out the other end. The new transportation energy market that Mr. Yergin hints at will shatter this model. Oil and its products--and corn and its fuel products--will play an important role for decades to come, but they will compete with synthetic diesel and jet fuel from natural gas, coal and biomass; biodiesel, ethanol and other alcohols from a wide variety of feedstocks and technologies; and electricity and hydrogen from a multitude of conventional and renewable sources, both centralized and distributed.

This new model will break three effective monopolies: of spark-ignition and compression-ignition internal combustion engines, of gasoline and diesel fuel as the dominant energy carriers for delivering transportation energy--and note that ethanol has so far only piggy-backed on gasoline's monopoly, rather than breaking it--and of petroleum as the source of primary energy for most forms of transportation. While the market shares of all three of these monopolies are in the high 90%'s today, the signposts of change are all around us. Biotechnology promises to break down the cellulosic material that gives plants their rigid structure and turn it into ethanol and other fuels, but it could eventually give us plants that excrete market-ready fuels. Better batteries will give consumers the choice between plugging in and filling up, but they could also facilitate the much wider adoption of renewable electricity from intermittent sources such as wind and solar power. And fuel cells running on hydrogen might yet provide a practical and more efficient way to turn chemical energy into useful work onboard the vehicle, powering electric motors that will become increasingly ubiquitous on all ground vehicles.

A decade ago, this scenario was just that, one possible future outcome of a number of competing trends and uncertainties. Now, thanks to the combination of concerns about climate change and energy security, and the practical problems of $130 oil, some version of it seems more plausible than the unchallenged continuation of those three "natural monopolies" for another generation. Whatever its other faults, the "farm bill" just passed by the Congress over the President's veto takes a step in that direction, by reducing the subsidy for corn ethanol, the so-called Blenders' Credit, from $0.51 per gallon to $0.45 and using the savings to fund a $1.01/gal. direct subsidy for producers of cellulosic biofuel.

As Mr. Yergin points out, oil "is not going to fade away soon." It will take time to turn over car fleets and move new fuel processes out of the laboratory, through demonstration-scale testing, and into full commercial production. But as frustrating as the wait for these new technologies and fuels may seem, while Americans pay $4 at the pump and Europeans pay the equivalent of $8 per gallon, this energy crisis--unlike the one of the 1970s and early 1980s--might just put in place the means of averting all foreseeable future energy crises centered on oil, by reducing the status of oil producers to that of merely one transportation energy source among many.

Rethinking Nuclear

Periodically, I've noted the growing interest in nuclear power as a source of low-greenhouse-gas energy, and the way this issue is splitting environmentalists. Designs for generating electricity safely and reliably using nuclear fission have improved steadily since the last big wave of nuclear construction peaked, at least in the US--Japan and France built dozens of plants after we stopped. But while the technology has advanced, I wonder whether our mindset about it remains stuck in the model that prevailed in the 1980s, when large, centralized power plants were the norm everywhere. Is it time to rethink the application of nuclear energy to maximize its leverage on reducing greenhouse gas emissions?

What if nuclear power hadn't been discovered before World War II, and instead had emerged from the laboratory only a few years ago? How might we consider exploiting an energy source with its properties today, without the baggage of the last sixty-plus years? Is it pre-determined that the only way to tap the energy of the atom is in 1,000 MW increments? The record of the US nuclear naval propulsion program suggests otherwise. Consider the difference between coal-fired power plants and those burning natural gas. There are important economies of scale in the transportation and handling of coal, and in the sizing of boilers, that create a strong bias towards large plants. In contrast, gas-fired power comes in a wide variety of sizes, from under 100 kW to hundreds of MW, at least in part because the fuel infrastructure is so simple. So is nuclear power more like coal or gas in this regard? It's probably somewhere in between, after you factor in the need to contain the radioactive fuel.

I can think of lots of ways to use a medium-sized source of intense heat that doesn't need to be re-fueled continuously, and making power is only a sub-set of those applications. Combined-heat-and-power (CHP) at facilities that need large quantities of process heat and currently burn huge amounts of fossil fuels might be a better, more efficient candidate. Oil sands production shares those characteristics, and there was a brief flirtation with this idea several years ago. And I recently ran across the website of a company that wants to use nuclear energy in an even more novel way, for coal-to-liquids. CTL requires both process heat and large quantities of hydrogen to upgrade solid coal to liquid hydrocarbons, and using nuclear power, rather than coal or natural gas for these purposes would shrink the net emissions of CTL fuels down to roughly the same range as petroleum products.

Realistically, we can't ignore the legacy of the Cold War or nuclear accidents, nor the prospect of further weapons proliferation or WMD terrorism. But that doesn't mean we shouldn't examine where a "clean sheet" approach to nuclear energy might take us, particularly if it involved smaller scales, quicker implementation, and fuel cycles that are less vulnerable to accidents or proliferation. That might not be sufficient to convince critics to turn in their "No Nukes" signs, but it would go a long way towards convincing the public that nuclear energy is a viable element of our low-carbon energy future.

Peak Preparation

Following on from Friday's posting on the uncertainty about how close we are to a peak in global oil production, I want to focus on a question I think is actually more important: However close we are to a peak--whether it is already here, or 5, 10, or even 20 years away--are we doing enough to prepare for the possibility of one? The short answer is no, but that doesn't mean we aren't doing anything. In fact, many of our strategies for addressing climate change and energy security also provide some insurance against the consequences of Peak Oil or its forerunner, a sustained period in which liquid fuel supply doesn't grow as fast as potential demand, and the oil-market discontinuity that would trigger.

On a basic level, oil is important for two main reasons. It is the source of most of our transportation fuels and many useful petrochemicals and lubricants, and it also accounts for 35% of the world's primary energy production. Preparing for a gap between oil supply and demand requires addressing both of these aspects of oil's value to the economy, and in that regard it dovetails neatly with the concerns about global warming and energy security that are prompting big changes in our energy policies.

For example, while improved energy efficiency is a primary strategy for countering climate change and reducing oil imports, it looks equally important in preparing for a future oil shortfall and price spike. Peak Oil worries could lend urgency to the debate over CAFE and appliance energy standards. At the same time, efforts to expand biofuels production and bridge electricity into transportation via plug-in hybrids and electric vehicles, though driven by emissions and energy security calculations, are also excellent prescriptions for mitigating Peak Oil's impact and even delaying its onset.

There are a few areas in which this one-size-fits-all logic fails. The conversion of solid and gaseous hydrocarbons into liquid fuels--CTL and GTL--looks quite useful from a Peak Oil perspective. Viewed through a climate change lens, however, it looks like an expensive diversion or downright counterproductive. And while natural gas has oddly fallen from favor with those most concerned about climate change, despite its relatively low CO2 emissions, improving our access to gas (imported and domestic) looks like another key leg of the energy security/Peak Oil axis. If Peak Oil is a significant risk, we would certainly not want to face it in the midst of an emerging natural gas crisis.

For me, all of this boils down to effective large-scale risk management. For the next decade Peak Oil remains a big uncertainty, not a given, but prudent planning must take it into account. Where it reinforces other concerns, it may prompt accelerated timetables. Where it conflicts, as on some aspects of climate change, we need a candid debate about which problem looms larger, and which consequences would be most damaging or costly. At a minimum, we should improve our monitoring capabilities, including the means of auditing global production and reserves data for all liquid fuels, not just the conventional oil on which most Peak Oil predictions are focused.

How Near Is the End?

Although Peak Oil has faded somewhat as a "front page" issue this year, after a couple of years in the limelight, yesterday I received a question suggesting that a peak was either imminent or already upon us. That prompted a quick review of global oil production data to see whether there had been any changes that might support that view. I'm generally agnostic on the whole idea of an imminent geologically-driven peak in production, as distinct from one that might occur as a result of OPEC policy or problems queuing up the necessary drilling kit, personnel and investments to keep production rising ahead of demand. As complex as this issue is, however, there is one statistic that I think provides a pretty good barometer for the proximity of a peak; based on that measure, at least, we're not there yet.

Without going through the whole Peak Oil argument again, technical and otherwise, I want to focus on one aspect of peak oil that ought to be fairly non-controversial, among both peak adherents and peak skeptics. The global distributions of oil reserves and current production are remarkably different, as a function of the upside-down economics of the oil industry, in which the low-cost producers constrain their output and the high-cost producers go flat out. OPEC countries (excluding the newest member, Angola) hold 60% of the world's proved reserves but account for only 40% of production. Fundamentally, if there is a geologically-based peak in oil production waiting for us, OPEC is much farther from it than the rest of us, so it must manifest first in non-OPEC production.

So what do the numbers tell us? Has non-OPEC production stalled or gone into decline, as many expect? After looking at the most recent data available from the Energy Information Agency (EIA) of the US Department of Energy, the International Energy Agency (IEA), and the just-released BP Statistical Review, the clear answer seems to be no. Between 2004 and 2006 non-OPEC production grew by an average of 0.5%/year, and the IEA expects growth >1% this year, in a predictably lagged response to four years of sustained oil high prices. I don't see how that would be possible if we were as close to a global peak as pessimists believe.

There are two important caveats about the above figures, and if I didn't mention them, I know my readers would keep me honest. If you subtract from non-OPEC production the contribution of Canadian oil sands projects and the rising output of Angola, the residual trend looks like a plateau, at least over the last three years. But it no longer makes sense to look at non-OPEC supply without including oil sands--which are now a fact of life--just as we routinely include natural gas liquids. For that matter, anyone looking at peak oil ought to be counting the growing contribution of biofuels and any CTL or GTL that comes along, because what matters to the market is total liquid fuel supply, not just conventional oil. As to the change in Angola's status, it highlights OPEC's recent cleverness and reinforces the significantshift in market power that is underway.

The net result of all this leaves us just as uncertain as we were before about the timing of a future peak in "oil" production, but increasingly vulnerable to OPEC's production decisions. While much of that vulnerability is the inescapable result of the maturity of the producing basins in North America and Europe, some of it is self-imposed, and we ought to be doing some serious soul-searching about the consequences of that choice. Improved fuel economy and more biofuels will help, but we could dig our way out of this hole faster with some help from the oil we've chosen to place off-limits to development.

Security vs. Emissions, Round I

It's probably premature to describe yesterday's Senate votes on energy as another turning point for coal in this country. Two separate amendments promoting coal-to-liquids (CTL) were voted down by healthy margins, as described in today's Washington Post. That doesn't automatically derail the industry's interest in producing liquid fuels from coal, but it seems to ensure that the final energy legislation coming out of Congress this year will include neither federal funding for CTL, nor a privileged place for its output within the liquid alternative fuels mandate of 35 or 36 billion gallons per year. While the US clearly can't ignore the energy bounty of the coal under our land, it looks increasingly likely that concerns about climate change will constrain coal's future contribution to sectors in which most of its CO2 emissions can be prevented from entering the atmosphere. That represents a real energy milestone.

In arriving at yesterday's decisions, it might appear that the Congress is expressing skepticism about the potential for Carbon Capture and Storage (CCS) technology to put CTL on an equal emissions footing with petroleum products. I don't think that's the case, because Senator Dorgan's SAFE Energy Act of 2007 (S-875), which is the centerpiece of the Senate's current debate on energy policy, spells out the importance of CCS in its charge to the Secretary of Energy to undertake R&D for CCS. If anything, the importance of CCS as an enabling technology for coal (and shale) has been elevated, at the same time that CTL has been recognized as a less attractive path towards low-emissions energy than biofuels or electricity.

I don't mean to rehash yesterday's posting, which addressed some of these same issues. Nor do I think that these votes rule out CTL entirely, because it could still emerge on a purely commercial basis. But I think it's worth noting that on its first opportunity to choose between the two main priorities that have emerged for national energy policy, enhancing energy security and reducing greenhouse gas emissions, the Congress has set the latter higher than the former. That could create a precedent that will carry beyond the current Congress and into the next Administration, regardless of who wins in November 2008, Democrat, Republican, or independent.

Do No Harm

The Senate and House of Representatives are both feverishly working on new federal energy legislation, and it's a reasonable bet that a bill will end up on the President's desk within a few months. However, it is still anyone's guess as to precisely what provisions will survive or be added along the way, as the process converges toward an eventual conference to iron out differences between the two bodies' differing energy visions. As the final legislation takes shape, however, we can only hope that our elected representatives will see the wisdom of adopting the credo of at least doing no harm. The potential for wasteful and counterproductive energy policy is enormous, particularly in two areas: the functioning of the petroleum products market and the promotion of alternative fuels.

The API ran a full page ad in today's Washington Post with a tag line of, "It's 2007, not 1977." That echoes a theme I've expounded here for several years. Many of the measures introduced to deal with the energy crisis of the 1970s were either ineffective or downright harmful. We should have learned from that experience, and from the much more successful market-oriented approaches of the subsequent decades. While fuel prices may be high again, we have seen none of the incredibly disruptive gas lines and runouts that plagued us then. In particular, the "anti-gouging" provisions espoused by some in Congress look like standby price controls, aimed at the point in time when the ability of the market to rebalance supply and demand is most essential, as we saw after the hurricanes of 2005. This idea clearly fails the "do no harm" test.

Turning to alternative fuels, it's rare that I agree with the editors of The New York Times on energy policies, but their editorial of May 30th on the impact of coal liquefaction on energy security and climate change was spot on. "A policy designed to solve one problem should not make the other worse," they said, citing the high greenhouse gas emissions associated with coal-to-liquids (CTL) plants. A recent posting on the Clean Car Congress site provides useful supporting data from a Carnegie-Mellon study comparing CTL, conventional fuels, and plug-in hybrids.

So in this regard, it is one thing to codify a greatly increased biofuels mandate that relies on production from unproven cellulosic ethanol technology to meet its long-term goals, but quite another to turn the understandable ambitions of coal-state legislators into a national policy that would double down our bet on the equally unproven technology of carbon capture and sequestration (CCS.) Even if cellulosic ethanol didn't take off as expected, we would still end up with liquid fuels that--however costly at the pump and the supermarket--could reduce both our oil imports and our greenhouse gas emissions by modest amounts. However, if we went ahead with CTL, but CCS proved either ineffective or uneconomical, we'd end up with a synthetic fuels industry that would roughly double our greenhouse gas emissions per gallon of gasoline or diesel. That would make a farce of any national effort to reduce those emissions via cap-and-trade or some other mechanism.

If we are indeed headed for a "grand compromise" on energy that would incorporate meaningful elements of energy efficiency and conventional and alternative energy supply, then those crafting a compromise must hold firm in excluding provisions that would sabotage either the ability of the fuel marketplace to respond to sudden shocks, or our first steps toward reducing our enormous greenhouse gas emissions. In the give-and-take world of Capitol Hill that won't be easy.

Multiple Choice Process

This morning's email included a link to a Business Week article about a slate of new, non-traditional ethanol projects, including several based on the gasification of biomass. One of the great things about gasification is its ability to handle a wide variety of feedstocks, extending beyond oil, coke and coal to include, in this case, yard waste, wood chips and citrus peels. Although the article isn't very clear in describing the path from feed to gasification to end product, it suggests that some of these projects will gasify biomass to make hydrogen, ethanol and methanol. This has me scratching my head, because once you have the synthesis gas, or "syngas", that a gasifier produces, you can make essentially any hydrocarbon you want, particularly the long, straight-chain hydrocarbons of diesel fuel. Why use this expensive process to make challenging and problematic fuels, instead of one that is 100% compatible with current energy systems?

One of the biggest problems plaguing alternative fuels is infrastructure and end-use compatibility. Ethanol cannot be shipped through petroleum products pipelines, so it gets to market via costlier truck, rail and barge routes. Methanol faces similar constraints, and is a neurotoxin, as well. The logistical challenges facing hydrogen are even more daunting. Then, once these fuels reach a retail facility, the only current option for their use is in specially modified gasoline engines--or in the case of ethanol in low-volume blends with gasoline, in which the lower energy content of ethanol reduces fuel economy and vehicle range. This hardly sounds like the way to maximize the energy benefit of biomass.

Synthetic diesel fuel is another story. Compression ignition engines are typically 30% more efficient than the spark-ignition ones used with gasoline, and the properties of ultra-clean diesel from the Fischer-Tropsh synthesis process allow them to run optimally, with lower pollution than from petroleum diesel. If European-style diesel cars, with state-of-the-art particulate cleanup technology, take off in the US, demand for this kind of fuel will grow rapidly. It would also blend nicely with biodiesel, which still can't be used year-round in many northern markets, because of its poor low-temperature properties.

In terms of greenhouse gas emissions, since every carbon atom in the biomass gasification feed will ultimately result in a molecule of CO2 emitted to the atmosphere--without sequestration at the gasifier--the differences in overall emissions for the various fuel options described above depend on the efficiency of transportation and end use. Diesel handily beats hydrogen and alcohols on both counts, unless the hydrogen is feeding a fuel cell. All of this suggests that we may need to rethink our definition of biofuels to encompass any fuel from biological sources, not just those that chemically resemble current-generation biofuels, such as ethanol from corn or cane.

Biomass-to-diesel looks like a promising way to tap the environmental and energy security benefits of biofuel, even though its product is hard to distinguish from diesel made from non-renewable feedstocks. However, before climbing on the "BTL" bandwagon, we should withhold judgment until some of these plants have been built and run for a while. The solids-handling end of the gasification business can be tricky, particularly when dealing with material of inconsistent quality and characteristics. That could turn out to be a much bigger challenge for biomass gasification than the molecular engineering process that sits on the back end of these facilities, but which has been proven in over 80 years of application to oil and coal.

Replacing Today's Oil

If money were no object, and we had unlimited access to engineering and construction capabilities and materials, I'm confident we could produce enough synthetic oil from unconventional resources and coal-to-liquids processes to displace the 10 million barrels per day of petroleum that the US currently imports. Wood Mackenzie, a British consultancy, recently released a report indicating that global unconventional oil resources--oil that is either too heavy and viscous to produce normally, or that is bound up in oil shale or oil sands--stand at 3.6 trillion barrels, roughly three times current estimates of proved conventional oil reserves. But as the Financial Times article above describes, producing this bounty will be no simple matter. Even if the technology were fully proved, as it is in case of oil sands and Venezuela's Orinoco heavy oil deposits, there are other significant barriers to reaching this energy future.

Aside from the direct environmental impacts of these conversion processes, which are inherently higher than for conventional oil and gas, and are beginning to receive serious scrutiny, there are fundamental problems of attracting the capital and other resources required to build such complex facilities on the required scale, and doing so rapidly enough to fill the expected gap between conventional oil supply and aggregate worldwide petroleum product demand. Some of this investment is already underway. The output of major new oil sands facilities in Canada has become part of the base-case assumption for global supply over the next decade, and it is already starting to have an effect on petroleum pipeline infrastructure in the Midwest, where much of it will go. But as I've discussed in previous postings on this topic, the oil sands projects also illustrate many of the practical constraints inherent in the large-scale production of synthetic fuels, in terms of their use of land, water and natural gas supplies, skilled labor and even housing. While the US could certainly supply more of these factors than Canada's smaller economy and population are able to, environmental and local permitting seem likely to create bigger hurdles here than up north. Interestingly, biofuels share some of the same potential limitations, as they scale up to compete as the incremental supply into the world's transportation fuels market.

As long as global demand for liquid fuels continues to grow, these challenges will compound. At the same time that geometric growth steadily increases demand, it drives cumulative consumption to levels that will approach even the enormous endowments of coal and unconventional oil within a few generations, while liberating a comparable tonnage of carbon into the atmosphere, which is on track to reach double its pre-industrial CO2 concentration sometime between mid-century and 2100. For these reasons, fuel efficiency remains a critical component of any energy security plan, whether it is based on biofuels, synthetic fuels, petroleum, or a combination of the three. But efficiency, too, will take many years to bear fruit.

Without resorting to central planning and "industrial policy", we will be asking the market to allocate $20 trillion in energy investments over the next couple decades, within a geopolitical context that looks at least as complex as anything we saw in the 20th century. Unless we want to make the problems we face today even worse, the result of all that investment can't just look like a bigger version of the status quo.