Friday, September 24, 2010

Does the fossil fuel industry need innovation?

Most of the posts that I write tend to deal with the technical side of energy production, rather than more philosophical discussions. However I was recently sent a copy of the new Strumsky, Lobo and Tainter paper on “Complexity and the Productivity of Innovation.” (H/t Nate Hagens). Since, in my day job, I have been one of those innovators in the field of energy (and since also I am going to address this as part of my talk at ASPO in a couple of weeks) I thought I might pen some thoughts that respond to the paper, and explain why I don’t think that the picture is nearly as bleak as the authors seem to suggest.

The point which the paper seeks to make is that, over time, innovation in a field becomes harder, as the early obvious inventions get made and it becomes more difficult to make significant further advances, requiring larger and more complex teams to work longer hours for less overall gain. The metric that is used in the paper is the number of patents that are generated in a field, with that assignation being made by the US Patents and Trademark Office. The authors note that, over time, the number of patents per inventor has declined, while the size of the patenting team has increased, and that the patents per inventor holds true in the fields of energy that include gas, power systems, solar and wind.

My field of knowledge in the area fits more into coal, oil and natural gas production – which may be a field so small that it falls under the radar as regards USPTO classification, certainly in many categorizations of engineering these days mining seems to be a forgotten word and fossil fuels a forbidden topic. But that gripe aside, let me talk about the process of innovation and invention and give some reasons why I disagree with the authors.

There is a saying that, though true, is not considered much because it may have been overused; “necessity is the mother of invention.” (And my thought strays to the day we were scheduled to run a demo for a TV station on a technique for landmine detection and when we ran a test that morning with an “improved” set-up it didn’t work. We had about three hours to work out why it did not, find a way to get it to work, rebuild the tool, and make it work before the crew arrived– our answer aired that night.)

In the field of digging things out of the ground there hasn’t been a whole lot of what might be called revolutionary thought in the past couple of millennia. This has been mainly because the old ideas of a pick and a shovel seem to work remarkable well. True, in the process the pick has moved from being a single point on the end of a manually wielded tool to one of a multiplicity of picks laced around a cutting drum of a rapidly advancing machine, but the principle remains the same. Energy efficiency in the change has gone down, but overall production rates have soared, as machines have got larger and more powerful, when mining underground. On the surface, the load that a single shovel can move went from the few pounds that a miner moved to the 165 tons of a single scoop of “The Captain” 180 cu yd mining shovel bucket.

Thus there has not been a lot of necessity for innovation. What we have is relatively cheap per ton, and it works. Now that is not to say that there isn’t an occasional need for innovation. Back in the 1970’s there were a series of coal mine explosions that killed miners as a result of sparks generated from the picks of mining machines. We developed a machine that used high pressure waterjets to cut the coal, rather than picks, and the technology was picked up by industry to the point of underground trials in Germany. We also worked with CalTech to develop a version for room and pillar mining. In both cases we showed that the tool could mine coal safely and productively without sparks or dust and at energy levels below that of existing equipment. But the world market was considered by GHH to initially be only 14 machines a year, not enough to justify the investment. The need was not great enough, and we looked for more marketable products than large mining machines. Which led on to the development of waterjets for drilling and cutting. But in terms of the overall impact on mining waterjets did not catch on. And patenting the technology proved difficult. Jason and his Argonauts visited a hydraulic mining operation, in what is now Georgia, millennia ago (and stole the sheepskins they were using to catch the gold in the bottom of their flumes). So what is new, apart from raising the pressure a bit?

As a result of a lack of need, there has not been a great deal of significant interest or funding in the actual digging process itself. Back in the energy crisis that ran into the 80’s, increased research funding led to the development of the polycrystalline diamond compact bit, which has since made drilling in hard rock much easier. But after the collapse of oil prices in the ‘80s funding and interest faded away.

So my first argument that I would raise is that there is, as yet, not that much necessity for invention in the fossil energy field, but that when it comes there are lots of different avenues that can be taken. As an illustration, we removed the rock for the Omnimax Theater under the Gateway Arch in St Louis by breaking it out into single large lumps weighing between 1,000 and 2,000 lbs each. Creating a single crack around the blocks took a whole lot less energy than breaking it down into the fragments of a typical mining operation. And it made it a whole lot easier to handle. The necessity was that we had to get the rock out without any of the visitors noticing, so we couldn’t use blasting or any more conventional technique. And as a second example, if you disintegrate rock into its constituent grains at the mining machine, you can separate the valuable mineral there, leave the rest behind, and only haul and process the valuable stuff. Saves a rather large amount of energy, over the increase needed for the conventional process.

Anyway, to get back to the paper, the other thing that is more integral to their discussion relates to how things get invented. In many of the cases I am familiar with there is originally the work of one person. That person has an idea, and some sort of drive to get this idea to work. It is only when a group of other folk has been persuaded that the idea is good that it starts to move toward commercial use – and I would agree with others that a realistic time from innovation to widespread use is about 20-years. And this is the path of revolutionary ideas, the ones that change the paradigm of an industry.

I do not think that this type of innovation is properly recognized in the paper. What the paper discusses is the more conventional evolution of technology once the original concept has moved into use. But it is the revolutionary concepts that will move us forward.


  1. It would appear that the payoff of innovation gets smaller as the state of the art approaches the ultimate limits.  Watt's external condenser was a huge improvement over Newcomen's one-chamber system, but today it's very hard to beat even the conventional supercritical steam plants.

    The one thing I can see which rewards innovation despite this is resource constraints.  There's no reason to innovate in transparent conductors while indium is plentiful and cheap, but now that it's looking scarce, innovation in graphene is pursued because of the likely rewards.  Flat-panel displays using graphene instead of indium are reportedly only a few years away, and PV cells are not likely to be far behind.

    There is also the experience curve to consider.  Many things are currently quite expensive compared to the materials needed to make them.  These tend to be new products, with less experience and less optimization in their manufacturing processes.  As e.g. traction batteries go from a market of a few million kWh/year to the order of a billion kWh/year, we'll see process improvements (rewarded by the size of the market) drive costs down.

  2. I have a lot of respect for Tainter. His 20-year old book, "Collapse of Complex Civilizations", remains a great contribution -- although his obvious conclusion that we need to simplify and contain the size of government has been ignored.

    But is it possible that Tainter & his co-workers have been using the wrong metric on innovation?

    Compare the automobile of today with that of 25 years ago. Same basic engine/wheels configuration, but there has obviously been a huge amount of improvement based on new ideas and innovation. Similarly, the flat screen serves the same purpose as the traditional cathode ray tube, but does so much more conveniently. Boeing is making a more efficient carbon fiber airplane. Mobile phones have gone from the size of a suitcase to the size of a wallet -- and now include a camera and an internet connection!

    These innovations have big impacts. Simply counting the number of patents issued and the size of the teams developing these innovations ignores the othe side of the equation -- the effects of those innovations.

    Eng-Poet is right that it becomes more difficult to innovate as a technology becomes more mature. And that is an issue with conventional fossil fuels. But we are only starting to see innovation on unconventional carbon-based transportation fuels.

    Back to Tainter's 20-year old point -- we need to cut the complexity of government (taxes & regulations), and unleash the full creativity of a planet full of human beings.