Half a century of pushing the Hydrogen Economy idea, some personal memoirs

Cesare Marchetti, keynote speech
World Hydrogen Technology Convention
Montecatini Terme, 6/11/2007


Some years ago I made a study of the history of basic inventions and innovations in the context of the social and economic system. [Full text, scan PDF 419 Kb]. What I discovered was that ideas, experiments, diffusion, and success are not only quantifiable   in time with neat logistic equations, but also occur under the strict control of a periodicity of about 55 years called a Kondratiev cycle. Furthermore, I discovered that successful inventions and innovations occur during precise periods of that cycle. To give an example:  I studied in detail the behavior of the nuclear energy industry in the context of other basic innovations like cars and trains. It was not opposition to nuclear power that caused nuclear construction to dwindle in the nineties.  In actual fact, nuclear construction strictly followed the rules internal to the cycle mentioned and will pick up speed in the current cycle that formally began in 1995.  [Full text, part1: scan PDF 888 Kb] [part2: scan PDF 1147 Kb].


The system treats    innovations in precisely the same way. A new idea has to fight to penetrate minds and opinions until it takes root and is finally accepted naturally with all the usual pros and cons. This process takes about 55 years or at least till the end of the current Kondratiev. In the next cycle, commercial penetration starts and reaches a certain level at the end of the cycle. In the cycle after that, penetration will be complete. As everyone knows the history of the automobile, this is easy to frame within a temporal scheme of this kind.  Today, the use of hydrogen as the main vector for energy is certainly a basic innovation; it was born in the right period within the previous Kondratiev that started in 1940. It should flourish now with the current cycle that began in 1995 and will end in 2050.


As to how the concept developed, I have been “swamped in hydrogen” since the beginning of the 1950s when I was working to develop processes to produce heavy water. These processes are based on the fine differential chemistry of hydrogen isotopes, and I had to learn about all the ticks and tricks of the hydrogen molecule. Working in a nuclear context at a time when there was blind enthusiasm for everything to do with nuclear, I was tinkering around with what to do about the production of food. Agriculture is a big machine with enormous territorial, political, and especially ecological impact. But chlorophyll is central to the biosphere because it produces hydrogen by splitting water using solar light. If nuclear could shut down the coal mines that have been so damaging to humanity, why couldn’t for the toils and spoils of agriculture do something similar?


At the time adenosyntriphosphate (ATP) was emerging as the general energy carrier in living things. I thus started looking at ways of recharging adenosindyphosphate (ADP) electrolytically.  This may work, but as I studied the process of photosynthesis, I saw that ATP is produced together with H2 and that, together, they run the energy of the plant. As the solutions that nature has adopted tend to be the final product of a stringent selection process, H2 did appear to be the candidate energy vector. A simple search showed that hydrogen studies had penetrated all sorts of areas where energy is used. Absolutely incredible: no stone had been left unturned. Even a phosphorous had been developed to fluoresce in air when some hydrogen is present. Obviously much of the attention was on transportation; after all, there were precursors—the diesel engine of a Zeppelin had been fed with air enriched with H2 from the ballast in 1927 when crossing the Mediterranean, and a German inventor, Erren, had rigged up trucks to run on hydrogen during WW2 in Germany.



However, there were some missing pieces in the jigsaw puzzle. The first one was in trying to create a system where all the pieces fit together in space and time. Trying to do that showed the very limited number of ways of producing H2 on an extremely large scale without possibly using fossil fuels. We were at the end of the 1950s when I went to Euratom to direct a research division in Ispra. Being in a nuclear environment I sold the argument that nuclear energy could not be the real primary energy source that everybody then thought it could be, if it was limited to electricity production, because electricity could end up absorbing half the primary energy.  With total energy consumption then doubling every 25 years or so, nuclear would spare just one doubling (i.e., delay primary energy problems by just 25 years).


Something that people tinkering with energy problems do not seem to have realized is the sheer size of the subject. Energy is by far the largest industry in the world in every way. Half the ships currently navigating the oceans carry fuels, coal, oil, and gas. As two- thirds of humanity will try to follow the first one-third, in a few years their consumption will have increased by an order of magnitude.  Think of India and China and the speed of their growth. At the time this was very clear in my mind, and I discarded all ideas that could not be scaled to very large capacities. The analysis I had done of the evolution in size of synthetic ammonia plants and power plants did show, in fact, that the size of   plants adjusts to the size of the market that they “see.”  They will thus grow even if, for example, transport costs decrease. As low-cost shipping has made the energy market into a single world market, plants and processes must also be seen in a world context. As we will see later, the “energy island” I proposed to the Japanese in 1973 as a final target for nuclear hydrogen production will export as much energy in the form of liquid hydrogen (LH2) as the Middle East in the form of oil and gas.


These general considerations strongly limit the technological choices we are left with. Nuclear as a primary energy source appears inevitable. It is very dense in terms of volume per kilowatt (kW), and it is always there working at full steam, reactors can have now 95% availability, and it is cheap as it scales. At the beginning of the nuclear age when costs were transparent, nuclear plants did scale according to the square root rule, like chemical plants. So if the specific cost per kW of an X4 plant are reduced by one-half the costs of an X1 plant. My friends at General Electric and I did sketch high- temperature reactors (HTR) 100 times larger than the present ones with, in principle, no stops. In the 20 years or so I spent with nuclear institutions I used to tease electrical engineers designing generators by asking what, technically, was the maximum size possible. I always got the same answer, double the present one. But in these 20 years reactors actually grew almost eightfold.  Incidentally, the Edison Jumbo Dynamo had a power of about 10 kW, and a little more than 100 years later, the current electric generators are one million kW.  Curiously, the overall size is not very different; the densification of power is monstrous. My simple criticism of renewables is that in comparison they are thin and unreliable. Incidentally, as we will see, the energy island scheme puts nuclear in the club of renewables.


The second inevitable choice for me was thermochemical watersplitting, a chemical process to decompose water using nuclear heat. As a chemical process it will scale, leading the technology to lower and lower costs. The most obvious solution, electrolysis, was discarded because it does not scale apart from the fact it is a two-level process with inefficiencies piling up. I keep  referring to a global system. If somebody wants to produce hydrogen for his car using a windmill and an electrolyzer, I will applaud.  Thermochemical  watersplitting did not exist at the time—I’m talking about the mid-1960s  and we had to invent even its name. So I asked one of my clever chemists, Gianfranco de Beni, to invent it. We had infinite discussion on thermodynamics but finally he came out with a process that we baptized Mark-1 after the British way of coding Jaguar models. The cycle was not very appealing but it did fit the constraints of using a heat source with a maximum temperature of 800°C and recycling all the chemicals used in the process. It was the process n1, and the prolific de Beni did invent another dozen schemes in the following years some of which are still being studied in various labs around the world.


In the meantime others had started working in the same direction, and in 1968 I felt there was sufficient material in existence to organize a roundtable on thermochemical watersplitting.           [Full text, part1: scan PDF 974 Kb] [part2: scan PDF 1369 Kb] [part3: scan PDF 1552 Kb].

The roundtable was a success as the participants were competent and enthusiastic. We all left with the belief that the problems were tricky but solvable. The central difficulty is that water is a hard molecule and requires a few thousand degrees to crack thermally. This was really a free energy problem, and the final solution may lay in identifying a molecule with a very large free energy change with temperature. We found something of the kind in sulfuric acid cracking that almost made it, so our research concentrated on processes to close the cycle. The exergy required to close the cycle is relatively small, but the chemistry is tricky and there are various versions of it. So most of the work in Ispra was concentrated on sulfur cycles. As I did publicize the work, other people became interested, for example, General Electric and General Atomics. There was in fact a rush of research in these years that later thinned out. Incidentally, we found a precursor, Professor Funk of Kentucky University who, during World War II, proposed a fuel plant for the army next to the front to reduce the transport of fuels based on nuclear and hydrogen. He is still fighting for thermochemical cycles in this second rush linked as usual to the new Kondratiev cycle that started in 1995. He recently reviewed about 120 cycles from Ispra, the United States, and Japan. You can see how widespread this research is if you call up “thermochemical water splitting” in Google. The Japanese are working hard and having good results, and their UTO-3 cycle, for which they have a pilot plant, is very clever and promising


At the beginning of the seventies the president of JAERI, the Japanese nuclear organization, did transit through Ispra and I had the honor of lunching and chatting with him. The Japanese were developing a high-temperature reactor and had the intention of carrying out methane steam reforming with the high temperature reactor they were building as the Germans were doing with their pebble-bed reactor. He was instantly excited at the prospect of reforming the steam itself, and that was the beginning of a fruitful interaction with Japan. For them I wrote what I consider to be the manifesto of the hydrogen economy. As they have the sun in their flag and the emperor at the helm, they like to have a very long term objective with a clear roadmap. I gave them the dream of becoming a sort of Middle East, exporting energy to everybody on top of becoming energy-independent. The following paper was published in a Japanese journal: 



Marchetti, C., 1973
Hydrogen and Energy,
Chemical Economy & Engineering Review, 5(1):7–15, January
Full text, scan PDF 968 Kb]


It contemplates an hydrogen economy based on energy islands located on Pacific atolls. The islands  export LH2 in cryotankers. Each island has an export capacity comparable to that of the Middle East in terms of energy. Uranium is extracted from the cooling water and the radioactive waste is collected in capsules that selfsink in the basalts of the atoll. This is a bootstrap operation importing only technology and exporting energy. The Japanese were flabbergasted and invited me for a tour of lectures and discussions, including geisha dinners every night. While they were fully convinced, they were reticent about acting; there was nothing to base their work on, and they had to break ground alone. However, they started research on two important subjects, the thermochemical cycles, where they have now a prominent position, and uranium extraction from sea cooling water. The effort was thin but the time long and the results very interesting. The uranium extraction they got at a reasonable price puts the nuclear system into the “renewables” league. They were also developing a high-temperature reactor that is a necessary complement to the thermochemistry.


In spite of my secular view of the development of technology I was a bit nervous about the fact that most research on hydrogen concentrates on end uses and, in my opinion, precious little is dedicated to non-fossil primary energy production. So I tried to suggest a hybrid that may start the process before  thermochemical cycles come on the market. The idea is simple. At a certain point a circuit for steam reforming methane was added to the German pebble-bed reactor that Professor Schulten had been promoting for decades and that had an excellent record of safety and reliability as a reactor.  The original idea was to transport high temperature heat by recombining somewhere the product of this reforming. But it is the first part that was interesting to me, that is, a proven technology of steam reforming associated to a HTR. So at a conference in Moscow I made a bold proposal, namely, to reform along the way some of the methane that Russia sells to Europe using nuclear energy. The endothermicity of the process will appear as hydrogen energy of nuclear origin. Furthermore mixing the hydrogen with the flowing methane, the network would distribute hydrogen around Europe


Marchetti, C., 1989
How to Solve the CO2 Problem without Tears,
International Journal of Hydrogen Energy, 14 (8):493--506
Abstract], [Full text, scan PDF 484 Kb]


I also took care to site the plant where carbon dioxide (CO2) could be sold for tertiary oil recovery. The paper was very well received but nothing happened.


The ancient used to look at the flight of birds or the liver of sacrificial animals to view the future. Less poetically, I look at my logistics that describe events faithfully from start to finish. At the beginning of my stay at IIASA in 1973 I stumbled upon a powerful description of the evolution of the energy markets using logistics, the full analysis of which is reported in this paper:


Marchetti, C., and Nakicenovic, N., 1979
The Dynamics of Energy Systems and the Logistic Substitution Model,
RR-79-13, International Institute for Applied Systems Analysis, Laxenburg, Austria
Abstract],[Full text, part1: scan PDF 1001 Kb] [part2: scan PDF 638 Kb]


In one of my fancy explorations I did look at the competition between H2 and carbon in the mean composition of primary energies in the last 150 years. It is well observed that the fuels we use become more and more “light” (i.e., richer in hydrogen) but my analysis was quantitative. The unexpected result was that hydrogen would penetrate the primary energy system logistically and reach the ratio of 4 in the next years. Because the highest ratio is in methane and is 4, this means that a new source of H2 is predicted starting from a non-fossil fuel source (i.e., water).


And here we are. When the system speaks, things happen.