Using A Fuel Cell To Produce Hydrogen

By Michael Johnston

Copyright 2003

From the time of Michael Faraday the idea of using hydrogen fuel to produce hydrogen fuel has lingered on the lunatic fringes of electrochemistry. For the most part it is dismissed out of hand as just another “impossible perpetual motion device”. The illustration below depicts what such an impossible system might look like.

At first glance this does seem to be impossible for several reasons. Mostly having to do with the Second Law of Thermodynamics and the Conservation of Energy rule. But after carefully considering the operating processes of all components involved in such a system some ideas for potential modifications come to mind which make it look less and less impossible and more like it just might be deserving of some serious experimental research. In this paper I will attempt to explain WHY I think this is so and detail my reasons for thinking as I do. And I will try to do so within the bounds of currently accepted scientific principles.

First let's consider the electrolysis reaction. I won't go into excessive detail here as I have already done that elsewhere in my collected papers on this site and am in the process of writing an even more detailed explanation which will be posted on the site as soon as I finish it. The essential facts are that the electrolysis reaction depends upon a certain minimum voltage (potential difference) being maintained between the two electrodes of the cell. This is for the operation of the cell only. The amount of Hydrogen and Oxygen produced by the cell is determined by the amount of current (amps) flowing through the cell. So at that same applied voltage You can produce a gram or a pound of hydrogen within the same time period depending only on the amount of current which flows through the cell.

In a fuel cell a voltage is produced. This voltage is 1.23 volts (0.8 volts under load) and it is produced and maintained whether there is any current flowing or not as long as Hydrogen and Oxygen gases are present and available to react. This almost seems to be two sides of the same coin and is too often interpreted that way but it isn't that easy. You see, unlike the fuel cell, the voltage required by the electrolysis cell is not identical in every case. It depends on the oxidation potentials of the reactants involved (including the electrolyte being used in the cell), the concentration of electrolyte, the electrical conductivity/resistance of the electrodes and the cell overall and the ratio of positive to negative ions present.

So the minimum voltage required by an electrolysis cell with NaOH electrolyte would be 1.47 volts. This is larger than the voltage produced by the H2 + O2 reaction in a fuel cell and is determined by the difference between the oxidation potential of the Sodium ion (+2.71 volts) and the OH- ion (-1.23 volts) or +1.47 volts. From this data comes the belief that it takes more energy to produce H2 and O2 from water than is given off by recombining the two. This is correct in this particular sense but misleading as it seems to imply that H+ and OH- ions are the only reactants in an electrolysis cell. This is not true as the added energy requirement of the above example proves. After all, the oxidation potentials of a reaction are the same no matter which direction it proceeds in, they just change polarity. So, if the only reactants involved in both the fuel cell and the electrolysis cell were hydrogen and oxygen then the amount of energy released in the fuel cell would be identical to the amount of energy required by the electrolysis cell (+/- 1.23 volts) ignoring resistance effects in both. Again, remember that the above example applies only to an NaOH electrolyte at 1 m concentration. Other electrolytes would be different. H2SO4 (2H+ [SO4]-- ) Sulfuric Acid, for example has an oxidation potential of only -0.98 volts.

The first fact then, which must be understood and accepted in order to see the veracity of the concepts set forth in this paper is that; the fuel cell reaction and the electrolysis reaction are actually not the same reaction. They are similar in that one combines H2 and O2 into water and the other separates water into H2 and O2 but different because in the combination reaction H2 and O2 are the only reactants involved and in the separation reaction the electrolyte is also involved and it is the electrolyte which determines the amount of energy required for the separation. So the hydrogen and oxygen produced are more accurately described as a by-product of the interaction of the electrolyte with the water molecules, not as a direct effect of the electric charge on the water molecules.

Even if the voltage required was identical to the voltage produced the above system would still not work, if constructed as pictured, due to the effects of resistance in the wires and the cell(s) and even if those resistance effects could somehow be eliminated, such a system could only ever produce exactly the amount of hydrogen and oxygen that it would need for it's own operation which would qualify it to be a scientific curiosity but nothing more. This is because of the current that flows in each. The amount of hydrogen produced by an electrolysis cell depends solely on the amount of current flowing through the cell (as long as the minimum potential difference between the electrodes is maintained) and the amount of current produced by the fuel cell is determined by the amount of H2 and O2 that are combined. In a straight DC circuit the current is the same at all points and so the fuel produced would be exactly equal to the fuel consumed in the above illustration.

Ok, so it takes more energy (in volts) to produce H2 by electrolysis (with most electrolytes) than is liberated (in volts) when combining the end products produced by the electrolysis cell. So what? That just proves that you can't use a fuel cell to operate an electrolysis cell, right? Not necessarily. Since we have two different chemical reactions (fuel cell/electrolysis cell) and we can easily see that the energy required/liberated is not the same in both then we realize that we aren't really trying to make a single chemical reaction be exothermic in both directions (the impossible scenario). What we are doing is trying to maximize the efficiency of an entire system of which the chemical reactions are but one component. We must remember too that the chemical reactions in the fuel cell and the electrolysis cell are independent of each other as both are complete oxidation/reduction reactions and both are also independent of the metallic portions of the circuit in that they don't directly react with the metals as they would in a battery.

This is a very simple and yet very important observation. The process of electrical conduction in a metallic circuit is similar and yet at the same time very different from electrical conduction in a liquid/chemical circuit. Once we have a charge flowing in the metallic portion of the circuit we are dealing with a whole different set of rules and we have many more options at hand that can be used to manipulate said charge in such a way that it will be to our advantage.

The illustration below offers a look at what a system which would enable a fuel cell to run an electrolysis cell might look like. Obviously this is not a completed blueprint for such a system, only a starting point but it gives the idea of what would be necessary to create such a system and I will go on to explain the theoretical reasons why I feel that such a system just might work as intended if properly engineered.

The most obvious change from the first illustration is that I have separated the fuel cell and the electrolysis cell by putting them on two different circuits. This will allow us to modify the way in which electrical energy is handled/utilized in each circuit. This will not work with straight DC current though. To overcome this obstacle I propose using a device such as an oscillator to turn the charge in the primary circuit into pulsed DC. It would do this by essentially turning the current from the fuel cell off and on very rapidly without either (a) using any of the available energy or (b) requiring any outside source of energy for it's operation. By doing this we open the way for the next modification; the addition of an induction transformer to transfer the energy of the primary circuit to the secondary circuit.

It is true that we are dealing with very small amounts of energy here and the method that I illustrate may not be the best or most efficient way to accomplish the task at hand so it is intended to be a simple illustration which communicates the concept in easy to understand terms. I have consulted with several people who are fairly knowledgeable in regard to electronics as to whether the circuit above could be made to work as described. In just the metallic portion of the circuit all agreed that indeed this is possible. Some suggested using a MOSFET charge controller and others predicted that it would require custom made components but all agreed that it was possible to do it. Below is one possible design for such a circuit.

What are the potential advantages of constructing a system like this though? Now that we have the system this far we can begin to look at the next set of relevant concepts. We would want to use the transformer to create a situation where the voltage in the primary circuit is increased. This will limit the amount of current that can flow in the primary circuit and, as a result, limit/reduce the amount of hydrogen and oxygen used in the fuel cell. Remember that current in the fuel cell depends on the number of atoms that are being combined in a certain amount of time. The lower the current the fewer atoms that react, the less H2 that is used. For example, let's say that we have a 10 watt fuel cell. Since volts (E) x amps (I) = watts (W) then that would be 1.23 volts x 8.2 amps = 10 watts at no load voltage or 0.7 volts x 14.3 amps if you use the voltage under load value. So if we then use our induction coil to boost the voltage to say, 6 volts, we would then only be able to pass 1.7 amps of current in the primary circuit and the total energy would still be 10 watts.

In the transformer's secondary coil we step-down the above (6 v) voltage from the primary circuit to 2 volts. The transformer changes the entire 10 watts from the primary into a different form and so, using Ohm's laws we see that we can pass 5 amps of current at 2 volts in the secondary circuit, through the electrolysis cell. The end result then would be that we are using 1.7 amps worth of hydrogen fuel in the fuel cell and producing 5 amps worth of hydrogen fuel in the electrolysis cell. In other words we are producing over 2.5 times the amount of hydrogen in the electrolysis cell as is being used in the fuel cell. This seems impossible at first but we are after all, producing 10 watts of energy in the primary circuit and using 10 watts of energy in the secondary so both conservation of energy and the second law of thermodynamics are left unchallenged and yet we are producing enough fuel to power our fuel cell with plenty to spare.

Now we are beyond the “scientific curiosity” stage with this concept and into really useful territory. Imagine the potential of such a system. If the “extra” hydrogen and oxygen are combined in another fuel cell then the electrical energy produced in that reaction is “free” energy. If the water produced by the electrolysis cells is recycled back into the electrolysis cells then you have a portable, self-contained power system which could be used anywhere that energy is needed such as homes, businesses, cars, boats, spacecraft, aircraft etc. Considering those potential benefits I believe that it will indisputably be worth the effort and expense to carry out the experiments necessary to confirm the validity of this concept.

I plan to construct a prototype of this system myself as soon as time and money permit. Until that time I intend this paper to be put into the public domain so that anyone who wants to experiment with the concepts outlined herein may do so freely. In initial experiments it might be easier (cheaper) to replace the fuel cell with a 1.5 volt flashlight battery or several of them in parallel in order to achieve the desired current/power level. This might allow you to work out the details of the electrical portion of the system more quickly. My only requirements are that anyone using these concept notify me of the results of their experiments and, if you either publish the results of such experiments or produce a (working) device intended for sale, based on this research, that you credit me as the originator of the concept.