of this page
The fuel for the Fusor is Deuterium gas, which is injected into the vacuum chamber through a precision regulated valve. Injected means here that the precision valve is slightly opened and the high vacuum in the vacuum chamber (reactor) draws Deuterium gas into the vacuum reactor. The applied high voltage in the reactor ionizes Deuterium to Deuterium ions (Deuterons).
For obtaining a maximally fusion rate a direct relationship exists between the pressure of Deuterium in the reactor (Deuteron density) the applied potential difference and the current. The higher each one of these three parameters, the higher the fusion rate, though an optimum appears to exist for these parameters.
When the parameters have been optimized, fusion for D-D occurs according to the following reactions:
The first stage reactions !a and !b have equal opportunities to occur and yield Tritium (as Tritons), Helium-3 (as a Helium-3 nucleus), a proton, a fast neutron and energy.
The second stage reaction is between Deuterons present in the reactor and Tritons formed in reaction Ia and yields Helium-4 (as nucleus) and a fast neutron. The neutron formed in reaction II has such a high energy that is carries away 80% of the energy formed in this reaction.
The second stage reaction has also the probability for fusion between Deuterons present in the reactor and Helium-3 nuclei formed in reaction Ib:
3He + 2D → 4He (3.6 MeV) + p+ (14.7 MeV) (III)
This reaction yields Helium-4 nuclei and a high energy proton, but the reaction is less likely to occur because the peak energy required for this reaction is much higher than for the D-D reactions.
In practice we usually deal with the reactions !a, Ib and II.
Deuterium can be obtained as gas in lecture bottles and as Deuterium oxide (heavy water). When obtained as heavy water, Deuterium needs to be liberated as gas by electrolysis of heavy water.
Deuterium gas is available in lecture bottles, typically e.g. 38 liters at a pressure of 100 kgf/m2 (atm). It is a highly flammable gas and air shipment of Deuterium cylinders is not allowed. This limits the possibility of buying Deuterium gas and it explains why Deuterium oxide is quite popular amongst fusioneers, despite the fact that an electrolysis unit will have to be built. The costs of an electrolysis unit, however, is more or less comparable with the costs of the pressure regulator that will have to be applied to extract the gas from the high pressure lecture bottle and convert it to a low pressure (image 1).
Image 1: Pressure regulator for Hydrogen
The pressure regulator should preferably be of the two stage type, intended for Hydrogen and produce an output pressure as low as possible.
When operating a spherical Fusor with a diameter in the range between 160 and 200 mm, a volume of Deuterium will be used of approximately 1 to 1.5 sccm, i.e 1 - 1.5 milliliter per minute at normal pressure and ambient temperature. A single bottle of Deuterium will therefore last for many years of Fusor operation experiments.
More details about the setup of the Deuterium gas line can be found on the reactor page.
When obtained as heavy water (image 2), Deuterium gas needs to be produced by electrolysis to make it suitable for feeding into the reactor.
The density of heavy water is 1.107 g per ml and the molar mass of heavy water is 20.0276 g per mole. One mole of Deuterium oxide will yield two Deuterium ions or one Deuterium gas molecule. From a commercial volume of 100 ml of heavy water therefore almost 101 liters of Deuterium gas, at standard pressure and temperature, can be electrolyzed as calculated below.
Mass D2O = 100 ml / 1.107 g/ml = 90.33 gram
Molar mass D2O = 20.0276 g/mole
Moles D2O = 90.33 g / 20.0276 g/mole = 4.51 moles
One mole D2O is equivalent with a volume of 22.4 liter D2O vapour
4.51 Moles equal 4.51 x 22.4 = 101 liters of D2O vapour
D2O gas <=> 1 (D2 gas) + 0.5 (O2 gas)
1 mole gas <=> 1.5 mole gas
mole ratio = 1.5
The volume of the elemental gases is: Volume D2O gas x mole ratio = 101 liter x 1.5 = 151.5 liter or 101 liter D2 gas and 50.5 liter O2 gas are formed by electrolysis.
The average price of 100 ml heavy water 99% is about 120 USD, hence one liter of Deuterium gas (after electrolysis) for amateur fusioneer use costs about 1.18 USD.
Under similar conditions as described in the paragraph about Deuterium gas, a volume of 100 ml of heavy water (when electrolyzed without losses and transferred into the Fusor) will enable continuous Fusor operation during more than 70 days.
Image 2: Bottle with 25 ml Deuterium oxide (© Supplier)
A method to extract Deuterium from heavy water is by means of electrolysis: an electrical current is passed through heavy water by using two inert metal electrodes, preferably made of platinum. Deuterium will appear at the negatively charged electrode, the cathode, and Oxygen will appear at the positively charged electrode, the anode, conform the overall chemical reaction:
2D2O → 2D2 + O2
The theoretical voltage to be applied is for starting electrolysis is 1.48V, but usually higher voltages will be used. As an electrolyte a solution of Sodium hydrogen carbonate (NaHCO3 [baking soda] analytical grade) needs to be added.
Do not add Sodium chloride as this will produce Chlorine gas together with Deuterium gas from your electrolysis cathode.
The highest production of Deuterium gas will be obtained when performing electrolysis in a 0.4M solution of NaHCO3 in heavy water, i.e. at pH 8.2 of the solution (reference 1). A 0.4M solution of NaHCO3 (molar mass is 84.00661 g/mol) will be obtained by dissolving 3.36 gram of NaHCO3 in 100 ml of heavy water.
From a strictly theoretical point of view adding (highly pure) baking soda to heavy water introduces Hydrogen atoms into the solution, which will join Deuterium at the cathode as an impurity in Deuterium gas. To avoid this, we could add NaDCO3, which is not commercially available and needs to be synthetisized by leading CO2 gas through a Sodium carbonate (soda, Na2CO3) solution, consisting of Na2CO3 dissolved in heavy water. A method to do this could be performed by dissolving a molar equivalent amount of Na2CO3 (i.e. equivalent to 0.4M NaHCO3) in a chosen volume of Deuterium oxide (heavy water) and bubbling CO2 gas through the solution (by means of an aquarium air diffusor) untill more or less a pH of 8.2 will be obtained. This is apparently quite a lot of work and it remains the question whether the created H2 impurity in Deuterium gas is of importance in the overall process of fusion in our Fusor.
Similarly to the H2 impurity in Deuterium, CO2 gas will join Oxygen gas as an impurity at the cathode, when we use either NaHCO3 or NaDCO3, but that is of no concern to us as gases at the anode are of no use for the Fusor and are allowed to escape in the air.
A more important disadvantage of the classical electrolysis process is that in the gas line following the electrolysis cell a (rather large capacity) gas dryer needs to be installed because the electrolysis produces moist Deuterium gas (saturated with heavy water). This occurs because the Deuterium gas bubbles up from the cathode through heavy water before it arrives in the output gas line. This is a very efficient way to saturate gas with heavy water.
A description how to construct an electrolysis cell for heavy water can be found in reference 2.
Electrolysis with a polymer electrolyte membrane (PEM), also called proton exchange membrane (PEM), is electrolysis of (heavy) water in a cell with a solid polymer electrolyte that is responsible for conduction of Protons (for heavy water read "Deuterons"), separation of product gases and electrical insulation of the electrodes (reference 3). The cell material is Nafion, a sulfonated tetrafluoroethylene based fluoropolymer-copolymer, with a thickness of 100-200 µm (reference 4).
The PEM cell is a highly efficient way to produce Deuterium gas with a very high purity (99.999%) with no cross over of oxygen. There is some doubt about the statement that the cell produces dry gas because the membrane itself is hydrated and therefore it is believed that the Deuterium gas produced in the cell can be saturated with (heavy) water and in that case it would not harm to add a Drierite column in the Deuterium gas line. It is a fact, however, that heavy water resides at the anode side of the (wet) membrane, where Oxygen is produced. There are therefore no doubts that the Oxygen gas is saturated with heavy water and it might be worthwhile to consider a cold trap in the Oxygen line in order to retain heavy water from the Oxygen gas outlet. A nice calculator to calculate how much heavy water is contained in a volume of saturated Deuterium gas can be found in reference 5.
Advantages of the PEM cell are further that it can operate at elevated temperatures (50-80°C), it has a high current density (0.6-2.0 mA/cm2), a cell operating voltage of 1.75-2.20V and it can produce gas at a high pressure (stack pressure < 30 bar). The latter is not possible with the cell from images 4 and 5 because the housing is made of plastic and will burst when subjected to high pressures.
For the production of Deuterium gas for a Fusor one single cell is usually sufficient because a (Nafion) cell surface of 10 cm2 produces about 2 liter Deuterium per hour at ambient temperature and pressure. The commercial PEM cell as shown in images 4 and 5 has a Deuterium production of 5 ml per minute when operated at 2.0 V DC at 700 mA current. That is a production rate of five times the required Deuterium input for a Fusor! In order to regulate the flow rate from the cell to the intake rate of the Fusor, a (preferably automatic) system needs to be developed for a PEM cell Deuterium line.
Fortunately for the fusioneer, PEM cells are easily available either as a single cell (approx. 40 - 45 EUR) or as part of an educational kit (approx. 55 EUR) as shown in image 3:
Image 3: Solar Hydrogen Educational Kit
The Horizon Solar Hydrogen Educational Kit contains the PEM cell, a solar panel, an electric motor, a propellor, reservoirs for collecting Hydrogen and Oxygen, a battery holder for two 1.5V AA batteries (useless, because the cell eats a lot of power and the batteries will not last long), wires and plugs, tubing, stands for the reservoirs and the cell and a CD-ROM with information and experiments. The dimensions of this PEM cell are 65x65x20 mm and it has a weight of 87 grams (empty). More about this and similar kits can be found in reference 6.
For the Fusor we are only interested in the PEM cell (images 4 and 5) , the wiring and the tubing.
Image 4: PEM Cel Hydrogen Side (© FRS 2014)
Image 4 shows the Hydrogen side of the PEM cell with two in/outlets and a black negative wire connector. For the fusioneers this is the Deuterium gas side. The in/outlets can be seen in image 4 at the 11 o'clock position near the hex screw head right under and at the 5 o'clock position near the hex screw head left top.
Image 5: PEM Cel Oxygen Side (© FRS 2014)
Image 5 shows the Oxygen side of the PEM cell with two white capped in/outlets and the red positive wire connector.
For those who want to construct their own PEM cell, the membrane is available at a price of approx. 16 EUR for a square piece of 10 cm2 (sides of 3.3.cm). The power requirements are 1 A per cm2 at 0.6V. The reason for constructing a PEM cell might be for producing Deuterium at high pressure or at a higher capacity by stacking cells. In that case the housing(s) of the cell(s) must be made of SS (stainless steel or inox or rostfrei) steel.
From the possibilities studied for production of Fusor suitable Deuterium gas we found two options:
When we take a look at the two most common available processes of electrolysis of heavy water we have again two options:
What remains now is the development of a setup of the system of the electrolyser unit based on a PEM cell.
PEM cell electrolyser unit
For the setup of the PEM cell electrolyser unit a choice can be made from different solutions used by fusioneers. The problem with developing an automatically working unit was found in the problem to get heavy water in the cell in a permanent but controlled flow. The PEM cell from images 4 and 5 has been constructed by the manufacturer for working with a single charge of (heavy) water per run. A volume of heavy water is injected with a syringe into the cell through the top inlet on the O2 side and after three minutes soaking the power plug is connected for starting of electrolysis in the cell. Because for our Fusor we permit O2 to escape into the air (through the bottom outlet), which presumably makes the O2 side a pressureless chamber, we may have a possibility one way or another to connect a permanent tube into the cell, permitting to add water. Than we need another tube connected to the outlet (with an y connector to permit O2 to escape)and we might be able to circulate heavy water through the cell and control the flow e.g. by means of a microprocessor.
On Fusornet a simple but efficient method was found for an electrolysis unit. The idea behind the Fusornet device has been changed slightly and is shown in image 6:
Image 6: Layout of PEM cell electrolyser (©FRS 2014)
The PEM cell electrolysis unit in image 7 operates as follows:
The PEM cell is under power and produces Deuterium gas. The D2 gas flow from the PEM cell arrives in the glass cylinder at the left, which is connected to the glass cylinder at the right. Both cylinders are filled to a certain level with (inert) vacuum oil. The production of D2 from the PEM cell is larger than the uptake in the vacuum chamber, which is regulated by a needle valve after drying the gas in the gas dryer. Therefore pressure will build up in the left cylinder and oil will be pressed into the right cylinder. The oil level in the right cylinder rises, air is allowed to escape through the air vent and a floater is pressed up until it activates an opto-electronic switch, which turns off the DC current through the PEM cell.
Gas production stops, but gas consumption by the Fusor slowly continues, the floater sinks and the opto-electronic switch connects the PEM cell to power, etc. When the opto-electronic switch is controlled by a microprocessor, e.g. an Arduino, a delay can be programmed to create a slight damping on the switching events.
As a safety measure an emergency vent tube has been added to the design: When the dryer might clog and interrupts the gas flow, the pressure will rise in the left cylinder and the floater will activate the switch and stop the process. When the opto-electronic switch also fails, pressure build up will continue until the oil level in the left cylinder drops under the level of the emergency vent tube, which is connected to outside air and blows off excess D2 gas produced.
The water supply to the PEM cell is performed by a recirculating heavy water system, driven by a peristaltic pump. The peristaltic pump (image 6) draws heavy water from the storage bottle and presses it into the PEM cell, which at the maximum water level has a return tube connected, which returns the surplus of heavy water to the storage bottle (see image 6).
Image 7: peristaltic pump (© Supplier China)
Contrary to the 12V mark in image 7, the peristaltic pump operates at 24V DC, which is available from the power supply used for the vacuum valves and solenoids in the vacuum lines of the Fusor. The pump draws a current of 80 mA and has a flow rate of 20 - 60 ml/min at a rotation speed of 0.1-100 rpm (motor: 5000 rpm). The tubing size is IDxOD: 2.5x4.7 mm.
As we have calculated in previous paragraphs on this page, 1 ml of heavy water after electrolysis yields approx. 1000 ml of D2 gas, whereas the Fusor uses a gas flow of 1 - 1.5 ml/min. The specifications for the peristaltic pump mention a minimal flow rate of 20 ml /min, which can be considered as far too high for our purposes. It probably would not harm the PEM cell to rush heavy water through it but a somewhat quieter flow seems more appropriate. Moreover, the outlet connector and the return tubing should be of a sufficient capacity to drain all the surplus water out of the cell. If not, the cell will overfill and heavy water will be pushed out and spilled through the O2 power connector when that is not a watertight connection.
Since we have chosen to use e.g. an Arduino for controlling the opto-electronic switch in the regulating reservoir for an automatic production of sufficient D2 gas to operate the Fusor, we can use the microprocessor unit also for controlling the peristaltic pump to deliver a reduced flow of heavy water by means of pulse width modulation (PWM) of the power and to stop the pump when the electrolysis is interrupted by an activated opto-electronic switch.The connection between the pump and the microprocessor is than made over a 5 V relay. For a microprocessor PWM controlled peristaltic pump it would be an advantage when we replace the electric motor from the (cheap) peristaltic pump by a stepper motor, which would enable an even more precisely regulated flow.
The power supply (image 8) is a mains connected stabilized DC unit that delivers 6 A max. at 3.3 V. The dimensions of the power supply are rather small, l x w x h is 80x50x30 mm. Two 2 W resistors, shunted as a divider lower the voltage to 2 V for the PEM cell (and generate some heat).
For the calculation of the values for the resistors in the divider we need to know the ohmic resistance of the PEM cell. This ohmic resistance is build up as a combination of the electrolyte resistance with a polarization resistance and a double layer capacitance as is shown in the Randles equivalent circuit (reference 7 and image 9). Reference 7 indicates that for a specific cell material a resistance was found of ca. 0.12 Ω/cm2.
The Horizon PEM cell of images 4 and 5 have a membrane with dimensions of approx. 2.25x2.25 cm or a surface of approx. 5 cm2, which would be equivalent to a resistance of about 0.6 Ω. for that cell assuming a resistance of 0.12 Ω/cm2. Of course it is not very scientific to assume that the value of 0.12 Ω/cm2 is also valid for our PEM cell because we are not sure that we are dealing with the same membrane material. Earlier we mentioned that the membrane material from the Horizon PEM cell can be bought per surface and that it requires 1A per cm2 at 0.6V, which suggests a resistance of 0.6 Ω/cm2. This is fully in line with the values given for the Horizon PEM cell, which operates at 2.0 V drawing 700 mA, suggesting a calculated resistance of 2.86 Ω for a surface of approx. 5 cm2, yielding a calculated 0.56 Ω/cm2 which is indeed equivalent to our 0.6 Ω/cm2.
Therefore a value of 2.86 Ω resistance for the cell could be used as a starting point for caculating the voltage divider and after some trial and error determinations we can more or less set our PEM cell on the correct voltage of 2.0 V. A voltage and current regulated 0-15 V benchtop laboratory power supply would probably be easier and handy but also more costly.
Image 8: 3.3 V 6 A DC power supply (© Supplier China)
Image 9: Randles equivalent circuit of PEM cell ohmic resistance (© FRS 2014)
Image 10 shows the voltage divider for the PEM cell circuit with the values for the resistors: 1.5 Ω and 12 Ω, each 2 W. This circuit with a PEM cell resistance of 2.86 Ω will deliver 2 V with a current of 700 mA through the cell.
Image 10: Voltage divider for PEM cell (© FRS 2014)
In the 2 V DC power supply line to the PEM cell a combined Volt/Ampère meter controls the voltage over the cell and the current through the cell (image 11).
Image 11: Combined Volt/Ampère meter (© Supplier China)
A last word can be said here about the design for the PEM cell electrolyser unit as shown above: The entire Deuterium gas unit has been designed for continuous production of D2 gas and for continuous operation, which is a situation that in practice hardly occurs. Usually the Fusor will be run for a short period and therefore most probably one single charge of heavy water in the PEM cell will be sufficient for an entire Fusor run. It is therefore not a bad idea for proof of concept Fusor runs to keep the electrolyser unit as simple as possible and to forget about the recirculating pump system until sufficient experience has been gained to judge the necessity of such a refined system. On the other hand, the recirculating pump system also has an advantage: it is not allowed to let the membrane run dry when it is under power, as this will destroy the membrane, and the recirculating pump system prevents running dry of the membrane.
Though the systems for Deuterium production, as described above, are quite efficient for feeding a Deuterium gas line, some problems may occur when the pressure regulator for a lecture bottle is unable to produce a low enough presssure of Deuterium gas or when the production of Deuterium gas from a PEM cell is not sufficient to feed the Fusor.
In such cases it my be required to add a storage bottle in the Deuterium gas line, where a sufficiently large amount at a low pressure of Deuterium gas is stored. The storage bottle (image 12) is placed after the gas dryer and before the needle valve.
Image 12: Swagelok stainless steel gas sample bottle (© Swagelok)
A storage bottle with a capacity of maximally 500 ml is more than adequate to store a few bars of Deuterium gas, which can than be used for endless Fusor runs.
However, for economy reasons we make use of a Whitey DOT-3E1800 304L-HDF4-75CC stainless steel cylinder with a capacity of 75 ml, equipped on both ends with 1/4" female NPT threads (image 13):
Image 13: Whitey DOT-3E1800 gas cylinder (© Whitey)
The low volume of 75 ml prevents spilling too much of the precious Deuterium gas, whereas with a Fusor Deuterium gas flow of 1 SCCM a filling of the cylinder at 2 bar pressure will be sufficient to operate the Fusor during two and a half hours before the gas runs out.
For a Deuterium gas system with a pressurized bottle of Deuterium gas and a pressure regulator lacking the second stage for regulating the gas to a low pressure, the layout of the system is quite easy (image 14):
Image 14: Deuterium gas storage (© FRS 2015)
The circuit in image 14 is operated as follows:
The needle valve and the inlety valve are closed. The flush valve is opened , which is connected to the vacuum system. The storage bottle is evacuated now to the best vacuum possible. After evacuation the flush valve is closed and the inlet valve is slightly opened in order to fill the storage bottle with a volume of Deuterium gas. The inlet valve is closed after partially filling the storage bottle an the bottle is flushed by evacuation to vacuum through the flush valve. This operation is repeated up to five times in order to obtain a filling of the storage bottle with deuterium gas without impurities from remaining air in the bottle.
With the inlet valve and the flush valve closed the Fusor can now be operated through the needle valve with Deuterium gas present in the storage bottle.
Image 14: Swagelok pressure gauges and components for the alternative circuit (© supplier Germany)
For a Deuterium production system with a PEM cell the same storage bottle circuit is possible but some extra precautions need to be taken. When opening the inlet valve for filling the storage bottle with Deuterium gas, the evacuated storage bottle could suck oil from the gas collecting glass cylinders into the gas dryer and/or cause damage to the PEM cell by applying a vacuum to it (see image 6). Moreover, the gas produced by the PEM cell is almost pressureless and should be compressed when filling the storage bottle.
The inlet valve therefore should be a needle valve in order to slow down the filling speed of the storage bottle under vacuum and additionally a pneumatically operated, level sensor controlled shut off valve should be added to the circuit (image 16).
Image 16: Deuterium gas storage from PEM cell (© FRS 2015)
The pneumatic valves as used are Swagelok bellow valves (image 17), which are controlled through a bidirectional pneumatic controller activated by a level sensor on the floater in the glass gas cylinder of image 6. This sensor should react on a minimum level of the oil in the right cylinder and then close the top pneumatic valve in image 16. When the pressure in the storage bottle is at ambient pressure, the top pneumatic valve remains closed and the bottom pneumatic valve opens, while the membrane pump starts pumping and builds up a few bars pressure pressure in the storage bottle.
Image 17: NuPro/Swagelok pneumatic bellow valves (© supplier GB)
For filling the storage cylinder with deuterium gas from a PEM cell we will need a mini-compressor which sufficient pumping capacity to obtain a final filling pressure in the storage cylinder of about 3.5 bar (50 psi). For compressing a clean gas without oil contamination we recommend a membrane operated mini-compressor. A 12VDC electric tyre inflator would be suitable because they can compress air up to a pressure of 7 bar (100 psi) but these usually operate by sucking air through large inlet ports, whereas we need an inlet port as well as an outlet port with a pipe fitting for connecting the compressor into the Deuterium gas circuit.
A suitable mini-compressor is shown in image 18.
Image 18: Mini-compressor (Source: Supplier)
Image18 shows a Topsflo Micro Diaphragm Liquid Pump, Model TS512/W12-1210, also capable of dry running and compressing gas (air) to a constant pressure of 1.0 bar, though for short duration up to 4.0 bar. The power requirements are 12 VDC at 0.8 A. At a constant rpm of 3000 it has a pumping speed of 1200 ml/min. The housing is made of polycarbonate (heat resistant up to 65°C) and the membrane is made of EPDM. The weight is 280 g, the inlet and outlet tube diameter is 6 mm (i.e. suitable for ID 6 mm tubing) and the dimensions in mm are shown in image 19.
Image 19: Mini-compressor; dimensions in mm (Source: Supplier)
A final word however, is to express the sincere opinion that the alternatives in this chapter are best to be avoided. The circuitry is more complicated, more costly items are added, more changes to fail are created and the requirement to flush the storage bottle several times waists a lot of precious Deuterium gas. Only when it is fun for you to create complex systems and when you possess a large number of Swagelok, Nupro, Parker Hannifin and Whitey components, like me, just proceed and create your preferred system!
Ref. 1: NaHCO3 in electrolysis of water: http://www.google.nl/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=0CCEQFjAA&url=http%3A%2F%2Fwww.scirp.org%2Fjournal%2FPaperDownload.aspx%3FpaperID%3D29168&ei=G-0hVJPOFY3isASj5IHgBA&usg=AFQjCNGj2UBZTDqZ2tr8ejzgchy-RBPY2g
Ref. 2: Electrolysis unit for heavy water: http://www.rtftechnologies.org/physics/deuterium-electrolysis.htm
Ref. 3: PEM electrolysis: http://en.wikipedia.org/wiki/Polymer_electrolyte_membrane_electrolysis
Ref. 4: Nafion properties: http://www.ias.ac.in/matersci/bmsjun2009/285.pdf
Ref. 5: Gas Saturation Calculator: http://www.hbuehrer.ch/Rechner/GasSatur.html
Ref. 6: Educational PEM cell kits: http://www.horizonfuelcell.com/#!science-experiment-kits/c1z3u
Ref. 7: PEM cell ohmic resistance: http://www.fuelcellmarkets.com/content/images/articles/Cooper-Smith-on-FC-ohmic-resistance-techniques-JPS-2006-final.pdf
|Last Updated on: Fri Nov 18 10:45:43 2016