NL1037618C2 - System for storing solid chemical hydrides dispersed in a liquid (oil) and a method for releasing hydrogen stored in the chemical hydrides. - Google Patents

Description

System for storing solid chemical hydrides dispersed in a liquid (oil) and 5 a method for releasing hydrogen stored in the chemical hydrides

The present invention refers to a system for storing solid chemical hydrides dispersed in a liquid (oil) and a method for releasing hydrogen stored in the chemical hydrides.

Hydrogen may be stored as a compressed gas, as a cryogenic liquid, in carbon, in solid state 10 materials or in chemical hydrides. High pressure storage requires special storage tanks that are able to withstand very high pressures. All storage materials tend to become brittle and porous to hydrogen, particularly at high hydrogen pressures. Cryogenic storage requires cooling the hydrogen down to -253°C. The maximum pressure that may be accomplished in a solid state storage tank is 1400 bar, since hydrogen becomes a liquid at higher pressures.

15

Storing hydrogen in carbon creates carbon - hydrogen bonds that may be burnt to supply energy, in which case also the polluting gas CO2 is formed. Reversible storage in carbon including high pressure storage and hydrogen release by heat is also possible. All these solutions have a very limited hydrogen storage capacity in weight and volume terms and all 20 these systems require dedicated storage solutions and dedicated control and release mechanisms, including energy to release hydrogen.

A key point in the development of an inhaler was finding a solution for storing the hydrogen required for driving the heating system for evaporating the medication. This development is 25 also used in the setup of a nicotine replacement inhaler and therefore the target size is that of a cigarette having the energy content of a packet of cigarettes, such that the user does not have to replace both the nicotine dose and the energy at the same time. This reduces the handling and the cost.

30 These prerequisites have had a major influence on the development of the inhaler, since they also had a bearing on the toxicity, weight, volume, safety, connection of different parts, the way hydrogen is made available but most of all the control rate of the process itself. The 1 03 76 1 8 -2- process, being the evaporation of a defined amount of liquid during the first 30 % of the inhalation in order to enable a deposition in the alveoli, demands that the application of hydrogen in that process proceeds swiftly and adequately. This actually calls for the application of hydrogen as a gas, while the dimensional requirements demand a much higher energy 5 density. The weight requirements exclude the use of heavy elements for storage under pressure and for reducing the pressure. Thus a small window of possibilities exists.

Chemical hydrides fit in this small window. The window is further narrowed in that only those hydrides can be used which do not require the supply of energy for releasing hydro-10 gen. The further demands regarding toxicity have even further narrowed the window and have resulted in the selection of sodium borohydride.

The major disadvantage of hydrides is that the reactions cannot be controlled to the point of the tiny amounts required by the application. Therefore a system for mixing such tiny 15 amounts had to be developed next to a more general understanding of the chemical reac tions taking place during the process of releasing from the hydride storage.

Hydrides may be stored in solid or liquid form. Storage in liquid form has quickly been acknowledged as the preferred form for the following reasons: 20 - Handling.

Pumping through lines is easy and largely comparable with present day refueling practices. Use is possible up to old age and including physical handicaps.

- Transport.

Pipeline transport and tanker transport of liquids is possible.

25 - Control

Liquids are more easily dosed than solids. The reaction can be influenced by the way of mixing and the complete reaction is better controlled.

A binary system employs a substance which reacts with another substance. In this specific 30 case the one substance is a liquid containing chemically stored hydrogen, also referred to as -3- the FUEL The other substance triggers the release process, also referred to as the ACTIVATOR.

The following conditions may be stated to influence the overall process and the reaction 5 rate: - Way of mixing.

Liquids are brought into a mutual intensive contact by mixing relatively small amounts. The reaction causes eddy currents which increase the contact area. Chemical composition.

10 Liquids may be water-based or oil-based. Water-based liquids contain a stabilizer which is subject to the following questions: o What is de minimum stabilizer concentration?

Investigations have demonstrated that high stabilizer concentrations require the activator to contain large amounts of additives which consume 15 stabilizer. This reduces the active fractions of both fuel and activator and thus the overall yield, o Which stabilizer is used?

In a sodium borohydride liquid, sodium hydroxide is used as a stabilizer. Initially hydrochloric acid was used to neutralize the hydroxide. For toxicity 20 reasons this has been replaced by acetic acid, o Which water quality is required?

Changing from tap water to pure water as activator had a major impact. Residues in the water slow down the reaction and result in an extended post reaction. The change to pure water implied that the synchronous mixing of 25 small volumes of fuel and activator results in an instant start and stop reac tion.

- Physical form and size.

Hydrides may be available as a liquid, such as aluminium borohydride, or as a water-soluble powder, such as sodium borohydride, or as an insoluble solid, such as most 30 other hydrides. Since hydrides normally react when in contact with water, and some even react with water vapour in the air, precautions are necessary to prevent spon- -4- taneous ignition. An effective precaution is the addition of a mineral oil, typically 30 %, to produce a liquid dispersion, which is pumpable and miscible.

The problem of solid hydrides is that reactions taking place at the surface may result in shielding of the interior hydride by reaction products, thus preventing the further 5 release of hydrogen. Several options are available to minimize the consequences thereof: o Reducing the surface to volume ratio.

By using a granulate having a maximum particle size of e.g. 1 pm, the surface to volume ratio is restricted. By dispersing the granulate in an oil a 10 viscous mass is obtained, which are easy to pump and handle.

o Degrading the solid.

By adding an acid to the activator, the crystal lattice of the hydride is degraded. As a result the granulate dissolves in the mixture and reacts completely with the activator, releasing all stored hydrogen and preventing any 15 residual stored hydrogen. As a disadvantage, this method may require addi tional heat to induce the effective degradation. A balanced approach is required.

o Combination of both options.

By adding acid to a fine granulate, while increasing the temperature, the op-20 timum result may be accomplished with a moderate addition of heat. The oil can be removed from the granulate with a jet wash, which simultaneously activates the hydride surface. During this activation part of the water is converted to hydrogen according to the known reaction. As a result the residual mass becomes dryer and thus more viscous.

25 By recirculating the residual mass, this is re-entered into the process, mixed with activator including acid and flushed, resulting in a further release of hydrogen. By sensing the required pumping power during recirculation and the pump revolutions, a measure of the viscosity can be given, which guarantees the completion of the process.

30 - Temperature range.

-5-

By mixing the hydride with an oil, a wide temperature range is accomplished. The activator may be protected from low temperatures by adding an antifreeze agent. The effects of these additions have been considered.

- Toxicity and availability.

5 The investigated hydrides have also been selected on availability and conflict of in terests. A high demand for lithium is for example expected to power electric vehicles. If this raw material would also be a carrier for fuel, the price will certainly go up. Beryllium, the prime candidate for storing hydrogen is toxic upon inhalation, which is the reason for not selecting this substance and selecting the number 3 from the list: 10 aluminium, with magnesium and calcium as alternatives.

In the presented setup the various systems have minor mutual differences. The selected hydride is a mere application detail. As a result a hydrogen gas generator is provided which complies with all the mentioned performance requirements of the DOE and the STORHY for 15 the year 2015. Such a hydrogen generator may serve a Fuel Cell powered vehicle as well as an Internal Combustion Engine powered vehicle both in a qualitative and quantitative way. Furthermore on this basis, a platform for refueling stations and a quality control system may be developed.

20 There is a need for a general purpose storage and release system for all chemical hydrides.

Simply said a chemical hydride is a hydrogen bound to a metal, as a result of which hydrogen is stored in the metal as a solid. Suitable metals include alkali, earth alkali and regular metals. Hydrogen atoms may also be bound to any combination of these metals (mixed hy-25 drides).

In principle all metals that are suitable to be used as hydride are known. Unknown, at least to us at this moment, are the binding capacities (amount of hydrogen that can be bound) and the binding strength (energy required to store and release hydrogen) of mixed hydrides. 30 A weak bond requires a relatively small amount of energy to couple hydrogen to the used metal from, a strong bond requires a relatively large amount of energy. This means that the -6- efficiency of chemical hydrides in storing, releasing and regenerating hydrogen is very closely related to the progress in hydride research. The best applicable hydride will be a winner.

There is no general overview of all combinations of metals that may be used as hydride in-5 eluding their binding properties. It is clear that metals are gaining interest as storage media for hydrogen. In 2008 the United Sates Department Of Energy (DOE) has indicated that more research is needed in this area, which in the meantime probably is taking place.

In order to release hydrogen from a chemical hydride an activator is required which, de-10 pending on the chemical hydride, may be moist ambient air or plain water. Water is a well known activator. In practice, upon mixing a chemical hydride with water, the hydrogen bound to the metal is exchanged for oxygen from the water and combines with a hydrogen from the water, leaving hydroxide bound to the metal. This is due to the different polarity between hydride hydrogen and water hydrogen.

15 A stabilizer such as strong alkali may be added to prevent chemical hydrides from reacting with water. In that case the chemical hydride may be dissolved in water. By adding a destabilizing agent, such as an acid (hydrochloric, acetic, citric etc.) the chemical hydride may react with the water. The spent fuel product from the reaction of the chemical hydride and water 20 in that case also contains the corresponding salts of the acid used and will be more complex to regenerate. The cost of regeneration may be twice those of regenerating a chemical hydride and water system.

In addition to water as activator, heat may be added as an accelerator, which may be pro-25 vided as steam (water and heat). Heat in this case provides the energy to release hydrogen from hydrides that do not react spontaneously. The principle of exchanging hydrogen may still apply, however the bond between the metal and the hydrogen may be so strong that energy is required to release the metal bound hydrogen. An onboard water reservoir required for the production of steam is approximately equal in volume to a reservoir for sup-30 plying water as activator. Therefore the reason for selecting a heat accelerated hydride may be safety rather than storage volume. The disadvantages of acceleration by heat include the -7- permanent heat requirement, the slow start and stop and the distinct operating temperature range. Heat acceleration of hydrides may work for industrial steady state applications, where the mentioned disadvantages do not play a role.

5 The key question is how to control the spontaneous reaction of a chemical hydride such that the amount of energy used for storing, releasing and regenerating hydrogen is minimized (most efficient process) and the safety is not jeopardized?

Upon thoroughly mixing a binary system of a - preferably liquid - chemical hydride and a 10 liquid activator, the spontaneous release of hydrogen may be controlled.

Aluminium borohydride is the preferred chemical hydride since it is a liquid; it has a high hydrogen storage capacity due to the 3 positions for bonding, each filled by boron having 4 positions for bonding hydrogen, making a total of 12 positions for bonding hydrogen; and it 15 is relatively cheap due to the abundance of both alumina and borax. The structure of alumi nium hydride prevents a proper orientation in a crystal lattice, which is the reason for being liquid.

Except for aluminium borohydride, most chemical hydrides however are solids, meaning the 20 bonded (gaseous) hydrogen atoms are stored as solids in a crystal lattice. And although solid hydrides (granulate) have the highest hydrogen density, they are difficult to control and handle.

Solid hydrides may be dissolved in water to obtain a liquid hydride. A stabilizer, such as an 25 alkali metal hydroxide is in that case required to prevent the dissolved hydride to react with the water and release hydrogen. To activate the stabilized hydride solution during the mixing of hydride and activator, a destabilizing agent must be added which neutralizes the stabilizer.

-8-

The spent fuel obtained by mixing dissolved or liquid chemical hydrides with an activator, may comprise salts with a limited solubility in the activator, which may deposit. To prevent solid deposits recirculation of the spent fuel is required.

5 By dispersing a solid hydride in an inert carrier fluid, such as an oil, a liquid hydride may be obtained. Oil does not react with the metal(hydride) or with hydrogen, but shields the metal hydrides from reacting spontaneously with moisture and/or oxygen in the air. Most hydride granulates typically have a bulk density of 450 - 600 kg/m3. Depending on the amount of oil added, the density of the resulting hydride in oil dispersion may be similar to the bulk densi-10 ty of the granulate due to occupation of the volumetric voids in the granulate (inefficient stacking of spheres) by oil.

The preferred maximum oil content of a hydride dispersion is approximately 30%. This allows a relatively high hydrogen storage density while preventing exposure of hydride par-15 tides to ambient air. To prevent hydride in oil dispersions from segregating, storage tanks are preferably provided with an impeller or a recirculation system. Alternative inert carrier fluids include natural oils and nonionics such as copolymers of ethylene oxide and propylene oxide. All carrier fluids may be regenerated for reuse.

20 Even if aluminium borohydride is a liquid, it may be mixed with oil to prevent evaporation at ambient temperature and reaction of the vapour with moisture in the air, releasing hydrogen.

The mixing of fuel and activator preferably creates a large contact area between fuel and 25 activator resulting in a fast release of hydrogen. In the case of hydride particles dispersed in oil, this means that the hydride particles preferably are as small as possible in order to prevent the formation of "oxide skins" at the surface of hydride particles and the inclusion of stored hydrogen.

30 The water used as activator preferably has a very high purity. Any contaminations in the water will decrease the reaction rate between water and hydride and will reduce the yield. The -9- contaminations may furthermore accumulate in the spent fuel and may complicate the process of regeneration. The water quality may be controlled by measuring the conductivity of the water at the refueling station. Preferably the maximum conductivity of the water is 0.5 microSiemens.

5

Apart from water, the activator may comprise substances that facilitate the release of hydrogen from the chemical hydride, such as an acid that may dissolve dispersed hydride particles. The dissolution process may be further facilitated by a suitable reactor design and/or by heating the reactor.

10

By adding more than the equivalent amount of water, the retained mixture may still be pumped and the conversion of hydrides may essentially be completed. This process may be controlled by checking the viscosity of the mixture. The viscosity of the hydride in oil dispersion preferably is 1 - 50 times the viscosity of water, more preferably 1-25 times, even 15 more preferably 1-10 times and most preferred 1-5 times the viscosity of water. The vis cosity of the mixture preferably is controlled within the same range.

Apart from the hydrogen released upon mixing a chemical hydride and activator, a mixture of spent fuel and activator or spent fuel is retained. The weight and volume of the spent fuel 20 implies that the amount of energy stored in chemical hydrides will always be lower than the amount of energy contained in a similar volume of gasoline. Any incomplete conversion of chemical hydride to hydrogen further reduces the energy efficiency. For that reason the conversion of chemical hydrides to hydrogen is preferably maximized. This is accomplished by using a liquid hydride.

25

Alternative metals for chemical hydrides having a high hydrogen storage capacity include beryllium (carcinogenic upon inhalation), lithium (large consumption by the battery industry may lead to price pressure), calcium and magnesium. All of these metals form solid hydrides. New combinations of metals and/or boron may have favourable properties for hydrogen 30 storage.

-10-

The quality of a chemical hydride may be determined anywhere in the process chain at any time by mixing a metered amount of a chemical hydride with a metered amount of activator in a certain period of time and determining the amount of hydrogen released within a predetermined time. In order to determine the amount of hydrogen released, the hydrogen 5 may be collected in a closed vessel and the pressure of the hydrogen in the vessel may be measured. Provided the resulting pressure remains within a predetermined range, the quality of the hydride and activator mixture is acceptable. By combining this measurement with the conductivity measurements of the water, the quality of the hydride itself may be determined. Storing the measurements provides a control log.

10

The quality control method may include the addition of water, acid and/or heat in any combination. Each parameter has an optimum window for completing the reaction, using one and the same algorithm. Each hydride and activator has a certain reference value that will be entered and used in any calculation. The result is a certain purity. Possible sources of error 15 include higher oil content of the hydride and lower regeneration percentage. In both cases the result is a lower yield than expected. This may be used as customer service information, where the customer pays for what het gets. This goes beyond the calibrated volume information provided by gas stations to the customer today. Refueling compressed hydrogen or liquid hydrogen is very similar to refueling liquefied propane gas. The mileage per volume of 20 hydride will vary with the ambient temperature similar to gasoline.

Hydrogen may be stored in chemical hydrides by adding compressed hydrogen to a molten metal or to a finely divided dispersion of metal particles in a carrier fluid such as oil. The hydrides form spontaneously and may be collected as a solid or in the case of aluminium boro-25 hydride as a volatile liquid. Heat will open up the crystal lattice of the metal (and upon melting destroy the lattice), allowing the compressed hydrogen to be "squeezed" into the lattice and bound to the metal. Upon cooling (under pressure) the hydrogen atoms are "captured" in the lattice.

30 The addition of acids during the mixing of hydride and activator, requires an additional regeneration step. Based on an analysis of the energy steps for generating hydrogen and regene- -11 - rating spent fuel the process of regenerating chemical hydrides is feasible, and the major expenses are those for energy.

Hydrogen may be compressed in a cost-efficient way using ultraviolet light by changing the 5 energy state of hydrogen and thus the bonding properties. This phenomenon may have a positive influence on the regeneration process of hydrides and thus increase the efficiency and reduce the cost by 25% to 33%. Since LED's may generate UV light with a very high efficiency, this may be the key to very efficient regeneration processes.

10 The preferred hydride selected from the following list of hydrides is aluminium borohydride. The reason for selecting an existing hydride rather than developing a new one is the time and money required for such development.

Hydrogen energy

Molar mass Hydrogen Water Water Binary system density

Substance Name [g/mole] [mass %] [mole/mole] [kg/kg] H2 [mass X] [MJ/kg] [kWh/kg]

LiH Lithium hydride 7.95 12.7% 1 2.27 7.8% 10.0 2.8

LiAIH4 Lithium aluminium hydride 37.95 10.6% 4 1.90 7.3% 9.4 2.6

LiBH4 Lithium borohydride 21.78 18.5% 4 3.31 8.6% 11.0 3.1

NaH Sodium hydride 24.00 4.2% 1 0.75 4.8% 6.1 1.7

NaAlhU Sodium aluminium hydride 54.00 7.5% 4 1.33 6.4% 8.2 2.3

NaBH, Sodium borohydride 37.83 10.7% 4 1.90 7.4% 9.4 2.6

MgH2 Magnesium hydride 26.32 7.7% 2 1.37 6.5% 8.3 2.3

Mg(BH4)2 Magnesium borohydride 53.99 15.0% 8 2.67 8.2% 10.4 2.9

CaH2 Calcium hydride 42.10 4.8% 2 0.86 5.2% 6.6 1.8

Ca(BH4)2 Calcium borohydride 69.77 11.6% 8 2.07 7.6% 9.7 2.7 AIH3 Aluminium hydride 30.01 10.1% 3 1.80 7.2% 9.2 2.6 AI(BH4)3 Aluminium borohydride 71.51 16.9% 12 3.02 8.4% 10.8 3.0 15 -12-

The selection criteria include: • Spontaneous release of hydrogen • Hydrogen energy density • Handling & availability 5 • Environmental risk profile • Toxicity profile

The compound has been selected according to the red arrows in the following decision tree: The reason for not selecting lithium borohydride is the high demand for lithium from the battery manufacture and its possible use as a raw material for fusion reactors, such as the 10 pilot reactor in Cadarache, France.

Sodium, calcium and magnesium are readily available minerals, but the borohydrides have a lower hydrogen energy content and invariable are solids. Solid borohydrides must be dispersed in mineral oil in order to facilitate handling and dispensing. In that case the particles 15 must be <lpm in order that all metal may be occupied with borohydride and all stored hydrogen may be released. Any oxide skins formed at the surface of the solid borohydrides may be dissolved by adding an acid. Obviously this will reduce the energy density of the system.

20 Aluminium is readily available as bauxite mineral and the borohydride is a liquid having a high hydrogen energy content, making it easy to handle and dispense.

Boron is readily available as borax.

25 Hydrogen is readily available from either methane and water.

The structure of aluminium borohydride (AI(BH4)3) comprises a central aluminium atom in a regular six surrounding with three borohydride groups, whereby three hydrogen atoms share a bond both with the central atom and a boron atom. The graphical representation is 30 as follows: -13-

The H2FUEL comprises a binary system of 70% aluminium borohydride in mineral oil (Fuel) and a water and glycol mixture (Activator). Fuel and Activator are mixed in the desired proportions under the appropriate pressure in a specially designed mixer in order to secure the 5 complete release of all stored hydrogen. Provided the mixer functions appropriately, 100% of the stored hydrogen will be released.

Fuel properties

Appearance liquid 10 Relative density (water = 1) 0.605

Temperature range -40.. 65°C

Miscibility with oil in all proportions

Miscibility with water reacts 15 The mineral oil in the Fuel shields the aluminium borohydride from moisture and decreases the vapour pressure of the aluminium borohydride. Provided the oil is inert (cannot absorb or react with hydrogen), it has no effect whatsoever on the release of hydrogen.

Activator properties 20 Appearance liquid

Relative density (water = 1) 1.056

Temperature range -40.. 65°C

Miscibility with water in all proportions 25 The glycol in the activator prevents the activator from freezing.

H2FUEL properties

Fueling rate 37 L/min

Hydrogen production 56 g/l = 63.6 g/kg 30 Hydrogen release rate depends on H2FUEL pump rate

Induction time for release < 5 ms - 14-

Spentfuel properties

Appearance emulsion/dispersion

Relative density (water = 1) 0.8 5 The spent fuel contains mineral oil, aluminium hydroxide, boric acid and glycol borates. Spent fuel may be stored for extended periods of time. Mineral oil and/or solids may separate upon storage. The spent fuel will be collected for recycle. Recycling includes the following steps: • Separation of solid aluminium hydroxide 10 • Drying of aluminium hydroxide to alumina (raw material for aluminium) • Separation of mineral oil for reuse • Conversion of residual borate mixture with methanol (raw material for boro-hydride), producing glycol for reuse 15 Spent fuel may be recycled without any limitations over and over again. Recycle losses are expected to be less than 0.01%.

A chemical partner will have to be selected for producing and/or recycling the H2FUEL. Currently aluminium borohydride may be produced on lab scale. Small scale and full scale pro-20 duction will have to be set up. Aluminium industrie Delfzijl (Corus) is a possible candidate for aluminium production and recycle.

(Boro)hydrides are and have been used as rocket fuel and aviation fuel.

25 Theoretically aluminium borohydride may be produced according to the following pathway: • Steam reformation of natural gas to hydrogen and carbon dioxide • Electrochemical conversion of alumina to aluminium metal • Addition of hydrogen to aluminium metal • Conversion of boric acid with methanol to trimethyl borate ester 30 • Hydride formation from borate ester and aluminium hydride -15-

Estimated price for full scale production of H2FUEL per ton hydrogen is USD 15,907 Component Amount Price/t Value

Alumina 8430 kg USD 300 USD 2,529

Borax 7668 kg USD 900 USD 6,901 5 Methanol 11921kg USD 200 USD 2,384

Glycol 6060 kg USD 550 USD 3,333

Mineral oil 1267 kg USD 600 USD 760

Estimated price for full scale recycling of H2FUEL per ton hydrogen ranges from USD 1,540 for grey hydrogen to USD 2,760 for grey-green hydrogen.

10

After manufacture, the H2FUEL is stored under a nitrogen blanket in a closed container.

Since aluminium borohydride is miscible in all proportions with mineral oil, the fuel does not separate upon storage. The activator is miscible in all proportions with water and does not separate upon storage either. Provided the H2FUEL is stored as indicated, the system is sta-15 ble for prolonged periods of time and the risks of unintended release of hydrogen due to moisture or high temperatures are negligible. Storage in this way is also applicable during development and does not affect the intentional release of hydrogen during use.

The H2FUEL may be transported in compartmented tankers by rail, road or vessel. Recycled 20 glycol may be simultaneously transported in a separate compartment for mixing with purified water on site. After unloading H2FUEL and glycol at a refueling station, spent fuel may be loaded either in a separate compartment or in the H2FUEL compartment after flushing.

The composition of the fuel allows temperatures up to 65°C. Higher temperatures up to 80°C 25 may be accommodated by increasing the pressure of the nitrogen blanket up to 3 - 5 bar max. Obviously the storage container must be able to withstand such pressures. At higher temperatures the release of hydrogen is accelerated.

Apart from investment in new dispensers and nitrogen blanketing, the cost of full scale sto-30 rage will be comparable with the cost of storing gasoline, with a premium of e.g. 10%.

-16-

The advantages of the H2FUEL include: • Ease of handling liquids • High transport and pumping efficiency during storage • Minimum transport and storage expenses due to high hydrogen content 5 • Low maintenance due to low pressure (no hydrogen embrittlement)

Storage of pure undiluted hydrides caries a high risk of unintended hydrogen release due to moisture or high temperatures.

10 In case of leakage of H2FUEL hydrogen may be released upon contact with water and hazardous fumes may be formed. Breath protection may be required. Water must not be used. Avoid contact since the H2FUEL is corrosive to skin, eyes and mucous membranes. Ventilate well and shut down any possible ignition sources. Prevent the H2FUEL from spreading over a large area and clean up spilled material according to the safety manual.

15

Leakage of activator does not pose any specific health hazards. Prevent leaked activator from spreading over a large area and clean up spilled material according to the safety manual.

20 In the unlikely event that fuel and activator come into contact unintentionally, hydrogen is released. Ventilate well and shut down any possible ignition sources.

The following risk and safety statements apply to both storage and transport:

Hazard Codes FJ

Risk Statements 15-24/25-34

Safety Statements 22-26-36/37/39-43-45 RIDADR UN1426 4.3/PG 1 WGK Germany 2 RTECS ED3325000 LEL 3.02%(V) - 17-

The safety measures at current refueling stations for gasoline are expected to be sufficient to handle H2FUEL

The main health hazard of H2FUEL appears to be boric acid. Based on mammal median lethal 5 dose (LD50) rating of 2,660 mg/kg body mass, boric acid is poisonous if taken internally or inhaled. However, it is generally considered to be not much more toxic than table salt The Thirteenth Edition of the Merck Index indicates that the LD50 of boric acid is 5.14 g/kg for oral dosages given to rats, and that 5 to 20 g/kg has produced death in adult humans. The LD50 of sodium chloride is reported to be 3.75 g/kg in rats according to the Merck Index.

10 Long term exposure to boric acid may be of more concern. Although it does not appear to be carcinogenic, studies in dogs have attributed testicular atrophy after exposure to 32 mg/kg bw/day for 90 days. This level is far lower than the LD50.

H2FUEL quality requirements include: 15 • Exclusion of moisture (fuel) • Hydrogen content (fuel) • Bacterial count (activator)

Hydrogen content may be controlled by determining the hydrogen release of a standard 20 amount of fuel with a standard excess of activator. Activator quality may be determined with a conductivity meter. The hydrogen yield may be monitored by measuring the pressure increase in the reactor from a known amount of H2FUEL.

Although the preferred activator of the H2FUEL is pure water, due to its high extractable 25 proton content and high reactivity, the temperature range of pure water is limited. For that reason glycol is added, extending the temperature range down to -40°C. Glycol also has extractable protons and therefore is a suitable activator as well.

Further alternative activators include: ammonia and alcohols such as methanol and ethanol. 30 Ammonia is a toxic gas. The alcohols have a lower boiling point than glycol.

-18-

Pure water may be produced on site by electro-deionization or by filtering. Usually a deionization system has twin columns in alternate operation, which may back up each other. As an alternative a Pall filter may be used as a back up.

5

The activator may contain acids to facilitate the dissolution of insoluble materials e.g. metal oxides. Heating the activator will accelerate the release of hydrogen. The activator may be re-circulated over a filter to prevent bacterial growth.

10 The heat released when combining Fuel and Activator, may be recovered through a heat exchanger and/or a heat pump. This heat does not in any way affect the hydrogen release process. By using a Seebeck element, the recovered heat may be converted to electricity that may be used to power e.g. an electric engine (most efficient) or to recharge a battery. This recovery increases the energy efficiency of the H2FUEL.

15

For a Seebeck element the investment is currently € 800 per kW. Depreciation in 5 years and 5% interest per year yields annual costs of 800/5 + 400*5% = € 180. Assuming 18000 km per annum, the cost per km is € 0.01 20 -19-

The type of hydrogen storage has major implications for the transport expenses: CGH2

System LH2 H2FUEL* (350 bar)

Tanker loading kg L kg L kg L

Gross 2267 34000 425.0 17000 27200 34000 loading

Hydrogen 2267 34000 425 17000 2720 loading

Losses 5% 113

Net delivery 2153 425 2720

Transport price

Basis €/ km €0.85 €0.85 €1.00

Per 1000 kg €/ km €0.38 €2.00 €0.37 hydrogen Per 1000 kg € / km €0.39 €2.00 €0.37 H2 net

Maximum driving range

km 158 I I

* assuming 40% H2FUEL and 60% glycol in separate compartments 5 The energy balance for H2FUEL has 4 values depending on the fuel type:

Grey Improved grey Grey-green Green-grey 46.9% 57.0% 54.0% 61.2%

The energy input in MJ/kg hydrogen for H2FUEL has 4 values depending on the fuel type:

Grey Improved grey Grey-green Green-grey 633.8 521.2 549.6 484.9 -20-

Efficiency losses occur in power generation (15%); steam methane reforming (17%); aluminium electrolysis (35 - 45%)

With a Seebeck element operating at 65% efficiency, 28,5 MJ electrical energy may be re-5 covered per kg hydrogen.

Compressed hydrogen (350 bar) energy balance has 2 values based on hydrogen source: Steam methane reform: 68.2%

Electrolytic hydrogen: 49.8% 10

The invention will be explained in more detail with reference to figures, wherein Fig.1 shows a first embodiment of a system for storing chemical hydrides and releasing hydrogen stored in the chemical hydrides;

Fig. 2 shows a second embodiment of a system for storing chemical hydrides and 15 releasing hydrogen stored in the chemical hydrides;

Fig. 3 shows a preferred embodiment of an inline mixer for mixing fuel and activator.

Corresponding numbers refer to corresponding parts of the system.

20 Referring to figures 1 and 2, the system comprises a fuel storage chamber (100), an activator1 storage chamber (200) and a spent fuel storage chamber (300). Each of the storage chambers (100, 200, 300) is provided with a sensor (21, 22, 23) respectively for sensing the fluid level, preferably a Hall sensor2 or an optical displacement measuring system based on the displacement of the moving walls.

25

Each of the storage chambers (100,200,300) preferably has a flexible volume3, and preferably is arranged such that a volume increase in one of the chambers is (partly) accompanied

Activator may be water, acid and/or a provision to maintain a certain temperature.or a combination thereof.

2 A Hall sensor is magnetic and therefore spark free, which is important for the fuel and the spent fuel in case of contact with oxygen.

3

Flexible storage chambers are only required if volume is an issue.

-21 - by a (simultaneous) volume decrease of the other storage chambers.1 The flexible storage chambers may be placed in a rigid 'tank' housing having a fixed volume.2

The storage chambers are preferably provided with a connector (114) for providing a blan-5 ket gas such as nitrogen (114-4) from an external supply source to the storage chambers. All the different sub-connectors in connector (114): fuel (1): activator (2): spent fuel (3) and nitrogen (4) preferably have unique couplings that prevent making any undesired connection. The sub-connectors preferably are completely free of any spills. A preferred type of connector includes the quick connect series of Swagelock.

10 Each of the fuel storage chamber (100) and activator storage chamber (200) is provided with a line (110, 210) respectively for supplying fuel and activator and/or a rinsing fluid from an external source of supply and each of the supply lines (110,210) is provided with a valve system (111, 211) respectively and a connector for connecting to an external supply source. The spent fuel storage chamber (300) is provided with a line (3310) for discharging spent fuel 15 to an external collection system and the spent fuel discharge line is provided with a valve system (331) and a connector for connecting to an external spent fuel collection system.

The nitrogen inlet is providet with valve (411).

Each of the valve systems (111, 211,331,411) and connectors may be integrated, however 20 the connectors (or the integrated valve system and connectors are different. The connectors and/or the integrated valve system and connectors may be combined in a single connector3 (114), which may be operated as one connector for all connections.

The storage chambers for fuel (100) and activator (200) may be provided with a recirculation 25 line (120,220) provided with a pump (125, 225) or an impeller (106,206) for homogenisation and/or pumping purposes. A screen (108, 308) or similar means may be provided in the upper section of a storage chamber, just downstream of any recirculation line outlet for

This works two ways: charging activator and fuel while discharging spent fuel and using activator and fuel while producing spent fuel.

2

This limits the volume of the overall system and maximizes the advantages of flexible storage systems.

3 A single connector joins three connectors into a one lever handle to which external sources can connect.

-22- even distribution.7 The outlet of the recirculation line (120, 220) is provided with a maximum pressure relief valve (121, 221), such that the maximum pressure in the fuel recirculation line is 8 bar and the maximum pressure in the activator line is 9 bar.

5 The fuel (100), the activator (200) and the spent fuel storage chamber (300) are preferably provided with a gas - fluid separation membrane at the inlet for nitrogen (150, 250, 350).

The nitrogen line (340) is provided with a mechanical pressure relief valve (341) for venting any excessive pressures (>7 bar) to the environment. The nitrogen line (340) is further provided with a bypass having a pressure transducer (343) which may actuate a control valve 10 (342). During refuelling nitrogen is supplied to the storage chambers through line 340 at a pressure of 3 bar. The controller (58) will control the pressure in line 340 by means of transducer 343 and valve 342 at a level between of 3.0 and 3.2 bar.

The system is provided with a reactor chamber (400), a mixing chamber (500) and a buffer 15 chamber (600) for released hydrogen. Each of the reactor chamber, the mixing chamber, and the buffer chamber are in open communication to one another, allowing the unrestricted release of hydrogen.

Each of the storage chambers for fuel (100) and activator (200) are connected to the mixing 20 chamber (500) with a line for supplying fuel (130) and activator (230) and each of the lines (130, 230) is providing the reactor (400) with fuel through a control valve1 2 (131, 231), having an opening pressure of 8 bar for fuel and an opening pressure of 9 bar for activator. Each of the lines (130, 230) is provided with a pressure transducer (132, 232) and may be provided with a minimum and maximum pressure switch. Each of the lines (130, 230) is preferably 25 provided with fluid control sensors3 (133, 233) and volume flow meters (134, 234).

An impeller is preferred for liquids while a circulation system is preferred for dispersions. Recirculation of a dispersion allows the fuel to be evenly distributed over the total storage chamber area thus minimizing difference in concentration, while circulating liquids may result in concentration gradients. Since the activator will always be a fluid and not a dispersion, an impeller is preferred.

2

The control valve is placed at the end of the line before the entrance of the mixing chamber.

3

By locating fluid control sensors between the control valves for activator and fuel and the mixing chamber, the presence of fuel and activator in the line is monitored (or checked) and no fuel can be dosed to the mixing chamber without the presence of any activator. Thus the reaction of additional activator (for reducing the viscosity measured via recirculation) with a very high concentration of fuel in the reactor room is prevented and as a result the pressure in the reactor chamber is prevented from exceeding way above the alarm limits.

-23-

The activator line (230) may be provided with a heating mechanism1 (239) for heating the activator in order to increase the reaction rate. The activator line may also be provided with a temperature sensor (17).

5

Referring to fig. 3 the outlet of line 230 is preferably shaped as a nozzle such that activator released from this line may be brought into the mixing chamber (500) as a jet flow2, while the outlet of the fuel line (130) is shaped as an open fluid passage3 connected to a dish. The outlet providing the jet flow may be located in line with the open fluid passage of fuel line 10 130 such that the jet flow will automatically mix with any fuel released from the open fluid passage4. The jet flow is arranged .such that (most of the) protecting fluids from the solid particles of a fuel granulate dispersion are removed.5

The mixing chamber (500) preferably has multiple stages which are in open communication 15 to one another and ultimately communicating with the reactor chamber (400) and the buffer chamber (600).6

The outlet for the activator (260) and the outlet for the fuel (160) are located in a first stage (501) of the mixing chamber.

20

The lower part of the reactor chamber (400) is shaped such that this lower part will receive non-gaseous reaction products from the mixing of fuel and activator, hereinafter referred to

This may be an in-line heater positioned downstream of the control valve and upstream of the entrance to the mixing chamber.

2

The outlet is positioned downstream of the control valve and the heater and inside the mixing chamber.

3

The outlet for the fuel is positioned downstream of the control valve and inside the mixing chamber.

4

The preferred arrangement is to place the open fluid passage beneath yet close to the outlet for the activator. However, it also possible to position the outlet of the activator inside the outlet of the fuel, so that a turbulent in-line mixing zone is created.

5

Any protecting fluid which is not removed by the jet flow in first instance, will be removed during remixing in the second stage of the mixing chamber.

6 13 By providing a multiple stage mixing chamber, wherein the different stages are separated by perforated plates having a decreasing flow resistance, a pressure gradient is created which causes turbulent mixing in each stage and which drives the reaction products from one stage to the next and so on. This means that the outlet for the gas discharge to the engine (including the safety valve) cannot be placed inside the mixing chamber and will be placed in the reactor chamber. The number of stages will depend on the desired reaction rate.

-24- as "receiver area" (450).1 The receiver area preferably has a convex shape. The receiver area (450) is preferably connected to the storage chamber for spent fuel via a first (430) and a second line (440) for transporting spent fuel. The second line is a backup for the first in case the first line would be blocked.2 Each of the lines (430,440) is provided with a dis-5 charge valve (431,441) respectively and each of the discharge valves (431,441) is positioned such that essentially all of the spent fuel collected in the reactor chamber (400) may be remixed prior to being transported to the spent fuel storage chamber (300).3

The control system for the reactor chamber (400) has four fluid control levels: a minimum 10 (24) and maximum (25) control level for the collected spent fuel, a low alarm level (26), which may be equal to the minimum level and a high alarm level (27) which will always be higher than the maximum control level (25).4

Each of the discharge valves (431,441) is located below the minimum level (24) while the 15 remixing outlet valve is located below the low alarm level (26) in the receiver area (450).5

The receiver area (450) is preferably connected to the mixing chamber (500) through a bypass (420) for remixing spent fuel. The bypass (420) is provided with a pump (425) and a control valve (421), which is located below the low alarm level (26) in the reactor room. The

The non-gaseous reaction products are process spent fuel, which is collected in the receiver area by gravity and the gas pressure in the reactor room. The convex shape allows easy transport of the spent fuel from the reactor room to the spent fuel storage room.

2

The back up prevents that any electrical and/or mechanical flow problems may cause malfunction. Furthermore, it prevents blockage of the discharge valve and/or the line due to sedimentation, which may be sticky in case a fuel dispersion is used.

3

The discharge valve is preferably located at the bottom of the convex shaped receiver area and the collected spent fuel in the reactor room is preferably re-circulated at all times in order to i) assure that the fuel is used in total and ii) to enable a viscosity measurement in the collected spent fuel through a tachometer and a power sensor attached to the remixing pump.

4

The high alarm level actuates the backup (second) discharge valve while the maximum level actuates the first discharge valve. At minimum control level all actuated discharge valves will be closed. At low alarm level however, the spent fuel remixing pump is stopped - a separate process. If the fluid level goes down to the low alarm level (and at the same reaches the minimum level) the remixing pump will stop pumping and the discharge valves are closed. When the fluid level rises, the remixing pump will immediately be actuated.

5

Discharge valves are preferably at the same height as the valve/outlet for the remixing line, in order to prevent blockage due to sedimentation. A separate remixing outlet and valve in the reactor room is preferred over a combination with discharge valves and lines in order to reduce critical malfunction. Other configurations may of course be used to remix (part of) the collected spent fuel.

-25- bypass (420) for remixing spent fuel connects to a second stage (560) of the mixing chamber which is close to the first stage (550) for mixing fuel and activator.1

The reactor chamber (400) may be provided with a connector (414) and control valve (411) 5 for adding standard activator or an alternative activator to clean the system.2 Control valve (414) may be connected to the activator supply line (270).

The buffer chamber (600) is provided with a gas release line (630), which is provided with a control valve (631) and a pressure reduction valve (632) downstream of the control valve.

10 Furthermore, the gas release line is preferably provided with a flame arrester (not shown) downstream of the reduction valve to prevent the propagation of any flame into the buffer chamber. A filter (603) is preferably provided between the buffer chamber (600) and the pressure reduction valve (603) for separating reaction products and allowing hydrogen to pass.3 The buffer chamber is preferably provided with a mechanical pressure relief valve 15 (602) for safety reasons. The temperature of the released gas is preferably measured by a temperature sensor (18).

The reactor chamber (400) is preferably integrated with and also used as the buffer chamber (400), including all provisions of the reactor chamber.4 The integrated reactor chamber 20 (400) is provided with a first (11) and a second (12) pressure transducer.5 The first pressure transducer 11 is preferably located upstream of the filter (603) and the second pressure

The bypass may enter the mixing room in a stage after the first stage where the fuel and activator are mixed. But it is also possible that all outlets (fuel, activator, remixed spent fuel) are all positioned in line, where a first mixing of activator and fuel takes place followed by a second mixing with remixed spent fuel. Thus the system may provide activator only which is mixed with remixed spent fuel. The different solutions for this first and second mixing step are not described yet.

2

The system may be cleaned by positioning the connector in the mixing chamber. Furthermore, the connector may connect to the activator storage chamber through a flush line, but may also connect to an external source. In case the activator storage chamber and a flush line are used then the system may be provided with an extra pump. Alternatively the activator line and pump may be used in combination with the bypass.

3 A small volume buffer chamber will easily splash liquid to the outlet and hence to the engine due to the relatively small length to cross, and therefore requires a filter. In industrial applications very large volume buffers may be used which do not require filters since the length to cross will be long. Splashes may however still occur.

4

If the buffer chamber is the reactor chamber then the pressure reduction valve and the filter are preferably located on top of the reactor chamber. The mixing chamber may also be placed in the mid section of the reactor chamber. This goes for the safety relief valve as well in order to maximize the distance between these valves and any spent fuel splashes. Placing a filter distant from the pressure reduction valve creates a second buffer chamber which has an open connection to the reactor room.

5 A first and second sensor provides a safe system through redundancy. The one sensor sense and checks the other and vice versa. In case of a separate buffer room the buffer room pressure is sensed as well.

-26- transducer (12) is preferably located between the filter (603) and the pressure reduction valve (632).1 The reactor chamber may further be provided with a temperature sensor (16).

The remixing line (420) is preferably provided with a viscosity meter (not shown) for sensing 5 the viscosity of the spent fuel.2 The pump (425) in the bypass (420) is preferably provided with a tachometer (not shown), more preferably a Nipkov disk.3 The power line of the pump (425) is preferably provided with a power sensor (not shown).

The system in fig 1 and 2 is preferably provided with a control system (50) for controlling the 10 mixing of fuel and activator, the flow of remixed spent fuel and the discharge of collected spent fuel, wherein the control of mixing fuel and activator is independent from the control of discharging the spent fuel.4

The control system (50) is preferably connected to fluid control sensors (133,134); pressure 15 transducers (11,12,132, 232, 343); temperature sensors (16,17,18); a viscosity meter and/or a tachometer and/or a power sensor. The control system (50) is preferably provided with a user interface/display (51) and an algorithm for controlling all sensors and actuators. The controller (58) may be provided with a wireless communication system (114-5) for communicating the filling status, fuel quality, pressure safety etc.

20

For quality control purposes refuelling stations may be provided with a micro reactor.

The transducer upstream of the filter may be (partly) blocked by splashes of spent fuel. The transducer downstream of the filter will be clean. A splashed transducer will provide a disturbed signal e.g. a slower response time on pressure variations than the clean one. Furthermore, the filter may be (partly) blocked by splashes of spent fuel resulting in lower pressure readings than the transducer upstream of the filter. In case of a defect transducer the system should react to the transducer with the highest reading. A safe system will use the transducer with the highest reading to regulate. In case, both the first transducer and the filter have splashes then the first transducer will be slow compared to the second and there will be a pressure difference resulting in a technical error signal and the selection of the clean transducer downstream of the filter.

2

The viscosity may also be determined by using the tachometer connected to the pump for spent fuel recirculation combined with the power consumption of the pump and comparing this with set values. This simplifies the system and reduces cost since the volumetric pumps for fuel and activator are already provided with a tachometer in order to monitor the ratio between fuel and activator.

3 A Nipkov disk is the most simple and cost effective solution.

4

Independent controls each having dedicated alarm limits cannot influence each other and will result in a safer overall system.

-27-

Fuel (130) and activator (230) lines are preferably provided with check valves (135, 235) respectively ensuring a proper flow direction and preventing any dripping of fuel and/or activator after closing of the control valves (131, 231).

5 The reactor (400) may be provided with a temperature sensor (17) and cooling means (not shown).1

The system preferably is arranged such that the electric resistance in the conducting metal parts is less than 0.1 ohm and that the potential difference between any conducting metal is 10 less than 10 mV.2

The heat generated in the mixing chamber (500) by mixing fuel and activator is preferably removed by a first cooling system (not shown), using water as a cooling medium, such that in the mixing chamber an operating temperature range of 130 - 200°C is maintained. A second 15 cooling system (not shown) may be provided, using water as a cooling medium, to maintain the receiver area (600) and the gas outlet (630) at a maximum temperature of 40°C. The total amount of heat to be removed is approximately 40 MJ/kg H2.

Storage tanks and tank operation 20 Referring to figure 2, each of the storage tanks (100, 200,300) is provided with a level sensor (21, 22, 23) respectively. All of these level sensors preferably are provided without electrical contacts, such as Hall type sensors.

In order to prevent sedimentation of the fuel (a granulate dispersed in mineral oil) a recirculation line (120) is provided with a pump (125) and the outlet of the recirculation line is pro-25 vided upstream of a sprinkler grid (108) in the fuel storage tank (100). Similarly the spent fuel tank is provided with a remixing line (320) with a pump (325) and a sprinkler grid (308). The activator storage tank (200) is provided with an impeller (206) for mixing the activator components. Referring to fig. 1, in case the fuel is a homogeneous liquid, recirculation line 120 and pump 125 may be replaced by an impeller (106).

The cooling means may comprise a ventilator or a Seebeck element, which converts heat into electric power.

2

This equipotential provision prevents the formation of sparks due to static electricity.

-28-

The fuel storage tank (100) may be pressurized with nitrogen to prevent moisture from penetrating during refueling. As a safety precaution an overpressure vent valve may be provided which may be integrated with the fuel inlet valve (111).

5

The spent fuel tank (300) may contain a slight hydrogen pressure from post reaction of the binary fuel system which has not yet fully reacted. A pressure transducer (343) is provided to sense the pressure in the spent fuel tank (300) and if the pressure exceeds a predetermined value, the controller (50) actuates control valve (343) to release the excessive pressure to 10 the environment. Any such actuation is displayed on the interactive interface/display (51). The spent fuel tank (300) is further provided with a mechanical overpressure valve (42).

Each of the storage tanks (100, 200,300) is provided with a supply line (110, 210,310) respectively and the three supply lines are connected in a one-handle-system (114) which 15 jointly connects to the different connectors of the refueling station dispenser in a single mode of action. The one-handle-system is preferably provided with unique self-sealing connectors which are not mutually exchangeable such that emptying the spent fuel container and refueling and/or flushing may take place simultaneously. The self-sealing provision minimizes the amount of fluid that may be released upon disconnecting and eliminates am-20 bient influences (moisture ingress). It may be desirable to flush the connector (114) with water from the one-handle-system in order to remove any minimal leakage.

An automotive design may comprise flexible tanks held in a rigid container wherein the space initially occupied by the fuel and activator due to consumption is gradually replaced by 25 the spent fuel. The volume of the spent fuel is always less than the volume of the corresponding fuel and activator.

Fuel and activator supply

The fuel and activator are pumped through lines 130, 230 and recirculation lines 120, 220 30 respectively by means of pumps (125,225). The pressure relief valves (121, 221) secure a constant pressure in the lines during the different process phases. Pressure transducers -29- (132, 232) sense the pressures in the lines in order to monitor and guarantee the working pressure of the nozzles (160, 260) through controller 50.

In order to measure the amounts of fuel and activator, mass flow meters (134, 234) are re-5 spectively provided in the fuel and activator lines. Based on the measured flows of fuel and activator, the controller (50) determines the actual fuel-to-activator ratio and compares that value with the initial set value. Optical sensors (133, 233) sense the presence of fluids at the valve systems (131, 231) respectively and the signals from these sensors enable the controller (50) to prevent the uncontrolled release of hydrogen gas due to an unbalance in the fuel-10 to-activator ratio as a result of the unintended release of just fuel or just activator.

An inline heater (239) and temperature sensor (16) may be provided in the activator line (211) upstream of the release opening. Control and set values for the heater are provided by the controller (50).

15

The activator line (210) is preferably provided with a filter (237) in order to ensure that the quality of the water in the activator meets a conductance value < 0.5 pS, such that the reaction between fuel and activator may be completed.

20 Mixer and mixing chambers

Referring to fig. 3 the inline mixer (500) comprises a first stage (550) where the activator line nozzle (230) sprays a relatively powerful jet of activator fluid into a flow of fuel fluid ("jet mixer") which is released by the fuel line (130) in a first outlet dish, thereby flushing the oil from the granulate and exposing the fuel to the activator. The reacting mixture flows 25 through a first perforated separation (558) to the second stage (560) where it is guided by guide (569) to a second outlet dish of remixing line (420), where it is mixed with spent fuel which is re-circulated from the receiver area (450) to the second stage (560) of the mixer (500) by pump 425 through bypass 420.

The reaction mixture from the second stage may flow through a second perforated separa-30 tion (568) to a third stage (570) in order to allow completion of the reaction, prior to flowing -30- through a third perforated separation (578) into the reactor (400). Alternatively the reaction mixture may flow from the second stage directly into the reactor.

The outlets of fuel line (160) and activator line (260) preferably contain check valves (135) 5 and (235) respectively to prevent leakage of fuel and activator. In this way a constant opening pressure is realized. The fuel-to-activator ratio is calculated from measured fuel and activator volumes and the required opening times of the valves for the fuel and activator are determined. For safety reasons the activator check valve is always opened prior to opening the fuel check valve and is always closed after closing the fuel check valve.

10

Spent fuel control

The spent fuel is the final product of the reaction between fuel and activator and the raw material for recycling to fresh fuel and activator.

15 In order to make sure that the spent fuel is completely exhausted, it is re-circulated to the mixer through line 420 by actuating pump 425 and valve 421. This intermediate process is controlled between minimum and maximum liquid level by level switches (24, 25). In operation a constant pumping rate of the re-circulation pump is maintained by the controller (50) using the input from a tachometer. By also determining the power absorbed by the pump at 20 that rate, a measure for the viscosity of the spent fuel is determined. By using a "lean" fuel-to-activator ratio, additional activator is required for the complete release of all hydrogen stored in the hydride fuel. The amount of additional activator can be controlled by the controller (50) based on the viscosity of the spent fuel.

25 Hydrogen pressure control

By mixing fuel and activator hydrogen gas is released instantaneously. Starting and stopping the simultaneous flow of fuel and activator implies starting and stopping the release of hydrogen gas. This allows the process to be controlled. The amount of hydrogen gas released depends on the amount of fuel injected, since completion of the process requires an excess 30 of activator to be present in the reactor. The spent fuel control is similarly adjusted. By actuating valves 131 and 231 in the fuel (130) and activator (230) lines, a pressure increase of -31 - pump 235 in the activator line (230) suffices to increase the amount of activator and thereby adjust the fuel-to-activator ratio and adjust the viscosity.

The chemical reaction between fuel and activator is independent of the pressure generated 5 in the reactor. This does not affect the intended control range. The hydrogen pressure in the reactor is also used to displace spent fuel from the reactor (400) to the spent fuel storage tank (300) through discharge lines 430 and/or 440. Such displacement is controlled by actuating discharge valves 431 and/or 441. The reason for double discharge lines and valves is to create redundancy for possible failures which may occur in the discharge system.

10

The outlet of the reactor (400) is provided with a gas/fluid filter (603) to prevent fluids to be released from the reactor. Since this filter may be blocked, a first pressure transducer (11) is provided in the reactor and a second pressure transducer (12) is provided in the hydrogen gas line (630). By comparing the recorded pressure curves of first (11) and second (12) 15 transducer, the algorithm of the controller (50) may signal any pressure differences indicating e.g. blockage of filter 603. Another safety precaution includes a specific algorithm of the controller, which continuously relates pressure increases to fuel dosage and signals any unexpected pressure increases.

-32-

Settings and action levels: Value S-l Fuel line fluid level (action if no fluid) Fluid S-2 Activator line fluid level (action if no fluid) Fluid S-3 Activator pump rate (default) 5 S-4 Fuel pump rate (default) S-5 Fuel-to-activator volumetric ratio (default) 100/90 S-6 Control valve default status (all control valves) Closed S-7 Fuel storage chamber lower liquid level S-8 Fuel storage chamber upper liquid level 10 S-9 Activator storage chamber lower liquid level S-10 Activator storage chamber upper liquid level S-ll Spent fuel storage chamber liquid alarm level S-12 Receiver area lower liquid control level 24 S-13 Receiver area upper liquid control level 25 15 S-14 Receiver area lower liquid alarm level 26 S-15 Receiver area upper liquid alarm level 27 S-16 Spent fuel viscosity default value (power at default pump rate)32 S-17 Fuel volume (to be defined) S-18 Fuel line bypass opening pressure 8.0 bar 20 S-19 Activator volume (to be defined) S-20 Activator line bypass opening pressure33 9.0 bar S-21 Reactor chamber lower pressure control level 4.5 bar S-22 Reactor chamber upper pressure control level 5.0 bar S-23 Reactor chamber lower pressure alarm level 4.0 bar 25 S-24 Reactor chamber upper pressure alarm level 6 bar S-25 Reactor chamber mechanical relief valve action level 8 bar S-26 Reactor chamber upper disabling pressure for activator >5 bar S-27 Reactor chamber upper disabling pressure for fuel >5 bar S-28 Reduction valve pressure 0.5 bar 32 This value corresponds with a fuel to activator ration of 100/110 33 Pressure difference between activator line and fuel line creates a shear stress in the mixing chamber that facilitates the break up of the oil film at the surface of the fuel particles -33- S-29 Fuel and spent fuel storage chamber upper pressure alarm level 4.5 bar S-30 Fuel and spent fuel storage chamber mechanical relief valve action level 7 bar

S-31 Reactor chamber upper temperature level 80°C

S-32 Nitrogen fill pressure storage chambers 3 bar 5 S-33 Control band nitrogen fill pressure storage chambers 3.0 - 3.2 bar

The following settings and action levels are part of the system control band: • Fluid levels for fuel line (S-l) and activator line (S-2) • Default fuel to activator volumetric ratio (S-5) 10 • Receiver area lower (S-12) and upper (S-13) liquid control levels • Default spent fuel viscosity (S-16) • Fuel volume (S-17) • Activator volume (S-19) • Reactor chamber lower (S-21) and upper (S-22) pressure control level 15 • Reduction valve pressure (S-28)

Parameters R-l Fuel valve 131 opening time R-2 Activator valve 231 opening time 20 R-3 Fuel-to-activator ratio (calculated from R-l/R-2) R-4 Rate remixing pump 4251 R-5 Power consumption remixing pump 425

For granulate type fuels only -34-

Algorithms:

Start operation 1. On power up, all sensors are checked by the control system 50.

2. The level sensors (21, 22, and 23) indicate the amount of fuel, activator and spent fuel 5 in the storage chambers (100, 200, and 300).

3. The pressure transducer 342 measures the actual pressure in the storage chambers 100-300.

4. The pressure transducers (132,232) sense the actual pressure in the lines for fuel and activator (130, 230).

10 5. The pressure transducers (11,12) sense the actual pressure in the reactor chamber (upstream and downstream of filter 603).

6. The upstream and downstream pressures according to pressure transducers (11,12) are compared (check) and the best value is selected.

7. If the pressures according to pressure transducers (11,12) differ more than e.g. 10% 15 this is signalled to a user interface/display 51 as an early warning of blockage of filter 603.

8. If this apparent pressure difference persists or increases, the filter must be replaced.

9. The level sensors 24-25 indicates the amount of spent fuel collected in the reactor chamber.

20 10. The optical sensors (133, 233) sense the presence of fluid in the fuel and activator lines.

11. If a level sensor senses no fluid (fuel or activator) the particular control valve and pump are actuated until the level sensor senses fluid or until a standard time has passed. If the level sensor still senses no fluid the system does not start and indicates (an) empty 25 line(s).

12. The temperature sensor 17 senses the temperature of the activator line.

13. The temperature sensor 16 senses the temperature of the reactor.

14. If steps 1-13 are within the control band the system starts.

-35- 15. The fuel pump 125 and activator pump 225 (and spent fuel impeller 306or remixing pump 3251) are started, starting the automatic recirculation of fuel and activator, based on the opening pressure of the valves in the fuel and activator bypass.

16. Based on the fuel-to-activator ratio S-5 the activator valve opening time R-2 is set.

5 17. A demand for hydrogen will cause the pressure in the reactor chamber to drop below the upper pressure control level S-22, as sensed by pressure transducers (11,12) whereupon the control valves (131,231) in the fuel and activator lines are actuated.

18. The fuel control valve (131) can only be actuated if the activator control valve (231) is actuated and the activator control valve can only be closed if the fuel control valve is 10 closed.

19. Fuel and activator start to flow as sensed by the fuel line volume flow meters for fuel and activator (134, 234).

20. The ratio between fuel and activator is controlled by the opening times of valves 131 and 232 based on the volumes measured by flow meters 134 and 234.

15 21. The fuel line pressure is limited to 8.0 bars and the activator line pressure is limited to 9.0 bars by the valves in the respective bypasses.

22. The (adjusted) fuel valve opening time R-l creates an offset between the calculated fuel to activator ratio R-3 and the default ratio S-5. The activator valve opening time R-2 is readjusted to compensate this offset.2 20 23. Due to the release of hydrogen the pressure in the reactor chamber increases.

24. During startup no action is taken when the pressure in the reactor chamber increases to a higher value than the lower pressure alarm level (e.g. 4 bars).

25. Upon reaching the reactor chamber lower pressure control level S-21 (e.g. 4.5 bars), the gas release valve (631) is actuated allowing the pressure reduction valve (632) to 25 supply hydrogen at e.g. 3 bars to e.g. an engine or fuel cell.

Fully operational 26. The-activator and fuel valve opening times are controlled in a master - slave fashion according to steps 17 through 22.

For granulate type fuels only 2

Activator valve opening time is master; fuel valve opening time is slave.

-36- 27. As a check on the control mechanism, the values from the volume flow sensors for fuel and activator (134, 234) may be integrated overtime, and adjusted to a fuel to activator volume ratio of 100/110.

28. During operation the spent fuel is collected in the receiver area 450.1 5 29. Upon exceeding the lower liquid alarm level S-14 in the receiver area 450 the control valve 421 in the bypass 420 is actuated and the spent fuel remixing pump 425 is actuated2.

30. The power consumption of the reactor impeller (or recirculation pump 425) is measured (as well as the pump rate (Nipkov disk)).

10 31. The actual viscosity of the spent fuel is determined based on the values in steps 28-29.

32. The actual viscosity determined in step 29 is compared to the set value (S-16).3 33. If the actual viscosity is higher than the set value, additional activator (on top of the default volumetric ratio) is pumped to the mixing chamber up to a fuel to activator volumetric ratio of 100/110, or until the actual viscosity equals the set value (S-16).4 15 34. Upon reaching the upper liquid control level (S-13) in the receiver area, the discharge valve (431) is actuated, allowing the pressure in the reactor room to drive the spent fuel through line 430 from the receiver area (450) to the spent fuel storage chamber (300).

35. After the spent fuel in the receiver area has reached the lower liquid control level S-12, 20 the discharge valve (431) is closed.

Operational Safety 36. The fuel storage chamber 100 is preferably flushed with nitrogen via connector 114-4, which is a sub-connector of connector 114, prior to charging fuel and after charging the 25 supply line (110) is preferably flushed with nitrogen such that contact between fuel and ambient air is excluded and a nitrogen blanket is kept over the fuel to prevent the formation of explosive hydrogen/air mixtures.

Spent fuel is the residual mixture of fuel and activator 2

Steps 29 through 33 apply to granulate type fuels only 3

The default fuel to activator volumetric ratio will be such that the spent fuel viscosity is lower than the set viscosity value, which allows the spent fuel to be pumped from the receiver area to the spent fuel storage chamber.

4

Fuel to activator ratio 100/110 ensures the complete conversion of all fuel -37- 37. All sensors and actuators are preferably explosion proof.

38. By pumping fuel and activator from the storage chambers 100 and 200 and discharging spent fuel to storage chamber 300, the total volume will vary and thus the pressure of the nitrogen blanket over the liquid. The valve 342 actuated by pressure sensor 343 5 will keep the pressure below 3.2 bar (S-33). Upon reaching the upper pressure alarm level S-29 in the spent fuel storage chamber, this is signalled to the user inter-face/display (51) as an early warning that the spent fuel contains unreacted fuel and the control valve (131) is actuated until a standard volume of activator is pumped into the reaction chamber1.

10 39. If despite step 38 the pressure in the spent fuel storage chamber further increases to the action pressure level S-30 (e.g. 7 bar), the pressure relief valve 341 opens, allowing hydrogen to be released from the system.

40. Upon reaching the upper pressure alarm level S-24 in the reactor chamber (e.g. 6 bars), hydrogen release must stop immediately, therefore the control valves (131, 231) in the 15 fuel and activator lines are closed and consequently the fuel and activator pumps (125, 225) are stopped.

41. If despite step 40 the pressure in the reactor chamber further increases to the action pressure level S-25 (e.g. 8 bar), the pressure relief valve (602) opens, allowing hydrogen to be released from the system and preventing dangerous pressures.

20 42. Any pressure alarm is signalled on a user interface/display (51) and implies that the fuel-to-activator ratio needs adjustment.

43. After the pressure in the reactor chamber has dropped below the upper disabling pressure for activator S-26, the activator pump (225) is restarted and the control valve (231) is actuated until a standard volume of activator is pumped into the reaction 25 chamber.2 44. After the pressure in the reactor chamber has dropped below the upper disabling pressure for fuel S-27, the fuel pump (125) is restarted.

45. After the pressure in the reactor chamber has dropped below the upper control level S-22, the system resumes normal operation according to step 26.

A fixed volume of activator will ensure that any remaining fuel will react 2 A fixed volume of activator will ensure that any remaining fuel will react -38- 46. The values from the liquid level sensors (21,22, 23) (fuel, activator and spent fuel storage chamber levels) are continuously compared.

47. If (100 minus the spent fuel storage chamber level) deviates more than e.g. 10%1 2 from the fuel or activator storage chamber level, this is signalled to the user inter- 5 face/display (51) as an early warning of sedimentation near the discharge valve (431), restricting the flow of spent fuel through line 430.

48. The discharge valve (431) may in that case be flushed by actuating control valve (411), allowing activator to flow from storage chamber 200 through line 270 into the reactor chamber (400)..

10 49. Upon reaching the upper liquid alarm level S-15 in the receiver area, the second dis charge valve (441) is actuated, allowing the pressure in the reactor room to drive the spent fuel through line 440 from the receiver area (450) to the spent fuel storage chamber (300).

50. Step 48 may in that case be repeated.

15 51. After the spent fuel in the receiver area has reached the lower liquid control level S-12, the discharge valves (441,431) are closed.

52. Any liquid alarm (26,27) is signalled on the user interface/display (51) and implies that spent fuel sediment blocks the discharge valve (431).

53. Upon reaching the upper temperature level S-31 in the reactor chamber, the cooling 20 means are actuated until the temperature is below that level.

Stop operation 54. The system may be stopped any time, in which case the system pressure automatically settles at the maximum values of the control band.

25

Charging and discharging 55. Low fuel (S-7) and/or activator (S-9) levels are signalled on the user interface/display (51)"

Any suitable value may be set in the control system 2

At low fuel and/or activator levels, the fuel and activator pumps may be stopped to prevent air from entering the fuel and lines -39- 56. At a refuelling station, the integrated connector (114) is connected to the connector of the external supply source for fuel, activator and nitrogen, as well as the external spent fuel collection all at the same time.

57. Preferably a data communication link such as a telemetry link is automatically estab- 5 lished for exchanging data regarding system pressure, liquid levels, fuel grade etc. The data communication link may have a manual override.

58. Simultaneously fuel and activator are supplied to the storage chamber for fuel and activator (100, 200) while spent fuel is discharged from the spent fuel storage chamber (300).

10 59. A nitrogen blanket is maintained in all storage chambers by an external supply of nitro gen at a pressure of 3 bars.

60. The values from the liquid level sensors (21, 22, 23) (fuel, activator and spent fuel storage chamber levels) are continuously compared.

61. If (100 minus the spent fuel storage chamber level) deviates more than e.g. 10% from 15 the fuel or activator storage chamber level, this is signalled to a user interface/display as an early warning of sedimentation in the spent fuel storage chamber, restricting the flow of spent fuel through the valve (331).

62. The discharge valve (311) and/or the spent fuel storage chamber (300) may in that case be flushed by pumping water through the integrated connector 114 into the spent fuel 20 storage chamber.

63. Upon reaching the maximum level for fuel S-8 and activator S-10, the supply of fuel and activator to the storage chambers (100, 200) is stopped.

64. Upon reaching the minimum level for spent fuel, the discharge of spent fuel to the external spent fuel collection is stopped.

25 65. The integrated connector (114) is disconnected from the connector of the external supply source for fuel, activator and nitrogen, as well as the external spent fuel collection.

-40-

Storage Safety 66. Upon reaching the upper pressure alarm level S-29 in the spent fuel storage chamber 300 (e.g. 4.5 bar), hydrogen may be released by actuating an additional control valve (301) at the top of the spent fuel storage chamber.

5 67. If despite step 66 the pressure in the spent fuel storage chamber further increases to the action pressure level S-30 (e.g. 7 bar), the pressure relief valve 302 opens, allowing hydrogen to be released from the spent fuel storage chamber and preventing dangerous pressures.

68. Any pressure alarm is signalled to the user interface/display (51).

10 69. Each of the nitrogen inlets of the storage chambers (100, 200, and 300) is provided with an additional membrane filter in order to strictly separate the liquids.

All algorithms are preferably provided with control tables listing the control parameters, control settings and action levels of the various sensors and actuators. This way control 15 loops may provide "Yes" or "No" values when comparing a sensed control parameter to a control setting or action level.

1 0 3 76 1 8

Claims (4)

55 1. Werkwijze om waterstof op te wekken die de volgende stappen omvat: a) het voorzien in een brandstof, waarbij de brandstof een vloeistof omvat die ten minste een of meer chemische hydrides omvat, vanuit een brandstof opslagkamer via een eerste toevoer naar een reactiekamer en 10 b) het voorzien in een activator vloeistof en/of een destabilisator vanuit een activator op slagkamer en/of een destabilisator opslagkamer via een tweede toevoer naar een reactiekamer c) waarbij het voorzien van alle vloeistoffen continu en/of met onderbrekingen kan plaatsvinden, in het bijzonder tegelijkertijd, 15 d) daarbij een menging bewerkstelligen van de brandstof en activator vloeistof en/of de de stabilisator.A method for generating hydrogen comprising the steps of: a) providing a fuel, the fuel comprising a liquid comprising at least one or more chemical hydrides from a fuel storage chamber via a first feed to a reaction chamber and B) providing an activator liquid and / or a destabilizer from an activator in an impact chamber and / or a destabilizer storage chamber via a second feed to a reaction chamber c) wherein the provision of all liquids can take place continuously and / or with interruptions, in in particular at the same time, d) thereby effecting a mixing of the fuel and activator liquid and / or the stabilizer. 2. Apparaat om waterstof op te wekken uit een vloeistof die een of meer chemische hydrides omvat a, in het bijzonder met een werkwijze volgens conclusie 1, gekenmerkt door 20 a) een reactiekamer voorzien van een eerste toevoer voor een brandstof en een tweede toevoer voor een activator vloeistof en/of een destabilisator, en b) middelen om een menging van de brandstof en de activator vloeistof en/of de destabilisator in de reactiekamer te bewerkstelligen.Device for generating hydrogen from a liquid comprising one or more chemical hydrides a, in particular with a method according to claim 1, characterized by a) a reaction chamber provided with a first feed for a fuel and a second feed for an activator liquid and / or a destabilizer, and b) means for effecting a mixing of the fuel and the activator liquid and / or the destabilizer in the reaction chamber. 3. Apparaat voor opslag van een brandstof omvattende een of meer chemische hydrides, een acti vator vloeistof en/of een destabilisator en de afvalproducten van een waterstof opwekkende reactie tussen de brandstof en de activator vloeistof en/of de destabilisator, gekenmerkt door a) een brandstof opslagkamer, b) een activator vloeistof opslagkamer en/of een destabilisator opslagkamer, en 30 c) een afval opslagkamer.An apparatus for storing a fuel comprising one or more chemical hydrides, an activator liquid and / or a destabilizer and the waste products of a hydrogen generating reaction between the fuel and the activator liquid and / or the destabilizer, characterized by a) a fuel storage chamber, b) an activator liquid storage chamber and / or a destabilizer storage chamber, and c) a waste storage chamber. 4. Apparaat volgens conclusie 3, waarbij alle kamers zijn gevat in een brandstoftank en in het bijzonder alle kamers een flexibel volume hebben. 1037618Apparatus as claimed in claim 3, wherein all chambers are contained in a fuel tank and in particular all chambers have a flexible volume. 1037618