FIELD OF THE INVENTION

The present invention is in the field of a redox flow battery for heat to power conversion. Said batteries comprise electrolytes, typically a solvent, electrodes, and a pump.

BACKGROUND OF THE INVENTION

A redox flow battery is a type of a battery for providing electrical energy typically in the form of a current. Energy is stored therein in the form of chemicals, hence the term electrochemical. In an electrochemical cell chemical energy is typically provided by two chemical components which may be dissolved in liquids, such as water, contained within a system. The electrochemical cell reversibly converts chemical energy directly to electricity, using e.g. electroactive elements in solution that can take part in an electrode reaction or that can be adsorbed on an electrode. Additional electrolyte is typically stored externally from the cell itself, such as in (small) tanks. The electrolyte is then usually pumped through the cell's reactor compartment. Flow batteries can be rapidly recharged, such as by replacing the electrolyte liquid whereas converted redox species may be recovered. The two chemical components are separated such as by a membrane. The electrochemical cell typically involves ion transport. Ion transport occurs through the membrane, such as an ion exchange membrane. Both liquids can circulate (hence flow) in their own respective flow path. Over the ion exchange part also a flow of electric current is established, when in use. An electrochemical cell voltage is determined by the chemicals used and is considered to follow the Nernst equation and ranges. In practical applications the (absolute) cell voltages may vary from 0.2 to 2.5 volts.

A flow battery may be used as a fuel cell and as a rechargeable battery. Some technical advantages over prior art rechargeable batteries are separable liquid tanks and extended use, present implementations are comparatively less powerful and require more sophisticated electronics.

Various types of flow cells exist, such as redox, hybrid, organic, metal hydride, nano-network, semi-solid, and without membrane. As mentioned above, a fundamental difference between conventional batteries and flow cells is that energy is stored not as the electrode material in conventional batteries but as the electrolyte in flow cells.

Clearly the energy capacity is a function of electrolyte volume, solvent, and type of electrolyte, whereas power is a function of surface area of the electrodes. Typical power densities are from about 1000-20000 W/m ² , a fluid energy density is from about 10-1500 Wh/kg, and a number of recharging cycles is from about 10-2000.

Redox flow batteries have certain advantages, such as a flexible layout, a long cycle life, quick response times, no harmful emissions, easy state-of-charge determination, low maintenance costs, good tolerance to overcharge and to over discharge, high current and power densities, which are suited for large-scale energy storage. However energy densities and efficiency are in general lower, compared to solid battery alternatives .

Flow batteries can be applied in relatively large (1 kWh -10 MWh) stationary applications. They may be applied for load balancing. Therein the flow battery is connected to an electrical grid to store excess electrical power during off- peak hours and release electrical power during peak demand periods. They may be applied for storing energy, such as from renewable sources as wind or solar, and for discharging during periods of peak demand. The may be used for providing an uninterrupted supply and for peak shaving. They may be used in combination, such as in power conversion. The electrolyte may be charged using a given number of cells and discharged with a different number of cells, or likewise cycles. The battery can be used in combination with a DC-DC converter. Power conversion can also be AC/DC, AC/AC, or DC-AC. Flow batteries can be used in vehicles. And they can be used as a stand-alone power system.

Apart from the above developments, redox flow batteries are according to the inventors not used in further applications .

Lee et al. in "An electrochemical system for efficiently harvesting low-grade heat energy", Nature Comm., 2014 use solid batteries, with Cu(CN) 6 and Cu ²⁺ /Cu as redox species for heat to power conversion.

Some further documents recite background art. For instance US 2010/003545 Al recites a redox Flow Battery System for Distributed Energy Storage comprising a large stack for energy storage of many types of renewable energy systems. Independent reaction cells arranged in a cascade configuration are configured according to state of charge conditions ex ^¬ pected in each cell, US 2015/099150 Al recites an electro- chemical system for harvesting heat energy, and associated electrochemical cells and methods, Ponce de Leon et al. in J. Power Sources, Elsevier, Vol. 160, Nr. 1, September 29, 2006, p. 716-732 recites energy storage technologies for balancing power generation and power consumption. Therein redox flow cells are designed to convert and store electrical energy into chemical energy and release it in a controlled fashion when required. Redox flow systems are compared in the light of open circuit potential, power density, energy efficiency and charge-discharge behavior, and US 2014/051002 Al recites an aqueous redox flow batteries comprising ionically charged redox active materials and ionomer membranes, wherein the charge of the redox active materials is of the same sign as that of the ionomer.

The present invention relates to an improved redox flow battery for heat to power conversion which overcomes one or more of the above disadvantages, without jeopardizing functionality and advantages.

SUMMARY OF THE INVENTION

The present invention relates in a first aspect to a redox flow battery for heat to power conversion according to claim 1. The power density is typically somewhat lower than prior art ones (10-5000 W/m ² ) , whereas a number of recharging cycles is higher (100-10000). The present battery has an increased power density, is very well scalable, has a low self-discharge, and so on.

The voltage of the present redox flow battery is dependent on the temperature. This feature is used to convert heat partly into electricity, wherein one flow cell operates at a low temperature and one at high temperature. In the present flow battery this process can be used in a continuous mode. The present invention also focusses on requirements of the redox species that are used for this heat-to-power redox flow batteries. In an example inventors use redox flow batteries based on hexacyanoferrate (Fe(CN)e ^4~ / ^3~ ) and polyiodide (I3 ^" to 31 ^" ) . This novel composition works excellent as heat-to- power device as the voltage differences are found to strongly depend on temperature, over-potentials are very small and they can operate in a wide temperature range. Moreover, the system is easy for practical operation because both active species used have the same charge (negative) , so can be separated well with a cation exchange membrane. Also, exemplary chemicals are widely and relatively cheap available. As the absolute potential of the flow battery is small, self-dis- charge is very limited. Inventors found this combination after elaborate research and consider the characteristics (as effectiveness) of the set of redox species a coincidence. The flow battery has been validated for converting a part of the heat into power. In a general perspective it is noted that an amount of waste heat produced in society is larger than the total energy consumption, the present invention provides a huge opportunity. The use of flow batteries allows a continuous and efficient process as flowing electrolyte can be easily exposed to hot/cold sources by using a heat exchanger. The present redox flow battery is found to be stable over time and during use. No deposits are formed. More specifically, compared to prior art, the present solution does not suffer from an incompatibility of redox species. It has been established that a redox flow battery for heat to power gen- eration needs amongst others two redox reactions that have an opposite dependency on temperature and at the same time are separable by a (ion exchange) membrane. No such compatible combination was identified so far. Using in an example

Fe(CN)6 and I3/I, this problem is solved. It is noted that the available literature predicted a decrease in potential for higher temperature for l3 ^" /I ^" , but surprisingly experiments showed the opposite. So the voltage dependency on temperature for l3 ^" /I ^" was not obvious. It is thus the first time a redox flow batteries are successfully used for heat to power conversion. The use of a flow battery is cheaper and faster than the use of solid bat ^¬ teries for heat to power conversion, as heating/cooling of a fluid is much faster in a heat exchanger than heat fluxes through solid materials.

In an example of the present redox flow battery heat is provided by a fluid, such as water and air, such as water from a power plant, by conduction, by radiation, by convec- tion, or a combination thereof. In order to drive the present flow cell, a flow of warm/hot fluid, such as water, is pro ^¬ vided. A flow 260 of e.g. hot water is typically passed over a heat exchanger 265 (see fig. 3), leaving the battery as a colder fluid 261. In an alternative, a redox flow battery op- erates in a hot chamber, where heat is provided such as via conduction, radiation, or convection. At the same time, another cell (which may be regarded as a redox flow battery as well) operates at low temperature. Not all heat can be con ^¬ verted into electricity, and part of the heat needs to be re- moved from the cold cell, such as via a fluid, conduction or radiation.

In an example of the present redox flow battery two flows can be distinguished, a first flow comprising a set of first redox species, and a second flow comprising a set of second redox species (see e.g. fig. 3a, flows 231 and 232). These flows are physically separated from one and another. Also in the hot and cold cell, respectively, these flows are separated, in casu by a membrane; these cells therefore may be considered to have at least two chambers, separated by the membrane. In the hot cell the first redox species is subjected to an oxidation reaction, therewith releasing elec ^¬ trons, and wherein the second redox species is subjected to a reduction reaction, therewith incorporating electrons, wherein in the cold cell the second redox species is sub- jected to an oxidation reaction and wherein the first redox species is subjected to a reduction reaction. Unexpectedly it has been found that the first set of redox species, having a first Seebeck coefficient, and the second set of redox spe- cies, having a second Seebeck coefficient, an absolute difference between said first and second Seebeck coefficient ( I Seebecki-Seebeck21 ) is > lmV/K, over the temperature range T _c -T . In order for the present redox flow cell to function properly in terms of redox reactions the difference in Seebeck coefficients must be large enough; a smaller difference is found to be insufficient. It is preferred to have a difference of > 2 mV/K, such as > 2.5 mV/K. In order to convert sufficient heat into electrical power also the solubility of the first and second sets of redox species is >0.1 M over the temperature range T _c -Th- The temperature range can be as large as 0°C-150°C, such as when very hot water (at elevated pressure) is used and cold water is released. This flow of fluid cools down over the present flow cell, typically from a tem- perature of about Th at flow cell 220 to a temperature of about Tc at flow cell 210; it is noted that an incoming flow 260 typically has a temperature less than T _h and an outgoing flow 261 typically has a temperature lower than Tc. The temperature T _h may be from 283-423 K ( 10 °C-150 °C) ; the tempera- ture T _c is typically 5-120K lower than the temperature T _h , preferably 10-100 K lower, more preferably 20-90 K lower, such as 40-75K lower. The temperature T _c is typically

slightly higher (2-10 K, such as 5K) than an environmental temperature. In view of heat management both flows are passed over a heat exchanger 250. Therewith the energy efficiency in the conversion of heat to electrical power is optimised. Also therewith a temperature of the hot and cold cell, respectively, is maintained at a substantially constant temperature. For providing flow typically a pump 14 is provided, preferably a pump per flow.

In the present redox flow battery the electrolyte is dissolved in a solvent, such as water. The fluid comprising said electrolyte is also referred to as "fluid electrolyte".

In the present redox flow battery an over-potential of the first and second sets of redox species half-reactions is < 0.1 V, and preferably < 50 mV. Such is required to minimize losses in the present battery. In electrochemistry the over- potential refers to a potential difference between a (first or second) half-reaction ' s thermodynamically determined reduction potential and the potential at which the redox event is observed in practice, the latter being slightly higher. Overpotential therefore is an important factor in the present redox flow battery electrical efficiency. In other words an overpotential implies that more energy is needed for a half- reaction than would be thermodynamically expected to drive said reaction. As a result less energy is recovered than thermodynamics predicts. Typically energy is lost as heat. An overpotential is considered to be specific to a cell design and may vary across cells and operational conditions. An overpotential can be experimentally determined, such as by measuring the potential at which a certain minimum current density is achieved, being representative for the start of the reaction.

In the present redox flow battery typically two electrical contacts 13, for current collection, are present. Typically one contact 13a, c is at a lower potential compared to the other contact 13b, d. Contact 13a, c may be at a negative potential, whereas contact 13b, d may be at a positive potential. Typically contacts 13a and 13 c of the hot and cold cell are in electrical contact with one and another, and likewise contacts 13b and 13 d of the hot and cold cell are in contact with a power providing unit. The present redox flow battery as such can provide an electrical power output. The contacts may be used to electrically connect the present battery such as to an electrical power requiring apparatus, or to a heat producing unit at opposite current direction.

In the present redox flow battery it is required that the first set of redox species and the second set of redox species have a same polarity (either both + or both -) . In order for the hot and cold cell, respectively, to provide a current a proper membrane needs to be selected; the membrane provides for ion transport and therefore the redox species have a same polarity. The ion exchange membrane has the opposite polarity of the first and second sets of redox species; if the species have a positive polarity the membrane allows transport of ions with negative polarity, and vice versa. The present system or present array may be used for generating electricity, such as when an electrical load is at ^¬ tached to the system or array.

In a second aspect the present invention relates to an array comprising two or more redox flow battery system ac ^¬ cording to the present invention in series and/or in paral ^¬ lel, such as 3-5 systems in series and/or 2-5 systems in parallel. An array, especially when comprising systems in series, can provide higher temperature differences between hot and cold reservoir, which is found to comply better to re ^¬ quirements of equipment producing heat.

In a third aspect the present invention relates to heat pump comprising a redox flow battery system according to the invention or an array according to the invention, adapted to convert power into heat. The present invention may therefore mutatis mutandis used to produce heat from power as well, which may for instance be advantageous when storing power during daytime and consuming heat during night time. In other words, during a certain period of time, typically hours, the present system may be used to convert heat into electrical power, and during a further period of time, typically hours, the present system may be used to convert electrical power into heat.

In a fourth aspect the present invention relates to a heat producing entity, such as a power plant, comprising the present redox flow battery or array. As such entities produce huge amounts of waste heat, the potential of stripping said heat and converting the heat into electricity is enormous.

Thereby the present invention provides a solution to one or more of the above mentioned problems.

Advantages of the present description are detailed throughout the description. References to the figures are not limiting, and are only intended to guide the person skilled in the art through details of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

The present invention relates in a first aspect to a redox flow battery for heat to power conversion according to claim 1.

In an exemplary embodiment of the present redox flow battery an average residence time of the electrolytes in the hot cell (220) and in the cold cell (210), respectively, may be 1-100 sec, such as 10-20 sec. For good results, e.g. in terms of energy conversion efficiency and power density, the residence time is preferably not too long and no too short.

In an exemplary embodiment of the present redox flow battery the solvent may be selected from water, polar organic solvents, such as alcohols, such as ethanol, and mix- tures thereof. It is preferred to use water, e.g. in view of the high solubility of charge carrying ions, compared to the organic solvents.

In an exemplary embodiment of the present redox flow battery solvent with the electrolyte may have a conductiv- ity of > 50 mS/cm, such as by addition of a 0.1-2M of a salt, such as KC1. In view of performance of the flow cell, and specifically of the hot and cold cell, respectively, the conductivity is high enough. Such can be achieved by addition of a, typically not interfering, salt, such as KC1.

In an exemplary embodiment of the present redox flow battery the redox potentials of both first and second sets of redox species may be in a range of 0 V-1.23 V with respect to a reversible hydrogen electrode (RHE) . Such is especially relevant when water is used as solvent.

In an exemplary embodiment of the present redox flow battery the first set of redox species may have a positive Seebeck coefficient (>0 mV/K) and wherein the second set of redox species may have a negative Seebeck coefficient (<0 mV/K) (over the temperature range Tc-Th) . Therewith to obtain a large difference in Seebeck coefficients for the two half-reactions, and the amount of power that can be harvested is increased.

In an exemplary embodiment of the present redox flow battery the first redox species may be present in a concentration of 0.1-2 M, such as 0.2-0.6 M.

In an exemplary embodiment of the present redox flow battery the second redox species may be present in a concentration of 0.1-6 M, preferably 0.2-5 M, more preferably 0.5-4 M, such as 1-3M.

In an exemplary embodiment of the present redox flow battery the first flow 231 may further comprise solid particles which form a first suspension.

In an exemplary embodiment of the present redox flow battery the second flow 232 may comprise solid particles which form a second suspension.

The above solid particles may be present in the form of solidified species, such as in crystalline form, or may be present as adsorbed to an absorbent, such as carbon black .

In an exemplary embodiment of the present redox flow battery the first suspension may be adapted to pro-vide first redox species in a molar equivalent of 0.1-20M, preferably 0.2-10 M, such as 5-8M. As such the suspension is capable of additional first redox species, e.g. when re-quired. Higher molarity provide better yields in terms of heat to power conversion, and vice versa.

In an exemplary embodiment of the present redox flow bat- tery the second suspension may be adapted to provide second redox species in a molar equivalent of 0.1-20M, prefer-ably 0.2-10 M, such as 5-8M, wherein the molarities are relative to the respective flows.

The redox species may be preferably present in an as high as possible concentration, e.g. close to a maximum solubility. It is important that the species are soluble over the entire temperature operation range; preferably no or minor deposits are formed. When the concentrations are too low the efficiency of the cells drops.

In an exemplary embodiment of the present redox flow battery the first redox species may be selected from

Fe{CN} ₆ , such as Fe(CN) ₆ ^4" and Fe(CN) ₆ ^3_ , and alloxazine car- boxylic acid (ACA) , and the second redox species are selected from iodide I-, polyiodide, such as I3 ^" , sulfide, such as S ^2_ , and polysulfide, such as S2 ^2" .

In an exemplary embodiment of the present redox flow battery the first set of redox species comprises (2 times} Fe(CN) ₆ ^4~ and Fe(CN) ₆ ^3" + le and the second set of redox species may comprise 31 ^" and I3 ^" + 2e. This combination is found to work particularly well.

In an exemplary embodiment of the present redox flow battery the membrane (10) may be a cation exchange membrane or an anion exchange membrane.

In an exemplary embodiment of the present redox flow battery the membrane may be selected from polymers, preferably polyethers, such as Sulfonated Poly (Ether Ketone) (SPEEK), ethylene-acrylate polymers in case of a cation exchange membrane, those comprising at least one quater- nary ammonium group, in case of an anion exchange membrane. It has been found that for instance other membranes, such as inorganic membranes, such as precipitate membranes, such as BaS04, do not function sufficiently in the present flow cell.

In an exemplary embodiment of the present redox flow battery a pH difference over the ion exchange membrane (10) may be <1, preferably < 0.5. It is preferred to have an as small as possible pH difference over the membrane, such as in view of energy efficiency and stability of the cells.

In an exemplary embodiment of the present redox flow battery at least one of a first flow 231 and second flow 232 may comprise a pH buffer, such as phosphate and borate. Therewith the pH is stabilized.

In an exemplary embodiment of the present redox flow battery a current density magnitude may be 0.1-1000 mA/cm ² , such as 1-100 mA/cm ² . Depending on the way the system is used, i.e. as battery or as heat producing entity, the density may be + or -.

In an exemplary embodiment of the present redox flow battery an output voltage (which is the difference in cell voltages) is 0.1-1.1 V, preferably 0.2-1.0 V, such as 0.4- 0.8 V. A maximum power output is achieved as such.

In an exemplary embodiment of the present redox flow battery the solubility of a mixed first and second sets of redox species may be >0.1 M over the temperature range T _c -Th. Typically over time some leakage may occur, especially over the membrane and openings which are difficult to prevent from existing. In view thereof it is preferred that if the first set of redox species is mixed with the second set of redox species no deposits are formed. Therefore a solubility of such a mixture is >0.1 M such that only minor deposits are formed of mixed redox species, and preferably >0.2M, more preferably >0.5M, such as > 1M. For example when Cu-NH ₃ as first redox species and Fe (CN) e as second redox species are used Cu2Fe (CNe) is formed within a day, typically within 6 hours. On the contrary, the present battery performs well at least over weeks' time (being the test period) .

In an exemplary embodiment the present redox flow battery may comprise at least one of a catholyte tank 11 and an anolyte tank 12. These tanks may suffer as a storage for the present electrolytes.

The invention is further detailed by the accompanying figures and examples, which are exemplary and explanatory of nature and are not limiting the scope of the invention. To the person skilled in the art it may be clear that many variants, being obvious or not, may be conceivable falling within the scope of protection, defined by the present claims .

SUMMARY OF FIGURES

Figure 1 shows principles of a prior art redox flow battery.

Fig. 2a shows schematics of a prior art flow cell.

Figures 2b-4 show a schematic representation of an example of the present battery.

DETAILED DESCRIPTION OF FIGURES

100 redox flow battery

10 membrane

11 catholyte tank

12 anolyte tank

13 contact (current collector)

13a negative contact cold cell

13b positive contact cold cell

13c negative contact hot cell

13d positive contact hot cell

14 pump

15 current flow

200 system of redox flow batteries 210 cold cell

220 hot cell

231 first electrolyte flow

232 second electrolyte flow

250 heat exchanger

260 hot fluid flow in

261 colder flow out

265 heat exchanger

Figure 1 shows principles of a prior art redox flow bat- tery. Therein a single cell is shown. The cell comprises a membrane 10, and contacts 13 (current collector) . Also a catholyte tank 11 and an anolyte tank 12 is shown. Two pumps 14 are provided for driving a flow. As a result an electrical current 15 flows.

Figure 2a shows changes in entropy and temperature for a prior art flow cell. In a first step a cell is heated. In a second step a cell is charged, therewith increasing the entropy. Than the cell is cooled, therewith lowering the temperature and slightly lowering the entropy. In a last step of the cycle present the cell is discharged, therewith lowering the entropy.

In a similar manner fig. 2b shows schematically the functioning of the present flow cell, comprising a hot cell and a cold cell. It is preferred to have a surface area being as large as possible, such as by selecting appropriate redox species and a large temperature difference.

Figure 3a shows a schematic layout of the present redox flow cell, as detailed through the description.

Figure 3b shows a cell wherein the membrane 10 is a cat- ion (+) exchange membrane and wherein the redox species are all negatively (-) charged.

Figure 3c shows a cell wherein the membrane 10 is an anion (-} exchange membrane and wherein the redox species are all positively (+) charged.

Figure 4 shows an exemplary embodiment of the present flow cell.

The figures are further detailed in the description of the experiments below. EXAMPLES/EXPERIMENTS

An exemplary flow cell was designed. A smaller flow cell has a circular electrode with a radius of 1 cm and was scaled up to 5 cm in the design. The design of the flow cell was done in parallel to choosing the electrolytes. The materials therefore had to be resistant to a large amount of chemical conditions (e.g. Alkaline, acidic or presence of bromine) at a wide temperature range. The ideal material for this is Teflon (PTFE) and has a maximum operating temperature of 260 °C. Teflon is a soft material, which is not suitable for thin (< 3 mm) parts. As second material, a polyamide (PA2200) was used. The flow cells contain three designed parts:

A PA2200 (a polyamide based plastic) 3-D printed spacer, manufactured by Oceanz 3D printing. Before building the flow cell it was sanded to smoothen the rough surface. A Teflon Back plate and an aluminum support plate is used. Created by drilling multiple holes in an aluminum plate. Teflon is a very soft material. In order to prevent it from deforming from the force of the screws, an aluminum plate was added to spread the force more evenly across the plates. In both Teflon back plates, graphite foil electrodes were placed. These electrodes were connected with two wires each through two holes in the back plate. To prevent short circuiting and leakages a rubber O-ring was added, to seal the wires from any fluid coming in. This O-ring was custom created by supergluing two ends of an O-ring cable together. On the top of the back plate, there is a hole for the reference electrode to measure the outflow. Ag/AgCl electrodes were used and were also sealed with O-rings and a custom made Teflon screw to prevent air coming into the electrolytes. The two sides were separated by a FKB-PK-130 cation exchange membrane from FuMaTech. All parts were pressed together with 300 μιη thick Silicone gaskets between the back plates and spacers and the spacers and membrane. Finally everything was screwed together with M12 steel screws.

The following aspects are taken into account:

Iron parts are protected, when using Fe ( C ) 6 ^3~ /Fe ( C ) 6 ⁴ ~ to prevent formation of Prussian blue. All materials are stable and do not undergo phase change on a large temperature interval (at least 0 - 80 °C) . In this temperature interval the materials are resistant under the operational pH conditions and possible corrosive electrolytes. The materials in contact with the electrolyte do not conduct electricity (apart from the electrode) .

To test the setup, the Seebeck coefficient of the Fe (CN) 6 ^3" /Fe (CN) 6 ^{4 "} was measured as -1.4 mV/K. For the ACA the Seebeck coefficient was measured as -1.5 mV/K. Contrary to literature the Seebeck coefficient of the Ϊ-/Ι3 ^" couple, turned out to be positive (+1.0 mV/K) .

When adding ethanol as solvent the following was noted. The Seebeck coefficient of Fe(CN)e becomes stronger negative with higher wt% of ethanol in the mixture. The solutions of 0, 10 and 20 wt% showed in a decrease of -1.3 to -2.3 mV/K. The Seebeck coefficient of I/I3 becomes stronger positive with higher wt% of ethanol in the mixture. The solutions of 0, 10, and 20 wt% showed in an increase of +0.9 to +1.9 mV/K. The Seebeck coefficient of I/I3 increased for lower concentrations (numbers still need to be calculated but the change is in the order of 0.1-0.3 mV/K) . So the addition of ethanol to the electrolytes may increase a full cell a for e.g. the Fe(CN)e and I/I3 system by almost a factor two.

It is found that the present process is reversible, as the open cell potential did not change (+0.5 mV} after run- ning the cell for almost an hour. A fit of the I/V curve resulted in a high resistance of 44.2 Ω cm ² , the ACA-Fe(CN) 6 had a resistance of 1.03 Ω cm ² , and the polysulfide/iodide had a resistance of 3-4 Ω cm ² .

After disassembling the battery components were still largely intact .

In a second experiment, the cell was connected to a heat bath and the in- and outflow temperatures were measured. Four measurements were done, with the heating bath at 30, 35, 40 and 45 °C, respectively. The Seebeck coefficient was found to be slightly lower than, but very close to, the expected +2.35-2.45 mV/K. The cell performed well.

When performing calculation an optimal current density was found to be 1.27 mA/cm ² . The maximum power output for the example is only 8.10 mW yet. When improving the design a Carnot efficiency of about 50% is considered feasible; for instance the temperature difference is preferably much higher than in the examples, such as > 50 K, a low current density, a low Ohmic resistance, etc.

The invention although described in detailed explanatory context may be best understood in conjunction with the accompanying figures.

It should be appreciated that for commercial application it may be preferable to use one or more variations of the present system, which would similar be to the ones disclosed in the present application and are within the spirit of the invention .