NO119186B - - Google Patents

Description

Thermoelectric heat pump.

The invention relates to heat pumps, and

more particularly heat pumps that include single or multiple connection points between different thermoelectric orga-

down.

A main purpose of the invention is

to obtain thermoelectric heat pumps that produce sufficiently strong heat transfer

ing and sufficiently high temperature differences for practical use.

Another purpose is to obtain heat pumps of this type with high efficiency

degree and good functional ability.

It is previously known that one can up-

now advantageous materials for thermoelectric

tric bodies, especially for cooling purposes, by using alloys whose composition of

deviates somewhat from perfect stoichiometric quantity ratios.

The present invention is based on

a thermoelectric heat pump where this precaution is used, but un-

where the use of materials that show significant advantages over those that have been previously known for the purpose. This will be explained in the following presentation with reference to the drawing. Fig. 1 schematically shows a heat pump according to an embodiment of the invention and provided with a single thermoelectric device.

Fig. 2 schematically shows a compound

or double thermoelectric heat pump according to an embodiment of the invention with thermoelectric devices of opposite polarity.

Fig. 3 is a graphical representation of certain lead-selenium-tellurium compositions which form

thermoelectric heat pump materials according to the invention.

Fig. 4 is a graphic representation of

certain lead-selenium-sulfur compositions comprising thermoelectric heat pump materials according to the invention. Fig. 5 graphically shows certain electrical characteristics of negative thermoelectric heat pump devices that can be used according to the invention (the meaning of positive and negative in this connection will be explained later). Fig. 5A graphically shows certain electrical characteristics of certain positive thermoelectric heat pump devices that can use

tes according to the invention.

Fig. 6 graphically shows operating characteristic

ker for heat pumps according to the invention, and

fig. 7 graphically shows certain electrical characteristics of additional positive thermoelectric heat pump devices in

stick to the invention.

When a direct current is sent through a junction between two different conductors, the junction is heated or cooled depending not only on the direction of the current,

but also the nature of the leaders. This heating or cooling of the junctions is called the Peltier effect and is a linear function

of the current through the connection point. The proportionality factor between the rate of heat absorption or heat generation and the electric current is called the Peltier coefficient and is equal to the amount of heat that is converted when a unit amount of electricity passes through the connection point.

the. The Peltier coefficient can be determined

directly as the product of absolute temperature and the derivative of the Seebeck emf with respect to temperature. Under the aforementioned conditions, this effect is accompanied by a conversion between heat energy and electrical energy that takes place in each individual conductor, and which is called the Thomson effect and is defined as the heat that is absorbed or emitted per unit volume under the influence of an electric current of unit strength that passes a temperature drop equal to one temperature unit. These effects are reversible.

When metallic conductors are joined to form a thermoelectric junction as discussed, they exhibit low Peltier coefficients and low thermoelectric power. Furthermore, such metallic conductors exhibit high thermal and electrical conductivity.

When an external voltage is introduced into the thermoelectric circuit as discussed, the electrical charges move in accordance with the laws of electrical conduction, and as they pass the junctions between different conductors, they cause the absorption or evolution of heat and thus a change in the electrostatic potential. This phenomenon is called the Peltier heat pump effect, disregarding the Thomson effect, c5)a the temperature drops in the thermoelectric devices are quite low under heat pump conditions.

In addition to the above-mentioned reversible thermoelectric effects, there are two other, more general effects that must be taken into account when one wants to understand the operation of a thermoelectric heat pump. These effects are the usual Joule heat or I-R watt consumption (where I denotes current and R resistance) and normal thermal conductivity. In most applications of the Peltier heat pump effect, these effects, which are irreversible, should be made as weak as possible, as they are harmful. Therefore, the specific resistance and thermal conductivity of thermoelectric materials used in heat pumps should be as low as possible.

In fig. 1, the reversible Peltier effect is illustrated for a single thermoelectric device 10, which at A and B is conductively connected with ordinary metallic conductors 11 and 12, which complete a circuit to a direct current source. The metal in the conductors can here and likewise everywhere in the following where not otherwise stated, be copper. The conductivity characteristic of the thermoelectric device 10 is such that when currents of unit strength pass the connection points A and B, P;1 heat units are absorbed at connection point A and P,, heat units are developed at B. The magnitudes of P., and P., depend on the magnitude of the Peltier coefficient between the thermoelectric element 10 and the conductors 11 and 12 and on the temperatures at the respective connection points. If the temperature of the connection points is the same, the heat absorbed at A is equal to that developed at B.

If the initial temperature distribution in the pumping body is constantly equal to T0, the absorption of heat by means of the flow at "A" will gradually lower this connection point's temperature Ta, while the development of heat at "B" gradually raises the temperature Th.

The extent to which Ta and Th differ from Tn depends on the strength of the aforementioned irreversible effects as well as on the magnitude of the present Peltier emf. The energy supplied to the pumping device in the form of I-R watt consumption within the volume of the thermoelectric device causes a slight increase in its temperature. This irreversible heat supply changes the temperature distribution in the thermoelectric device until it is offset by an equally strong heat flow out through the bypass points. The other irreversible effect, heat conduction, causes a flow of heat from the area with the high temperature to the colder parts. This heat flow is denoted Q..

The generation of temperature differences between the junctions and their external surroundings causes heat exchange with them in accordance with the laws of heat conduction. The relevant heat flows shall be denoted Qa and Qh. When the amperage is kept constant and the temperature distribution assumes a stationary state, the total heat flow into connection point A is equal to the Peltier absorption per time unit at connection point A (Pn = Q., + Q,. + Vz I-R)- In a similar way, the heat flow from connection point B is equal to the Peltier emission per time unit at connection point B (Ph = Q>, + Ql; — I-R/2).

As will be clear, the important quantities for the work of the pump are Q.|( Q.,, T;l and Th. Q., expressing the rate at which heat can be extracted from an external source in contact with the cooling compound A. Qh expresses the rate at which heat can be emitted (exhausted) to an external source that is connected to the heat-generating compound B. T,, — T., is the temperature difference by means of which the heat energy is transferred. These parameters will in the following be used when describing the work of heat pumps. The other conditions with regard to phenomena that occur in the pumping body were mentioned only to explain the pump effect and will not be mentioned in the following. It should be noted, however, that the difference between Q;l and Qh is only equal to the electrical power needed to operate the pump unit (Q,,-Q, = EI = F R + I. de). In the latter expression, "de" denotes the differential of the thermal emf (differential thermal emf) which is written from the induced temperature difference

T,, — T.,.

The direction of the heat transfer occurring in the thermoelectric device in fig. 1, is characteristic of materials that have positive polarity of thermoelectric power and conductivity when current flows in the indicated direction. For thermoelectric devices with negative polarity, the heating and cooling of the junctions is exactly the reverse. This is illustrated in fig. 2, where there is shown a thermoelectric device 13 with positive polarity and a thermoelectric device 14 with negative polarity, arranged in series in relation to each other and to a direct current source. The arrow indicates the direction of flow in the conventional way — the electron flow naturally goes in the opposite direction. Consequently, heating occurs at the connection points C and D between the thermoelectric devices 14 and 13 and the electrical conductors 15 and 16 respectively, while cooling occurs at the connection points E and F between the thermoelectric devices 14 and 13 on the one hand and a metallic conductor 17 on the other hand. In the device shown in fig. 2, the heat transfer and the consequent heating of the junctions C and D and the cooling of the junctions E and F act additively, as will be readily understood, thus providing a heat pump with a higher efficiency than one comprising only a single thermoelectric element as shown in fig. 1. The heat pump in fig. 2 is of course only possible thanks to the use of thermoelectric devices 13 and 14 which have Peltier emf and conductivity with opposite signs. The conductors 15, 16 and 17 are of metallic composition and therefore exhibit electrical and thermal properties similar to those discussed above for metals.

On the basis of the above-mentioned parameters, an expression has now been derived which forms a convenient measure for comparing thermoelectric materials and assessing their applicability in heat pumps, and which will henceforth be referred to as the heat pump's "function factor" ("Per-formance Factor”). This expression is jt<2>/ 2K o, where

jt = Peltier coefficient in volts K = Specific thermal conductivity in watts cm-<1>"C-1

o = Specific electrical resistance in Ohm cm, which is therefore given the dimension °C.

This expression is derived directly from an analysis of heat flow in thermoelectric heat pumps. The occurring parameters are determined by conventional techniques and the above expression ("function factor") is a measure of the heat pump's ability to create a temperature difference and transfer heat through this (convect heat through this difference).

The absorption or development of heat at a junction as discussed above is directly proportional to %. The irreversible heat flow down through the temperature drop is proportional to K, and the Joule watt consumption (FR) is proportional to q. The values of n and o are interrelated through the nature of the conduction processes in the compositions that will be described later. From the above expression, it will be noted that the best heat pump will include thermoelectric devices that at the same time have high values of Peltier emf (n) and low values of heat conduction (K) and specific resistance (q) or high specific electrical conductivity.

The function of thermoelectric heat pumps where the invention is used can also be expressed by the heating and cooling that is obtained at the connection points, in relation to the amount of electrical energy required for this heating or cooling. An expression for this factor will be referred to here as "function coefficient". The coefficient of performance for a heat pump (with regard to the cooling connection) is defined as the ratio between absorbed heat at the connection point and the electrical energy consumption in the pump and can be stated as follows:

A heat pump's function coefficient (with regard to the heating connection) is defined as the ratio between the heat emitted at the heating connection point and the electrical energy consumption in the pump and can be expressed as follows:

The function coefficient for heating is equal to the function coefficient for cooling plus 1.

In order to obtain heat pumps with strong heat transfer and high efficiency as mentioned, and comprising one or more thermoelectric devices (as visualized in figures 1 or 2 respectively), thermoelectric devices have been produced with new compositions and with electrical properties in the above-mentioned respects, which with a view to for heat pump purposes are superior to known metal alloys and thermoelectric devices of a semiconductor nature with a known composition.

It has been found that certain alloys or intermetallic compounds exhibit very advantageous relationships between Peltier coefficient and specific resistance and at the same time advantageous values for specific thermal conductivity. Furthermore, it has been found that the values for Peltier coefficient and specific resistance can be changed arbitrarily in an advantageous way to give an extended value range for the function factor without significant changes in specific thermal conductivity. Metals and metal alloys exhibit low values for the Peltier coefficient and specific resistance and, at the same time, high values for specific thermal conductivity. On the other hand, semiconductors exhibit low values of specific thermal conductivity and high values of Peltier coefficient and specific resistance. The function factors for metals are low due to the high values of specific thermal conductivity. The function factors of semiconductors are low due to their high specific resistance. The present invention is based on the fact that a way has been found to change the electrical properties of certain intermetallic compounds (which in their pure state exhibit a high Peltier coefficient and specific resistance and low values for specific thermal conductivity) in order to reduce their specific resistance without change the Peltier coefficient in the same ratio or effect any significant increase in specific thermal conductivity. These intermetallic compounds, which in their pure state are semiconductors, are changed by the addition of a small concentration of beneficial impurity, which gives the semiconductors a somewhat more metallic character, whereby the function factor is improved to a very high degree. The resulting compositions show high values of Peltier coefficient and low values of specific resistance and specific thermal conductivity in combinations that cannot be achieved either with pure metals or with pure semiconductors. These compositions will be called "semi-metallic" alloys or compositions (as distinct from metallic alloys on the one hand and semiconductor alloys on the other). Such semi-metallic alloys or compositions provide, for the first time, heat pumping by Peltier effect with sufficiently high induced temperature differences and heat pumping effects to be used for practical purposes.

The aforementioned semi-metallic alloys or compositions can, according to what has been found, be characterized as binary metallic compounds having a slightly imperfect composition, in that they contain beneficial impurities which condition deviations from perfect stoichiometry as a result of an excess of one of the metals in relation to the other, and/or contains added beneficial pollutants which will hereinafter be referred to as "activators" ("promoters"). Such semi-metallic compositions have semiconductor-like conductivity (both electrical and thermal as mentioned above). Semi-metallic alloys or compositions also include mixtures of such binary metallic compounds, which may be termed ternary metallic alloys or compositions.

Thermoelectric materials which, more particularly, were found to be particularly advantageous for heat pump purposes, consist of lead and at least one component of the group of tellurium, selenium and sulfur in proportions that will be discussed later, and possibly with the addition of beneficial pollutants in the form of activators , which will also be discussed in more detail later. As will be seen, some of these alloys or compositions exhibit negative electrical characteristics, while certain exhibit positive electrical characteristics, whereby, if desired, a compound or dual heat pump assembly as shown in FIG. 2.

Fig. 3 graphically shows a variety of examples of semi-metallic compositions or alloys for thermoelectric heat pump materials comprising lead plus tellurium or selenium. It will be noted that the abscissa of the diagram indicates the different percentages of tellurium and selenium in atom percent, varying linearly from tellurium containing only a "trace" (as detailed later) of selenium, on the left of the drawing, to selenium containing only a "traces" of tellurium, on the right. The ordinate scale on the left (in percent by weight) indicates the amount of lead that can be alloyed with the tellurium, selenium or tellurium-selenium component for any ratio in this, while the ordinate scale on the right conversely indicates the amount by weight of tellurium- , the selenium or selenium-tellurium component for arbitrary quantity ratios for this in the final composition, while the rest is of course lead.

Fig. 3 graphically shows compositions or alloys consisting of lead and either tellurium or selenium or both, as it was discovered that selenium and tellurium, when alloyed with lead within the stated proportions, are mutually soluble within the range of compositions shown, and that tellurium and selenium are interchangeable in providing suitable thermoelectric heat pump materials falling within the above class of binary metal compounds, and because of this mutual solubility such binary compounds of the indicated constituents can be mixed, as is also contemplated graphically, to form the ternary alloys or compositions also mentioned above. The discovery of this mutual solubility of selenium and tellurium is important for the manufacture of thermoelectric devices of the above-mentioned nature, as it eliminates the necessity of separating selenium from tellurium and vice versa (these two constituents invariably occur together as contaminants of each other ), which denotes a difficult and expensive process and is actually impossible to carry out completely. Consequently, even the extremes for the compositions illustrated in fig. 3, i.e. the composition at the far left of the scale (35.0 to 38.05% by weight tellurium, remainder mainly only lead), and the composition at the other end of the scale (25.0 to 27.55% by weight selenium, remainder mainly only lead ) are considered to contain at least a "trace" of selenium and tellurium respectively. When, in the following, in the presentation and the drawing, "traces" are mentioned, this expression must therefore be understood to mean such small amounts of the relevant constituents and/or contaminants that they are difficult to detect, but must be assumed to be present because it is impossible to achieve absolute purity.

As it is again referred to, e.g. to fig. 3, it will be noted that a thermoelectric organ of lead. selenium and tellurium which exhibit the desired properties, as mentioned could consist of a selenium-tellurium component in which the selenium only occurs as traces. If this is the case, this component must make up 35.0 to 38.05 percent by weight of the composition, while the rest (65.0 to 61.95 percent by weight) is lead. At the other extreme, where the selenium-tellurium component consists almost entirely of selenium, with only a trace of tellurium, the component shall comprise from 25.0 to

27.55% by weight of the final composition, while the rest (from 75.0 to 72.45% by weight) is lead. These extreme compositions can, for the purposes of the present presentation, be termed "limit" compositions and can be regarded as binary metallic compounds of slightly imperfect composition, as will be seen later.

As a further example, mention can be made of the case where selenium and tellurium occur in equal amounts (in atomic percentage) in the selenium-tellurium component. If this is the case, this should make up from 30.0 to 32.8 percent by weight of the composition, while the rest (70.0 to 67.2 percent by weight) is lead. This composition and all other compositions which are visualized in fig. 3 between the boundary compositions mentioned, are mixtures (in different proportions) between the boundary compositions, and as they contain both selenium and tellurium as well as lead, they form ternary alloys or compositions.

Reference is now made to fig. 4, which further illustrates a variety of examples of semi-metallic compositions or alloys for thermoelectric heat pump devices, comprising lead plus selenium and/or sulfur. As in the case of fig. 3 represents the abscissa of fig. 4 the different process proportions of selenium and sulphur, calculated in atomic percentage and listed on a linear scale, from the extreme selenium containing only a trace, as defined above, of sulphur, on the left of the drawing, to the extreme sulfur containing only a trace of selenium, right. The ordinate scale on the left (in weight percent) indicates the amount of lead that can be alloyed with the selenium, sulfur or selenium-sulfur component for arbitrary percentages thereof, while the scale on the right conversely indicates the weight percent amounts of sulfur, selenium or selenium-sulphur the constituent for arbitrary percentages of this in the final composition, while the remainder is of course lead.

Fig. 4 further graphically illustrates compositions or alloys comprising lead and either selenium or sulfur or both, it being found that selenium and sulphur, when alloyed with lead within the given proportions, are mutually soluble over the entire shown range of compositions, and that selenium and sulfur are interchangeable for the purpose of providing thermoelectric heat pump materials falling within the above class of binary metal compounds, and because of this mutual solubility these binary compounds of the stated constituents can be mixed, likewise as shown, to form ternary alloys or compositions, likewise as mentioned above. The discovery of this (possibility of replacing selenium and sulfur with each other) is important for a salary-like production of thermoelectric bodies of the above-mentioned kind, as it makes it unnecessary to separate selenium from sulfur and vice versa (these two components invariably occur together as impurities into each other), which is a difficult and expensive process and is actually not possible to carry out completely. Consequently, even the extremes of the compositions shown in Fig. 4, at the left end of the scale, can

(25.0 to 27.55 weight percent selenium, rest lead)

and at the opposite end of the scale (12.80 to 13.37 weight percent sulphur, the remainder mainly just lead), are considered to contain at least a trace of sulfur and selenium respectively.

Referring again to fig. 4 as an example, it will be noted that a thermoelectric organ of lead, selenium and sulfur which gives the desired properties, as previously mentioned, could consist of a selenium-sulphur component in which the sulfur only occurs as a trace. If this is the case, this component should make up from 25.0 to 27.55 percent by weight of the composition, while the rest (75.0 to 72.45 percent by weight) is lead. At the other extreme, where the selenium-sulfur component consists almost exclusively of sulphur, with only traces of selenium, the component should comprise from 12.80 to 13.37 percent by weight of the final composition, while the remainder (87.20 to 86.63 percent by weight ) is lead. These extreme compositions can also, for the purposes of the present presentation, be termed "borderline" compositions and can be regarded as binary metal compounds with a slightly imperfect composition, as will be seen later.

As a further example, mention should be made of the case where the amounts of selenium and sulfur calculated in atomic percent in the selenium-sulphur component are equal. In this case, this should make up from 18.90 to 20.46 percent by weight of the composition, while the rest (81.10 to 79.54 percent by weight) is lead. Such a composition and all other compositions which are visualized in fig. 4 between the mentioned boundary compositions, are mixtures (in different proportions) of the boundary compositions, and as they contain both selenium and sulfur together with lead, they form ternary alloys or compositions.

It was also discovered that the above-mentioned boundary compositions, which practically exclusively consist of lead and tellurium, can also to a limited extent contain some sulfur (i.e. to the extent that sulfur is found as a contaminant in commercially available tellurium) and likewise selenium. In a similar way, the above-mentioned boundary compositions which mainly consist of lead and selenium may, to a limited extent, also contain tellurium and sulfur as appropriate, just as the above-mentioned boundary compositions which mainly consist of lead and sulphur, also to a limited extent may contain tellurium as mentioned, likewise the harness. Likewise, any of the intermediate or ternary alloys or compositions consisting of lead and at least two constituents of the tellurium, selenium, sulfur group may contain such limited amounts of the third constituent of the group.

The above-mentioned quantity ratios and proportions for the various components which are visualized graphically in fig. 3 and 4, must be considered critical if the compositions shown are to have the aforementioned electrical properties that are desired for heat pumps. The minimum limits for the lead component in the composition according to the invention (graphically visualized by the lower curve in fig.

3 and 4) must be considered critical, then there if

the lead component is significantly less than the shown minimum value for an arbitrary combination of quantity ratios of the components, desirable values of Peltier emf and specific resistance will not appear, and the electrical and mechanical properties will not be reproducible. On the other hand, it has been found that the resulting composition, if the lead content for an arbitrary composition significantly exceeds the maximum limit (graphically illustrated by the upper curve in Figs. 3 and 4), acquires a too metallic character to exhibit the electrical properties that satisfy the purpose of the invention.

Not only the quantities and areas described above must be considered critical, but also the purity. More particularly, the limit for tolerable metallic contamination in the final composition has been found to be of the order of 0.01%, and the composition must be substantially free of oxygen, if the desired mechanical and electrical properties are to be achieved and reproducible. Such purity can be achieved by using lead, tellurium, selenium and sulfur which do not contain metallic impurities exceeding the order of magnitude 0.01%. Or you can also use starting ingredients of lower purity if you let the formed composition undergo a crystallization step, which will be described later, to obtain a final or finished composition with a purity of the specified order of magnitude. Accordingly, the starting ingredients or, in any case, the final composition, when they are referred to in this presentation with claims, are to be understood to have a purity of the specified order of magnitude. Various impurities which are usually found in commercial grades of all four components will reduce the Peltier emf which compositions according to the present invention exhibit, and must therefore be essentially removed by purification. Thus, copper is an example of a pollutant with a destructive effect.

Compositions of lead and at least one of the substances tellurium, selenium and sulfur as described above and exhibiting the desired electrical properties mentioned above can be produced in the following way: The starting components, which are assumed to be free of metallic impurities as mentioned above and preferably in the reduced state, are mixed with each other in the indicated proportions and enclosed in a tube or container, preferably of quartz or Vycor, which has been previously evacuated. The tube with contents is then heated to the contents' melting point, which is at a temperature from approx. 920° C for the boundary compositions shown graphically on the far left of fig. 3, to 1085° C for the limit compositions shown on the far right in fig. 3 or far left in fig. 4, and further to 1115° C for the limit compositions shown as extremes on the right in fig. 4. The particular temperature for an arbitrarily given composition which is visualized in fig. 3 and 4, depends on the melting point of the composition in question, which experts will be able to derive without further ado from the melting points indicated for example for the boundary compositions. During the heating, the molten mass is preferably stirred to ensure good mixing, after which it is cooled.

After the composition has been mixed as mentioned above, the solidified piece can be removed from the tube and cast into molds of graphite or the like under an atmosphere of inert gas. More particularly, it has been found expedient during casting to cover the mold with an inert gas such as argon or carbon dioxide under positive pressure. This gas counteracts evaporation of the molten composition and thus porosity in the casting.

The alloying and casting processes described above should be carried out in crucibles that do not react with or contaminate the composition, as rather small amounts of harmful contaminants can have a very harmful influence on the body's electrical and/or physical properties. It has been found that suitable crucibles can be made from coal, alundum (alumina), burnt lavaite (lavite) and "Vycor" (borosilicate glass containing approx. 96% bound silicic acid) or quartz.

After casting, the workpiece can be machined if necessary. The shaped blanks are then preferably annealed in a reducing atmosphere at 540-815° C for 20 to 10 hours. This tempering treatment ensures homogeneity of the blanks and improves their electrical and physical properties.

An alternative method of preparing the compositions consists in melting the aforementioned ingredients in an open crucible under an atmosphere of argon or any inert and/or reducing gas. As the vapor pressure of selenium and sulfur is relatively high, one may be exposed to a partial loss of selenium or sulphur, and adjustments must then be made to the original amount of the component in question, when it is used, to take this into account loss. In all other respects, this method is similar to the one previously described.

Thermoelectric heat pump bodies comprising alloys or compositions according to the invention can, as previously mentioned, be obtained by using starting components of such purity that the compositions and the resulting alloy do not contain greater contamination than previously mentioned. Alternatively, an alloy with such purity can be obtained by, as described below, making the composition from less pure starting components and reducing the content of impurities by recrystallization from the smelting. As will be understood by those skilled in the art, this process consists of melting the impure composition and causing it to solidify slowly and progressively from one end of the melt to the other. This leads to a concentration of the impurities of the starting components in the area of the last place of the melt which solidifies, after which this lot can be discarded and the process, if further purification is necessary, can be repeated. When this cleaning method is used, a small additional amount of lead must be added sufficient to bring the resulting composition within the above limits.

The used alloys of lead and at least one component of the group tellurium, selenium, sulfur can best be described metallographically as two-phase alloys. When cutting through and microscopically examining these two-phase alloys, it has been observed that they comprise a main phase comprising crystal grains, usually varying from 1 to 10 mm in size, and where between these grains there are thin, darker areas of another phase . The grains in the primary phase are crystals of the intermetallic compounds lead-telluride, lead-selenide and lead-sulphide (or mixed crystals thereof), which respectively contain approx. 61.89, 72.41 and 86.60 weight percent lead. The darker secondary phase, which can be clearly observed at the grain boundaries, is lead which contains a very small concentration of selenium, tellurium or sulphur. The effect of the secondary phase in these alloys is assumed to be threefold. Firstly, the thermal equilibrium between the two phases established by the aforementioned heat treatment causes negative Peltier emf and conductivity in the primary lead-telluride, lead-selenide or lead-sulphide phase, which due to its high concentration in the alloy is decisive for the two-phase alloy's electrical properties. Secondly, the thin layers act as a sealant for the primary phase grains and thereby improve the mechanical strength of the alloy compared to the pure inter-metallic compound. Thirdly, this putty effect of the secondary phase causes a good electrical conductivity in the polycrystal-line alloy by reducing the intergranular component of the electrical resistance to negligible size. It has been found that the real concentration, the secondary phase, is not critical as long as the composition is kept within the ranges indicated above.

With regard to the specified ranges for the various above-mentioned compositions, it will be noted that in each case there is an excess of lead beyond the amount needed to satisfy the stoichiometric quantity ratios in the compound formed with the other constituent(s), i.e. tellurium, selenium or sulphur. If you take e.g. the limit compositions indicated, it will be noted that the first limit composition, which essentially consists of lead and tellurium, contains from 61.95-65.0 weight percent lead, while the stoichiometric quantity ratio is 61.89% lead, 38.11 % tellurium. For the basic composition 61.95% lead, 38.05% tellurium, the amount of lead that stoichiometrically corresponds to the contained amount of tellurium is thus equal to

There is thus an excess of lead, calculated in relation to the stoichiometric composition, equal to

Calculated in a similar way for 65% lead, 25% tellurium, an excess of 8.9% by weight is obtained in relation to the stoichiometrically necessary amount. In a similar way, the boundary composition which essentially consists of lead and selenium contains from 0.15-10.4 weight percent lead, calculated in the same way, beyond the amount of 72.41 weight percent lead required stoichiometrically for connection with selenium. The same of course applies with regard to the limit composition which mainly consists of lead and sulphur, where the amount of lead specified for the composition according to the invention is from 0.23-4.7% by weight in excess of the amount required for stoichiometric connection with the corresponding to sulphur, namely 86.0 weight percent lead. In a similar way, for any of the compositions that lie between the aforementioned basic compositions in the areas indicated on fig. 3 and 4, an excess of lead beyond the stoichiometrically necessary for connection with the tellurium-selenium or selenium-sulphur component, and the excess in weight percentage varies correspondingly to the location of the intermediate composition in relation to the extremes.

The above-mentioned excess of lead which occurs in all the specified compositions within the areas shown graphically in fig. 3 and 4, can be described as a contamination in relation to the primary lead-telluride, lead-selenide or lead sulphide phase. However, these contaminants must be distinguished from the harmful contaminants discussed above in connection with the requirements for the purity of the compositions or alloys. Such intentional excess amounts of lead occurring in the compositions will therefore be termed "beneficial impurities" in the present presentation with claims. As mentioned above, it is the presence of such beneficial impurities which gives the compositions according to the invention their desired electrical properties and distinguishes the semi-metallic alloys or compositions according to the invention over the intermetallic compounds lead-telluride, lead-selenide and lead-sulphide which occurs in the primary phase. More particularly, these beneficial impurities impart to the inter-metallic compounds, whether these are binary compounds of boundary composition or ternary compounds of intermediate composition, a lower specific resistance without suffering the same loss of Peltier emf or with respect to low thermal conductivity , and leads to compositions with what can be described as semiconductor-like electrical conductivity resp. specific conductivity. It should be noted that said excess in any of the described compositions introduces a Peltier emf of negative polarity as well as negative conductivity and thus provides a thermoelectric heat pump device which can be advantageously used e.g. as a simple thermoelectric device in the heat pump in fig. 1 or in combination with a thermoelectric heat pump device of opposite polarity in the double or combined heat pump of fig. 2.

Furthermore, it was discovered that the electrical properties which are desirable in thermoelectric devices for use in heat pumps and consisting of lead-tellurium, lead-selenium, lead-tellurium-selenium, lead-sulfur or lead-selenium-sulfur within the indicated ranges with regard to composition and purity (hereinafter for simplicity's sake termed "basic compositions"), can be changed perceptibly in an advantageous and reproducible way by adding measured amounts of substances other than the constituents of the basic compositions. These additions can also be described as "beneficial pollutants" in contrast. from harmful pollutants as discussed above. For the sake of simplicity, these additives will hereafter be referred to as "activators", since, as will be seen later, they tend to enhance the electrical properties desired in thermoelectric heat pump devices of the base composition. Since the quantities of these activators, expressed as a percentage by weight of the base composition, as will also be apparent later, are very small when these activators are to be as effective as possible and condition reproducibility of the electrical properties, the base composition to which they are added should still have higher purity, i.e. contain even less harmful and uncontrolled contaminants. As a practical rule, it has been found that the basic composition should have such purity in this respect that it does not contain more than 0.001 weight percent of unwanted contamination. Similarly, when these activators are added, the lead content of the base composition should be slightly less, e.g. a maximum of 63.0, 73.5 and 87.10 percent by weight for the respective limit compositions.

As previously noted, all of the above base compositions exhibit negative Peltier emf and negative electrical conductivity. By adding the activators that will be described later, these negative properties can be enhanced by using certain activators, while by adding certain other activators it is possible to change the polarity of the basic composition's electrical properties. Therefore, certain activators will be termed "positive activators" and certain others will be termed "negative activators", as further defined later, and the resulting alloy or composition may be a "positive" or "negative" alloy or composition, likewise according to a definition which will be specified in more detail.

By "negative" compositions or alloys, in the present presentation with claims, an alloy or composition that exhibits negative conductivity, which can be demonstrated by measurements of the Hall effect or thermoelectric effect, both at room temperature, shall be understood. In a similar way, "positive" compositions or alloys shall be understood to mean an alloy or composition that exhibits positive conductivity as demonstrated by measurements of Hall effect or thermoelectric effect, both at room temperature.

"Negative activators" are those which, when added to the above-mentioned base alloys, change their electrical conductivity without changing the polarity of the base alloy's conductivity or Peltier emf (which is negative according to the above definition). "Positive activators" are those which, when added in quantities to be specified to the above base alloys or certain base alloys which will be discussed later, act to influence the electrical conductivity in a positive direction as defined above, and only cause a small simultaneous reduction of

the activated base alloy thermoelectric power. Certain of these "positive activators", when added to the above basic alloys, first - with very small amounts of addition - cause a reduction in the alloy's conductivity to a

minimum value, after which a further increase

of the concentration of "positive activators" causes an increase in the conductivity of the alloy, accompanied by a shift in the polarity of conductivity and Peltier emf, i.e. from negative to positive.

The difference in effect between these negative and positive activators shall, for the sake of clarity, be explained as follows: 1. Increasing concentrations of the negative activators causes an increase in conductivity and a decrease in Peltier emf in the resulting alloy in relation to the base alloy, while the negative polarity of its conductivity and Peltier emf are maintained. 2. Increasing concentrations of the positive activators initially causes a decrease in conductivity and an increase in Peltier emf in relation to the base alloy until a minimum of conductivity is reached, after which Peltier emf and conductivity change polarity to a positive sign and so on increase in the concentrations of the positive activators causes increase in conductivity and decrease in Peltier emf of the resulting alloy.

The activators which have been found to be effective for the purposes of the present invention when added in very small amounts to the above-mentioned basic compositions will, for the sake of simplicity, be discussed in relation to the limit compositions modified with regard to purity and lead content as described earlier, as it will be understood that these activators can also be added to any of the intermediate compositions with advantageous results. In that case, the activators should be adapted both in terms of type and amount to the relative concentrations of the limit compositions in the intermediate composition, which includes a mixture of these.

The table below indicates in the first column the additives which have been found to be effective as negative activators when added to the above-mentioned base alloys or compositions of lead and tellurium. The second column in table I indicates the order of magnitude of the maximum limits for the concentration of these activators, calculated as a percentage by weight of the base alloys which are effective for achieving the object of the invention. It will be understood that these concentration limits denote the maximum for effective change of the base alloy's electrical properties. Concentrations in excess of the specified amounts of the additives have no tangible effect with regard to beneficial changes in the electrical properties the invention is concerned with, and in this sense are

the specified limits to be considered critical.

The third and fourth columns of Table I indicate

the electrical properties, measured at room temperature, of lead-tellurium alloys activated with the maximum useful concentrations of the negative activators according to

high temperature tempering as described earlier.

As previously mentioned, certain positive activators can also be alloyed with the above-mentioned lead-tellurium base alloys, and these activators are listed in the first column of Table II. Another column in table II indicates, like the corresponding column in table I, the order of magnitude of the maximum concentration limits of the relevant activators calculated in weight percent of the base alloys that are effective for achieving the invention's purpose. It will again be noted that concentrations of the positive activators, calculated on the lead-tellurium base alloys, in amounts beyond those appearing in the second column of Table II, have no appreciable effect with regard to

to effect some advantageous change in the electrical properties with which the invention is concerned, and in this sense the specified limits are to be considered critical.

The third column in Table II indicates in weight percent the minimum concentration of the enumerated positive activators which provide useful heat pump organs with positive polarity.

The fourth and fifth columns respectively indicate the Peltier emf and specific resistance at room temperature for the alloy or composition that results from the addition of the mentioned positive activators in quantities as indicated in the second column, after tempering at high temperature as described above.

As mentioned earlier, the above-mentioned lead-tellurium base alloy is a two-phase alloy. When the aforementioned activator additives are introduced into the base alloy, they are distributed between the two phases. It has been discovered that the nature of this distribution has no significant effect on the electrical properties of the composition in all cases, except for bismuth, thallium and arsenic. Consequently, the maximum effective concentration in the case of bismuth, thallium and arsenic depends on the lead content of the lead-tellurium base alloy within the ranges indicated for it in Fig. 3. It was found that 1.20 weight percent bismuth is the maximum effective concentration for lead-tellurium base alloys containing 63.0% lead. For base alloys containing less lead, the maximum effective bismuth concentration is somewhat smaller and drops to 0.60% by weight when the lead content drops to 61.95%. Similarly, the maximum effective concentration in the case of thallium depends on the lead content of the lead-tellurium base alloy within the range specified for it. It was found that 1.00 weight percent thallium is the maximum effective concentration for lead-tellurium base alloys containing 63.0% lead. For base alloys containing less lead, the maximum effective thallium concentration is somewhat smaller and decreases to 0.25% by weight when the lead content decreases to 61.95%■ Furthermore, the maximum effective concentration in the case of arsenic is similarly dependent on the lead content of lead -the tellurium base alloy within the range specified for it, and varies from 0.25% for base alloys containing 63.0% lead down to 0.07% for base alloys containing 61.95% lead. As shown in Table II, the minimum effective concentration for heat pump bodies in the case of thallium-activated base alloys ranges from 0.01 to 0.04 weight percent when the lead content of the lead-tellurium base alloy ranges from 61.95 to 63.0 weight percent. Similarly, the minimum effective concentration in the case of arsenic activated base alloy ranges from 0.002 to 0.006% when the base alloy lead content ranges from 61.95 to 63.0%.

This relationship in connection with bismuth, thallium and arsenic is believed to be due to the formation

respectively of a bismuth-lead-tellurium, a thallium-lead-tellurium and an arsenic-lead-tellurium complex in the aforementioned intergranular phase which

occupies part of the addition. All the other additions of a third element, both positive and negative, form complexes with the aforementioned secondary or intergranular phase to a much lesser extent than is the case with bismuth, thallium and arsenic, and for the purposes of the present invention the effects of the complexes of the mentioned other additions in this phase without consequences. Consequently, no changes are needed in the limit concentrations for these additions when the proportions of lead and tellurium in

the base alloy varies within the specified range.

In the above tables I and II, the values for the Peltier emf and specific resistance in both cases are given for a concentration of 61.95% lead, the rest mainly only tellurium and the relevant activating additive in the maximum quantity indicated in the tables (in the case of bismuth, thallium and arsenic for the lowest stated value of maximum effective amount).

The table III below lists in the first column the additives which have been found to be effective as negative activators when added to the above lead-selenium base alloys or compositions. The second column in table III indicates in percentage by weight the maximum limit concentrations of such activators in the base alloys which are effective for achieving the object of the invention. It will be understood that these concentration limits are the highest concentrations that effectively change the electrical properties of the base alloy. Concentrations beyond the specified amounts of the additives have no tangible effect in the form of any beneficial change in the electrical properties the invention is concerned with, and in this sense the specified limits are to be considered critical. The third and fourth columns of Table III indicate the electrical properties, measured at room temperature, of lead-selenium alloys activated with the maximum useful concentrations of the negative activators after high-temperature tempering as described earlier.

As previously mentioned, certain positive activators can also be alloyed with the above-mentioned lead-selenium base alloys, and these activators are listed in the first column of Table IV. The second column of table IV indicates, like the corresponding column in table III, the order of magnitude, in weight percent, of the maximum concentration limits for the activators in the base alloys, which are effective for achieving the object of the invention. Again, it will be noted that concentrations of the positive activators in the lead-selenium base alloys in amounts beyond those indicated in the second column of Table IV have no appreciable effect in the form of any beneficial change in the electrical properties with which the invention is concerned, and in this sense the stated limits are to be considered critical.

The third column in Table IV indicates in weight percent the maximum concentrations of the enumerated positive activators which give useful heat pump bodies with positive polarity.

The fourth and fifth columns indicate the Peltier emf and specific resistance at room temperature for the alloy or composition obtained by adding the indicated positive activators in an amount as indicated in the second column, after high-temperature annealing as previously described.

As mentioned above, the previously described lead-selenium base alloy is a two-phase alloy. When the aforementioned activator additives are introduced into the base alloy, they are distributed between the two phases. It was discovered that the nature of this distribution has no effect on the electrical properties of the composition in all cases, except for bismuth and antimony. Consequently, the maximum effective concentration in the case of bismuth and antimony depends on the lead content of the lead-selenium base alloy within the ranges indicated for this in fig. 3. It was found that 2.5 weight percent bismuth is the maximum effective concentration for lead-selenium base alloys containing 73.5% lead. For base alloys containing less lead, the maximum effective bismuth concentration is somewhat smaller, varying down to 0.40% by weight when the lead content drops to 72.45%. Similarly, the maximum effective concentration in the case of antimony depends on the lead content of the lead-selenium base alloy within the range indicated for it. It was found that 1.5% by weight antimony is the maximum effective concentration for lead-selenium base alloys containing 73.5% lead. For base alloys containing less lead, the maximum effective antimony concentration is somewhat smaller, varying down to 0.20% by weight when the lead content drops to 72.45%.

This relationship in connection with bismuth and antimony is believed to be due to the fact that in the above-mentioned intergranular phase a bismuth-lead-selenium-resp. an antimony-lead-selenium complex that occupies part of the addition. All the other above-mentioned substances which can be added as a third element, both the positive and the negative, form complexes with the secondary or intergranular phase to a much lesser extent than is the case with bismuth and antimony, and for the purposes of the present invention these are the effects of the other additions mentioned without consequences. Consequently, there is no need to change their concentration limits when the percentages of lead and selenium in the base alloy vary within the range specified for them.

In the above tables III and IV, the values for the Peltier emf and specific resistance in both cases are given for the composition 72.45% lead, the rest mainly only selenium, however with the relevant activator addition in the maximum effective amount indicated in the tables ( in the case of bismuth and antimony the lowest stated maximum effective amount).

The table V below lists in the first column the additives which have been found to be effective as negative activators for the above-mentioned basic alloys or compositions of lead and sulphur. The second column in table V indicates the order of magnitude of the maximum concentration limits, calculated in weight percent, of the activators in the base alloys which are effective for achieving the object of the invention. It will be understood that these concentration limits denote the highest concentrations which effectively change the electrical properties of the base alloy. Concentrations beyond the specified amounts of the additives have no tangible effect with regard to beneficial changes in the electrical properties the invention is concerned with, and in this sense the specified limits are to be considered critical. The third and fourth columns of Table V indicate the electrical properties, measured at room temperature, of lead-sulphur alloys activated with the maximum useful concentrations of the negative activators after high-temperature tempering as described earlier.

As previously mentioned, certain positive activators can also be alloyed with the above-mentioned lead-sulphur base alloys, and these activators are listed in the first column of table VI. Another column in table VI indicates, like the corresponding column in table V, the order of magnitude of the maximum concentration limits, calculated in weight percent, of the activators in the base alloy which are effective for achieving the object of the invention. It should again be noted that concentrations of the positive activators in the lead-sulfur base alloys in amounts beyond those indicated in the second column of Table VI have no appreciable effect with respect to beneficial change of | • the electrical properties the invention deals with, and in this sense the stated limits are to be considered critical.

The third column in Table VI indicates in weight percent the minimal concentration of the specified positive activator which provides useful heat pump materials with positive polarity.

The fourth and fifth columns indicate the Peltier emf and specific resistance at room temperature for the alloy or composition that results from the addition of the mentioned positive activator in the quantity indicated in the second column, after high-temperature tempering as described above.

As mentioned, the previously described lead-sulphur base alloy is a two-phase alloy. When the described activator additives are introduced into the base alloy, they are distributed between the two phases. It has been found that the nature of this distribution has no significant effect on the electrical properties of the composition in all cases, except for bismuth and antimony. Consequently, the maximum effective concentration in the case of bismuth and antimony depends on the lead content of the lead-sulfur base alloy within the ranges indicated for this in : fig. 3. It has been found that 3.0 weight percent bismuth is the maximum effective concentration for lead-sulfur base alloys containing 87.10% lead. For base alloys with a lower lead content, the maximum effective bismuth concentration is somewhat lower, as it varies down to 1.0% by weight when the lead content drops to 86.63%. Similarly, the maximum effective concentration in the case of antimony also depends on the lead content of the lead-sulphur base alloy within the range indicated for this. It has been found that 3.0 weight percent antimony is the maximum effective concentration for lead-sulfur base alloys containing 87.10% lead. For base alloys containing less lead, the maximum effective antimony concentration is somewhat smaller, varying down to 0.50% by weight as the lead content drops to 86.63%.

This relationship in connection with bismuth and antimony is again believed to be due to the fact that in the aforementioned intergranular phase a bismuth-lead-sulphur complex resp. an antimony-lead-sulphur complex, which occupies part of the addition. All the other substances mentioned as a third addition, both positive and negative, form complexes with the mentioned secondary or intergranular phase to a much lesser extent than is the case with bismuth and antimony, and for the purposes of the present invention these effects are the aforementioned other appointments without consequences. Consequently, there is no need to change their concentration limits when the proportions of lead and sulfur in the base alloy change within the range specified for these.

In Tables V and VI above, the values for the Peltier emf and specific resistance are in both cases given for the composition containing 86.63% lead, the rest mainly only sulphur, subject to an addition of the relevant activator in the maximum effective amount stated in the tables (in the case of bismuth and antimony the lowest stated maximum effective amount).

From the foregoing description, it will be seen that the maximum beneficial contaminant concentrations for activated boundary compositions range from 4.22 weight percent (ie, 3.02% lead excess plus 1.20% maximum for any activator) in the case of the boundary composition lead -tellurium, to 6.6% by weight (ie 4.1% lead excess plus 2.5% maximum for any activator) in the case of the limit composition lead-selenium, and further to 6.9% by weight (ie

3.9% lead excess plus 3.0% maximum for any activator) in the case of the lead-sulphur boundary composition. Thus, the maximum amount of beneficial impurity for any activated composition is 6.9% by weight; while the maximum amount of beneficial impurity for any unactivated composition, as previously described, is 10.4% by weight. Accordingly, it can be said that the maximum amount of beneficial contamination for any of the aforementioned compositions around the rain is about 10 percent by weight.

Typical data for the mentioned compositions are illustrated, for example, in fig. 5 and 5A, where the Peltier emf in volts (left scale) and in watts per cm (right

Q

scale) are listed as a function of specific resistance in Ohms. cm, given on a logarithmic scale.

In fig. 5, the curves G, H and I apply to compositions consisting mainly of lead + tellurium, lead + selenium and lead + sulfur respectively within the specified ranges, and graphically illustrate how the mutual relationship between Peltier coefficient (jt) and specific resistance (p) varies with the addition of negative activators to the respective basic compositions (the right end of the curves represents the basic compositions). In a similar way, the curves J, K and L apply to compositions consisting mainly of lead + tellurium, lead + selenium and lead + sulfur respectively within the specified ranges and graphically illustrate variations in the quotient when negative activators are added to the basic composition (right end of the curves still represent the basic compositions).

In fig. 5A applies to the curves M, N and O for compositions essentially consisting of respectively lead + tellurium, lead + selenium and lead + sulfur within the specified ranges, and illustrates graphically how the relationship between Peltier coefficient (jt) and specific resistance (q) varies at addition of positive activators to the respective basic compositions (the right end of the curves represents the minimally effective concentration of the mentioned positive activators). In a similar way, the curves P, Q and R apply to compositions essentially consisting of lead + tellurium, lead + selenium and lead + sulfur respectively within the specified ranges and graphically illustrate how the quotient varies

Q

by adding positive activators to the basic composition (the right end of these curves always represents the minimally effective concentration of the mentioned positive activators).

In fig. 5 and 5A show the curves J, K, L and P maximum values for the quotient

P Jt2

Q

The curves Q and R indicate the extreme values at the maximum concentration of positive activator. The specific resistance at the right end of fig. 5 and 5A (0.1 Ohm. cm) are characteristic of semiconductor materials. The value for specific resistance on the far left of fig. 5 and 5A (0.0001 Ohm. cm) is characteristic of a metal. The above-mentioned maxima occur at values of Peltier coefficient and specific resistance that lie between those that apply to metals and to semiconductors. Furthermore, semi-metallic materials according to the present invention show low thermal conductivity which approaches the thermal conductivity of a semiconductor. Simultaneously occurring high values of the quotient and low values for heat-o

the conductivity (k) provides the high function factors desired for heat pump materials.

Further illustrative data for some of the mentioned compositions is shown as an example in fig. 6, where the left scale expresses the function coefficient (cooling)

QQ

—£ = ?s i!—-. The right scale indicates the values EI Qb — Q.,

of Q, (pumping rate — "pumping rate") in watts. The abscissa is in both cases the current density in amp/cm-. In fig. 6, curve S is a representative function-coefficient curve for compositions essentially consisting of lead and tellurium, while curve T is a

representative curve for the pump effect (Q;1) for the same compositions. Similarly, curve U is a representative coefficient of performance curve for compositions consisting essentially of lead and sulphur, while curve V is a representative pump-effect curve (Q.,) for the same compositions. From fig. 5 and 5A, one will easily see that representative curves for compositions essentially consisting of lead and selenium will fall somewhat within the curves in fig. 6 for the compositions of lead and tellurium and of lead and sulphur.

As will be apparent to those skilled in the art, all of the semi-metallic compositions described as examples of thermoelectric heat pump materials comprising binary metallic compounds and ternary compositions or alloys thereof with semiconductor-like conductivities exhibit data sufficiently high for practical application in heat pumps. As an example, the values in Table VII below are illustrative of these improved heat pump properties.

In table VII, the first column lists various examples of compositions of the type described with an indication of composition in percentage by weight. The second column indicates in Volts the Peltier emf that can be obtained with each of these compositions. The third column indicates for the respective compositions the thermal conductivity in watts per cm. per degrees Celsius. Fourth column indicates the specific resistance of the compositions in Ohm. cm. The fifth column indicates the function factor for the respective compositions. It will be noted that the selected examples of compositions include examples of each of the base alloys and at least one example of each of these with one added activator.

A comparison of the values in the function factor column in Table VII shows that the activated alloys exhibit better pumping properties than the base alloys. The legitimacy of this conclusion is confirmed by the table VIII below, which shows as an example real pump data for a bismuth-activated lead-tellurium alloy.

The values in Table VIII show that the function coefficient and the heat pump effect increase with increasing bismuth concentration. Similar effects can be observed in connection with other base alloys and with other activators for a given base alloy. It will also be noted that for a given basic composition and current density there is a relationship between induced temperature difference and heat pump effect. When one rises, the other falls. In the area shown, the induced temperature difference and the heat pump effect increase with increasing current density, while the function coefficient decreases. As it is generally desirable for most of these parameters to be as high as possible, there is no special optimum current density. Instead of the current density, one should choose a parameter according to the application in question.

Table VIII indicates data typical of all the compositions described above and is equally valid for positive and negative polarity thermoelectric heat pump materials. The following Table IX provides an example of heat pump data for a particular positive polarity thermoelectric heat pump body.

In the case of a heat pump with a single heat pump body of the type shown schematically in fig. 1, a temperature drop in the cold connection point of 25° C and a temperature increase in the hot connection point of at least 100° C have been achieved. In a similar way, as indicated in tables VIII and IX, a function coefficient for cooling as high as 17. In fact, considering fig. 6, it is noted that the function coefficient, when the current density approaches lower values, rises steeply as a sign that even higher function coefficients can be obtained at lower current densities. As mentioned, similar function coefficients can be obtained at the hot connection point, as the function coefficient for heating is equal to the function coefficient for cooling plus 1. Furthermore, it will also appear from fig. 6 that with pump powers approaching 0.2 watts for a single unit with a diameter of 6.3 mm (Vi"), a current density above 25 amp/cm can be achieved. As the pump power rises with decreasing function coefficient, the current density for practical use should be chosen so that it simultaneously gives the highest possible values for function coefficient and pump power, depending on the nature of the application.

It was further found that further

suitable positive thermoelectric materials which exhibit positive electrical characteristics as stated above and can be advantageously used for heat pump purposes may include lead and tellurium, in which there is an excess of tellurium over the amount required to satisfy the stoichiometric quantity ratios in the lead-telluride compound. Such additional alloys or compositions may contain added beneficial pollutants, and as before, these will also be referred to in the following as "positive activators". These additional alloys or compositions may also be characterized as binary metal compounds having oxygen imperfect composition due to the excess of tellurium over lead and/or containing added beneficial contaminants. These additional alloys or compositions may be termed semi-metallic alloys or compositions and have semiconductor-like conductivity (both electrical and thermal) similar to the binary and ternary alloys already discussed.

More particularly, the following materials were found particularly advantageous for thermoelectric bodies with positive electrical characteristics: Base alloys or compositions consisting essentially of lead and tellurium, in which lead is present in the range from 58.0 to 61.8 weight percent and the remainder in the range from 42.0 to 38.2 percent by weight is tellurium.

The latter quantity ratios for the components lead and tellurium must be considered critical, if the compositions are to have the electrical properties desired in heat pumps.

As before, not only the latter area for the tellurium-rich lead-tellurium basic compositions must be considered critical, but also the purity. Again, the limit for permissible mechanical contamination in the final composition was found to be of the order of 0.01%, and the composition must be essentially oxygen-free, if the desired mechanical and electrical properties are to be achieved and to be reproducible. This purity can be achieved in the manner already described for the manufacture of the aforementioned compositions of lead and respectively tellurium, selenium and sulphur, which contained an excess of lead. A significant difference in the method for the production of the alloys of the last mentioned tellurium-rich lead-tellurium compositions lies in the cooling rate after tempering. In this respect, it was found that a cooling rate of 50° C per hour is sufficiently low to supply the body with the required mechanical strength.

The tellurium-rich lead-tellurium base alloys can, as in the case of the first described alloys, also be termed metallographically as two-phase alloys. The excess of tellurium in the tellurium-rich compositions behaves to a large extent in the same way as the excess of lead in the first mentioned alloys, in that it determines the polarity of the conductivity and of the thermoelectric power in the primary lead-telluride phase, and it also forms an observable secondary phase at the grain boundaries, as already discussed earlier. The compositions also have satisfactory stability for use at temperatures up to 400°C.

It will be noted from the foregoing that the tellurium-rich lead-tellurium base compositions contain from 38.20 to 42.0 weight percent tellurium, or from 0.14 to 6.7 weight percent excess tellurium, calculated as indicated above over the tellurium content which is stoichiometrically necessary for connection with lead.

The above-mentioned excesses of tellurium in the last-mentioned compositions can be described as impurities in relation to the primary lead-tellurium phase. As before, these contaminants must be distinguished from any unwanted contaminants, likewise as discussed above. Consequently, these intentional amounts of tellurium present in the composition will, in the present presentation with claims, still be described as "beneficial impurities" in a similar way as before. Likewise, it is the presence of these beneficial impurities which gives these latter additional compositions their desired electrical properties and separates these semimetallic alloys or compositions according to the invention from the lead-telluride intermetallic compound in the primary phase. These beneficial impurities give the lead-telluride intermetallic compound a lower specific resistance without correspondingly sacrificing Peltier emf or low thermal conductivity, and lead to a composition with what can be described as semiconductor-like electrical conductivity or specific conductivity. It should be noted that the excess of tellurium in the described compositions induces a Peltier emf of positive polarity and positive conductivity and thus provides a valuable thermoelectric material for use in heat pumps as described above.

It was further discovered that the electrical properties of thermoelectric materials for heat pumps within the latter range of excess tellurium as a contaminant (these will again for the sake of convenience be referred to as "basic compositions") can be changed perceptibly and reproducibly in an advantageous manner by the addition of measured quantities of substances other than the constituents of the basic compositions. These additions will also be referred to here as "beneficial pollutants" in contrast to unwanted pollutants as discussed earlier. These additions will also be referred to as "activators", since, as will be apparent from the following, they act to strengthen the electrical properties desired in thermoelectric heat pump materials with the latter basic composition. As before, the amounts of these activators as a percentage by weight of the base composition are very small, and in order for them to be as effective as possible and provide reproducibility of electrical properties, the base composition to which they are added should have an even higher purity, i.e. contain even less unwanted and uncontrolled contaminants . As a practical rule, it was found that the tellurium-rich lead-tellurium base compositions should be of such purity in this respect that they do not contain more than 0.001 weight percent of undesirable contamination. In a similar way to the first mentioned activated compositions, the lead content of the present basic composition shall lie in the range from 59.0 to 61.8 weight percent lead, in contrast to the lead content of 58.0 to 61.8 weight percent that was discussed above for the non-activated compositions .

As already mentioned, the above-mentioned tellurium-rich lead-tellurium compositions exhibit positive Peltier emf and positive conductivity. By adding the activators that will be described in the following, these positive electrical properties can be enhanced by the addition of certain activators, and these positive activators cause, in connection with this base alloy, an increase in the alloy's conductivity and a slight decrease in the Peltier emf . Table X below lists in the first column the additives that have been found to be effective as positive activators when added to the latter tellurium-rich lead-tellurium compositions. The second column in Table X indicates the maximum effective concentration limits, calculated in weight percent, of the activators in the base alloy, which act to achieve the object of the invention. It will be understood that these concentration limits denote the maximum concentration which effectively changes the electrical properties of the indicated base alloy. Concentrations of the relevant additives in excess of the stated amounts have no tangible effect with regard to a beneficial change in the electrical properties the invention is concerned with, and in this sense the stated limits are to be considered critical. The third and fourth columns in Table X indicate the electrical properties at room temperature of the above-mentioned tellurium-rich lead-tellurium base compositions activated with maximally useful concentrations of the positive activators after tempering.

From the preceding description it will

it is noted that the maximum beneficial impurity concentrations for the last-mentioned activated tellurium-rich lead-tellurium compositions range from a minimum of 0.28 weight percent (namely, 0.14 weight percent excess tellurium plus 0.14 weight percent sodium) to 5.5 weight percent (namely 4.9 weight percent teilur excess plus 0.60 weight percent thallium), which falls well within the value of approx. 10% by weight which is mentioned above as the maximum amount of beneficial pollution.

Typical data for the last mentioned compositions is shown as an example in fig. 7, which corresponds to fig. 5 and 5A, with the ordinate values on the left indicating the Peltier emf in volts

H2 and the ordinate values to the right indicate

o

in watts per cm, while the abscissa indicates specific resistance in Ohm. cm.

In fig. 7 applies to curve X for compositions of tellurium-rich lead-tellurium within the above-mentioned range and graphically shows the variations in the relationship between Peltier coefficient (jr) and specific resistance (q) when adding positive activators to the last-mentioned basic compositions (right end of the curve represents the minimally effective concentration of the last mentioned positive activators). In a similar way, curve W applies to the last mentioned tellurium-rich lead-tellurium compositions within the specified range and graphically shows varia-

jt2

sj<p>nen in the value of the quotient at

y addition of positive activators to the base composition (the right end of the curve indicates the minimally effective concentration of the positive activators). The curve W shows maximum values for the quotient •

9 The value for specific resistance on the far right of fig. 7 (0.1 Ohm. cm), is characteristic of semiconductor materials. The value for specific resistance on the far left of fig. 7 (0.0001 Ohm. cm), is characteristic of metal. The above-mentioned maximum and extreme values occur at values of

Peltier coefficient and specific resistance first column constituents and weight pro- which lie between the values for metals and cent amounts for different examples of semiconductors. Furthermore exhibits, as early- compositions of the last mentioned species. are mentioned, semi-metallic materials in Second column indicate in Volts the Peltier according to the invention low heat element that can be obtained from each of these coupling ability, which approaches heat elements. The third column indicates for each the switching capability of a semiconductor. The simultaneous compositions the thermal conductivity in occurring high values of the quotient watt per cm per degrees Celsius. Fourth column

jt2 , , denotes the function coefficient for each

——- and low values for heat conduction composition. It will be noted that the ability (K) conditions the valuable operating characteristics examples of compositions in-create that are desired for heat pump doctor-contains tellurium-rich lead-tellurium-base alloy-more and further each of the last mentioned I the table XI below indicates positive activators.

Claims (14)

1. Thermoelectric heat pump where i at least one of the thermoelectric components consists of a semi-metallic compound with a proportion deviating from the stoichiometric composition, characterized in that this compound contains as basic components lead and a further component consisting of tellurium, selenium, sulphur, tellurium plus selenium or selenium plus sulfur and the deviation from the stoichiometric quantity ratios for improving the electrical properties of the compound is provided by means of an excess of lead in respectively lead-telluride, lead-selenide and/or lead-sulphide or an excess of tellurium in lead-telluride and/or by means of an effect-promoting additive and constitutes less than 10.4% by weight of the alloy, and that the effect-promoting additive in the case consists of at least one of the substances bismuth, tantalum, manganese, zircon, titanium, aluminium, gallium, chlorine, bromine, iodine, uranium , potassium, thallium, arsenic, silicon, indium, copper, gold, antimony, fluorine, niobium, sodium, silver. 2. Thermoelectric heat pump as stated in claim 1, characterized in that the said compound without effect-promoting addition contains lead in an amount that can vary from a value between a minimum of 61.95 and a maximum of 65.0 percent by weight when the other component is essentially only tellurium, and to a value between a minimum of 72.45 and a maximum of 75.00 weight percent when the other component is essentially only selenium, and further to a value between a minimum of 86.63 and a maximum of 87.20 weight percent when the other component in the essential only is sulphur. 3. Thermoelectric heat pump as stated in claim 1, characterized in that the said composition, using the effect-enhancing additive as mentioned, contains lead in an amount that can vary from a value between a minimum of 61.95 and a maximum of 63.0 weight percent when the other component is mainly only tellurium, to a value between a minimum of 72.45 and a maximum of 73.50% by weight when the other component is mainly only selenium, and further to a value between a minimum of 86.63 and a maximum of 87.10% by weight when the other component is mainly only is sulfur. 4. Thermoelectric heat pump as specified in claim 1 or 3, characterized in that the effect-enhancing additive, depending on the basic components, is selected from among the substances sodium, potassium, thallium, arsenic and silver in such a way that it gives the component positive polarity in relation to the other component that interacts with the heat pump. 5. Thermoelectric heat pump as specified in claim 1 or 3, characterized in that the effect-enhancing additive depending on the basic components is selected as follows from among the substances bismuth, tantalum, manganese, zircon, titanium, aluminium, gallium, chlorine, bromine, iodine, uranium, silicon, indium, copper, gold, antimony, fluorine and niobium that it gives the component negative polarity in relation to the other component of the heat pump that interacts with it. 6. Thermoelectric heat pump as specified in one of claims 1-5, characterized in that the compound contains no more than 0.001 weight percent of contaminants in addition to any effect-enhancing additives as mentioned. 7. Thermoelectric heat pump as stated in claim 1, where at least one of the components essentially consists of lead and tellurium, characterized in that the lead is present in an excess of no more than 3.02 percent by weight in relation to the stoichiometric composition of lead-telluride. 8. Thermoelectric heat pump as stated in claim 1, where at least one of the components essentially consists of lead and selenium, characterized in that the lead is present in an excess of no more than 4.1 percent by weight in relation to the stoichiometric composition of lead selenide. 9. Thermoelectric heat pump as stated in claim 1, where at least one of the components essentially consists of lead and sulphur, characterized in that the lead is present in an excess of at most 3.9 percent by weight in relation to the stoichiometric composition of lead sulphide. 10. Thermoelectric heat pump as stated in one of the claims 1-7, characterized in that the effect-promoting addition in the event that the relevant component of the heat pump essentially consists of lead and tellurium, includes one or more of the elements bismuth, tantalum, manganese, zircon, titanium , aluminium, gallium, chlorine, bromine, iodine, uranium, sodium, potassium, thallium and/or arsenic. 11. Thermoelectric heat pump as specified in claim 1 or 8, characterized in that the effect-promoting addition in the event that the relevant component of the heat pump essentially consists of lead and selenium, includes one or more of the elements iodine, chlorine, zircon, bromine, silicon, titanium , indium, tantalum, gallium, aluminium, copper, gold, fluorine, bismuth, antimony, niobium, sodium and/or potassium. 12. Thermoelectric heat pump as specified in claim 1 or 9, characterized in that the effect-promoting addition in the case that the relevant component of the heat pump consists of lead and sulphur, includes one or more of the elements zircon, indium, bromine, chlorine, titanium, iodine, tantalum, bismuth, gallium, fluorine, niobium, uranium and silver. 13. Thermoelectric heat pump as stated in claim 1, characterized in that the tellurium is present in stoichiometric excess in relation to lead with a value below 4.9% by weight in relation to the stoichiometric composition of lead-telluride. 14. Thermoelectric heat pump as stated in claim 1 or 5, characterized in that at least one of its components essentially consists of lead and tellurium in an amount of 58.0 to 61.8 weight percent lead, rest tellurium.