Document xYY1d3xyJp0VDN15dOv6ReNb

LIA24753 LEAD INDUSTRIES ASSOCIATION, INC. 1SS M A 011 N AV KNUI NKW YORK IT, N. V. Maxell 2, 1962 SUBJECT! THnWQEtZCBUC MATERIALS To Mesber* of the Lead Iivlurtria* Association, Inc.: He axe attaching an Informative article 00 thermoelectric aterial*, vhich Appear*! In a recent lsfue of Material* la Dealga Engineering Magazine. It explain* the possibilities of the** material* and Bake* a good caae for lead tellurld*. V* he11ere you vlll find It lntereatlng. It la our plan to distribute the balance of our supply to people rlsltlng our booth at the Design Engineering Show to ba held In Chicago's HrCord ck Place, April 30 to Kay 3, 1962. Very truly yours. RLZ:m)t be. Executive Vice President . . . ---------------- - 1.. ............ ............... -. . ............ ' Look AhnJ with L--d i npni.ipm^B LIA2S75A i ijutija ts I \W 4444 444 & 4A4 44 4^ X4444444 fi 44 44 4 4 4 4 4 4 4 44 4 4 444^ 444** ,,*&*VAYi** . -1 iiid^*4. * UH taltundt /< <%*' ?wm A*ef fimm }**'."* '*<* fn| +n4*J f f afrw* /r fit /rtCr Thermoelectric Materials Today \t`w and improivd material* are1 bringing tin 1 inoettctrie devias closer to commercial realitg. frjr D Tk*rmofU<lr*e J'twdmrlt I*rf4^ Af-f4trJ Rtrt+rck, JI/iKWn-ft Mtmtmf 4c Mf$. ('. TKUtMOruenue MaTIXULS have Ike ability to I) directly convert thermal energy into ebctrical en err?. *nd 2) transfer thermal er ergy by mean* of an eb-ctnc cur rent. Althourh thermoelectricity i* more than 100 years old, it was oot until the 194J`a with the ad vent of semiconductor technology, that its possible u m\* l>rgan to be reali led. Recent materials ad vances have enabled engineers to tome up with many unique and competitive devices for power generation and heat transfer. Although several different ma tenals are currently taing used for generators, lead telluride (sc*' Table 1) appears to he the brV tssu i-eetirM(i sr u a s ULIVSIN SUSVt 0lb u mx tK>v* Vieepk. pa Vtvsgsa. pa Vw( i iWrin, pa Cart T Krr f ip pel *f O.'V loco* 10000* ?. 10 4 1C. 10 because of ita high efficiency and its low coat. Germanium telluride exhibit* alight)/ greater (ffWirncy than p lytic lead telluride, but when costa are considered th** ratio of power to cost-jiri-pound is at>out one sixth that of lead telluride. Other compounds such ns crsnim sulfide can be nf-erabd at higher temperatures than lead telluride but their efficiencies arr lower and material costs higher. Most manufacturers have adopt* ed bismuth telluride as a heat pump material. This is because bismuth telluride exhibits the most desirable thermoelectric properties between " 50 and 400 K. the temperature range where heat pump* are most widely used. Cost an Important factor Although thermoelectric device* have been successfully developed for many spcclaliied application*, widespread commercial use can only be achieved if thermoelec tricity becomes economically com petitive with conventional ayafeme Thus, materials cost it an : ~'1* - - N 1747.01 * LIA24755 Ak^Wiifir via. lm|>orUitt deiyn factor, a* shown in Table 2 which compare* the electric materials. However, on the bast* of antiri|iated Improvt- the square of lemjwralure differ* ence. This accounts for the fact costs and efficiencies of thermorlec trie generator materials. Note that thf ftyure* in the table reflect only the basic rout of the elements and do not Include I rocesMn* cnf*. Some of the ma terials are mote difficult to process than other* ;nd this factor will affect the flnu.'ied |rice of ihermo- rm-nt* in processing technique*, the dominant cost factor will I* raw materials col. Tampe/atura aftacts proportipt The efficiency of the ihernnwlectrie material is proportioral to thr temjwraturr difference across it, whereas power is proportional to that a material with low efficiency can still haic a hijrh power rating if it can be o|<eratrd at a hifh enough lemj>rraiurr. Fig J shows how conversion rfficien.) vf |iTrs rnt thermoelectric generator ma trrials vane with ojeratm# ternjwraturr. The end of each rurv* represent* the hiirhest tempera ture at which the material ran maintain chemical stability. Although efficiency values as hlrh as 13f. ran be obtained with material* such aa lead tellundr, we must remember that these are theoretical efficiencies fot convert, in* thermal energy into electrical eneriry. In practice not all of the heat produced by the source flows through the thermoelectric ms tens) Furthermore, resistance at contact points w ill detract from unable output power Actually, the highest efficiency w have achirved in a practical dev ice is h.5r*. Thi value represents true overall effi ciency. i f., the ratio of net elec trical cut|>ut to total thermal 1--<T**ct#f rfl/ifirv k trmptn Imrt f+r 1wpir+i (A*rmm*l*c*rir +rm*r ml#e mmlmml*. power input. Theoretically, it i* possible to obtain efficiency values as hifh as tK% by usin* continuous seg ment* of different thermoelectric 2-- Effrt / Jwr*ng Unit mm prep. rrft / Ud ttltvtryd*. ToN>S'<ltrl { r~*nrf rf Umptrntr# fmr feW *//->/* r* eediwm and Irwd doy*ng mgrml*. BvTTW*--ru- r* trmprrmtmrf tuing |rW d>pp prf. materials. By joining different materials in series electrically and thermally it is possible to use each material over the temperature ranye where it Is mot efficient. However, this procedure has it* drawliacks. principally hreau** tare is n<-*d-d to match the rn fficient of expansion of various materials and because diffusion raw.* r-Mia t asaanjcrsic ma t u u ii c s w a s c Msteng % 1 | Cend*. Tee* Dff IflKIffKf ;tint? tp UUI** <*> Cm*. (ftiomcv ref CeU 1 , Ceil *j M ft*tw Its* fsflvcvU <nU) i* Ir Tfl*nd* (tdole) r J* kr*-r**,4t (7*St) r g<V*vtA Te^itfiiOw tls.if ei> 1 N C*m*iwn Tsdso^t (Ulr) f Isd'Vjm (ItsAi) ... ft t*l"wtt TeHanM iB.dh) r CrPv<* Sv't 4* fCe^i 1010 US toao II 5 ! }lt s IK /SI 19 I too tt 7? 031 > 1?0 M 94 i 0?l 4i 4 V 4 i 0.9/ 4J ms If , 079 907 IMO rn 2*1 , OO.H 144 im 10* U w KX to; 6tt ! 0 007 ( c la-tween the materials can cause degradation Also, the technique Is more <t'tly and In practice ha* seldom produced more than a Iff'f Increase in output. How performance It measured fn addition to efficiency, the relative merit of thermoelectric materials I* sometime* established by a "figure of merit**: Z-SVeK *C*WWv<*r*4 *S I*tmU*%k S*WM>e*WrtW I* #Wtrlf| artWr Vtlww mw*14 W ** It m*. **>W*l *r*#tl<-*J W--rrrv'M'* 4in*<-,* *HS KU-fl*r iaVltiif wMfc n>U fwfcrtW* autaUlMj t I f. <Ai*e44 w*i>f(ur( ii r taHfiMr 4B**4 arwSvcUua where S is Seebeck coefficient in g ia eh*ctrical resistivity in johm<m, and K la thermal con ductivity In /rm T. mm&tm . u1 n'n iy rutipfMgwii >n ii Jl sZL. LIA24756 A high figure of merit indicate* a good material for either a he*t generator or a heat pump. BcfauM* the three properties in the equation are Dot completely inde pendent, a compromise i usually Decenary to achieve the highest figure of merit Each of the prop* ertiea is dependent oo operating temperature. Although temperalure dependency U not too import* ant with hmt pyxnpe which usually operate over a small raojre of temperature difference (e.g., 12V T), it ran be an imjxjriant ma* teriala selection factor for thermo* electric yrneraton where a tem perature difference of 800 to 1000* P ir not uncommon. A material that performs well over one tern* prraturv rar.ge may be eomp4r1ety Unacceptable over another range. Concentrated materials develop* merit in the past decade has pro duced dramatic increases in the figure of merit values for thermo* electric materials. As a result of this progress some rather extrava gant predictions have been made for the future. Figure of merit \ slues of 0.005 or 0 00* may he obtained writh future material*, but we should not expect these a* a certainty. At the present state of the art we should be realistic and expert normal progress in processing of present materials. Thermoelectric generators ws* nme*** / mf. .tub fe mmrtrt k*mt mt ftwtnir pexyr. *prrtMr*Mf pea /r*4 grrr*f*r j>rnJm<r* to e ( pmpnurr rtimnd es *A/rr rr/ar he <//cw/l nnd mntig ( fU-rumUetrx !<* tush fts fvmr-*trmfmt *it U/or rea I# ee*AiwiW It mrr*nsi patr^e, //ref df pntrrr in wrwiv nreft4 th*rmn*t+ctrie rfrmrmt. )nrl*#x pfefra Three processing methods Binary and ternary thermoelec* trie materials can usually be formed by casting, cryrtai grow* ing or powder pressing tech* biquea. The best method to use de pend* on cost and performance re quirements. The ccwipounds are somewhat more difficult to form than elemental semiconductor ma* teriaK Although reasonable puri ty Jevrl* must be maintained durinjr processing. purity require ments are not ax high aj for germanium and silicon materials used in transistors and diode*. In gen eral, the state of the art has pro* greased to the point where xre can l now ea*ily produce a material that it chemically stable oxer its oper ating ran re and does not deterio rate with time. Because of their susceptibility to oildation, most 1  M !* > LIA24757 How Thermoelectrics Work The basic phenomena of ther moelectricity were discovered In the IVth century by Thumaa Seabeck end itsn Peltier. Today's Ihermoelcclnc generators stem from the discortry by Sodx-ck la 18Z2 that when the Junction of two dissimilar metal* la boated, 4 vvlUifo ia produced across the open terminala which la propor* iional to the temperature differ* .*ce between the open terminals and the boated junction. Pm.et day heal pump* item from the discovery by Peltier a abort time later that an electric current flowing acroaa a Junction of two diaaimilAr metal* cauae* beat to be ab*ubed or evolved at the Junction, depending on the direction of current flow. Actual);, thermoelectric fen eration and heal pumping are merely two apcct of the aame phenomenon. Thermoelec* trie gtneretien ia an energy rocversion proctA* and heat pumping ia an energy transfer process. Thermoelectric generator* and beat pump* usually operate with ptype and n-type material*. (Note: n type material! Have concentrations of electrons; pt)-pe matcnals have concentra tions of "bolas" or locations de void of electron!.) In a thermo electric generator, the tempera ture gradient cauwe electrons in the a-typ* material and hle* in the p-t>pv maiertsi to move toward the cold end*, thus cre ating a potential difference. Conversely, la a beet pump, when an electric current ia passed through a heat absorbing Junction it la carried by elec tron* ia the a-type material and bote* ia the p-type material These charge carrier* both move away from the junction and carry away thermal energy, thereby reducing temperature at the junction. thermoelectric material* must he thermoelectric materials to their devices shown in the photo*. 3M hermetically scaled If they are to electrodes are in genera! use. engineers have also developed: be o j k -rated above floo or 600 F. Some of these are: casting the ma a An experimental electric gener Thermoelectric materialj art terials on to the electrode; solder ator for satellites and space ve usually more heavily doped than ing; and use of pressure contact. hicle* which directly converts the other aemicondurtor material*. Each has it* advantages, depend heat of a radioactive isotope into The effect that doping has on prop ing on the niaterial involved and electric power for communications erties is illustrated in Fig 2 which the application. Care is required a A lighting system for buoys and show* how an increase in doping when selecting an electrode ma other navigational equipment level reduces both the resistivity terial to prevent contamination which eliminates battery systems and frbeck coefficient of lead tel- Certain metals, for example, will by directly converting the energy lunde. The proper amount of react with thermoelectric materi from the combustion of natural doping to use depend* on the par als and degrrie their properties gas into electrical cnerg: ticular application. *f a high cmf or cause a high resistance contact. Spot cooler.* lorated near the cen and low power are needed, then i small amount of doping produces Current uses ter of fluorescent light* which cool a small area on the wall of each 1 the best results. If power is more The most Important advantages lamp bulb, thus maintaining mer important then heavy doping is of thermoelectric devices are their cury vapor pressure at the opti often more suitable. long life and maintenance-free mum value and increasing light Many method* of contacting operation. In addition to the novel output by a* much a* 72^. /Kttt Ay LEAD IMH'VTHIM AWKItTION, IV. 292 A tent** New Voei 17, N*w Verb RtftiufeJ ftfm MATt RIALS in DiMt.s f s u m ; mn ., RcinliolJ l*uhl(dmi Corporation t in Pjrl Aunoi . Ww Volk 22, N. Y. 1962 hike |>VU