稀有金属(英文版) 2019,38(01),35-41

Pressureless sintering behavior and properties of Ag-SnO₂

Henri Desplats Elodie Brisson Philippe Rogeon Patrick Carré Alexandre Bonhomme

Department of Materials Engineering, IRDL, FRE CNRS 3744,University of South Brittany

Schneider Electric-Electropole,31 rue Pierre Mendes France

作者简介：*Henri Desplats e-mail:henri.desplats@univ-ubs.fr;

收稿日期：20 September 2016

基金：financially supported by the French National Research Agency REF ANR(No.ANR-09-MAPR-0007-MAPR);

Pressureless sintering behavior and properties of Ag-SnO₂

Henri Desplats Elodie Brisson Philippe Rogeon Patrick Carré Alexandre Bonhomme

Department of Materials Engineering, IRDL, FRE CNRS 3744,University of South Brittany

Schneider Electric-Electropole,31 rue Pierre Mendes France

Abstract：

In this study, the results of measurements on pressureless sintering behavior of Ag-SnO₂(88%wt Ag,12%wt SnO₂) pellets were reported. Dilatometric measurements, relative densities, hardness values, rupture transverse strength and electrical conductivities function of sintering temperatures were presented. A constant thermal expansion coefficient was determined, and a threshold temperature of densification(T_d) was exhibited. Sintering kinetics were reported for different temperatures. Hardness values were measured, and no increase in hardness is found under Td. Three-points bending tests were used to determine the transverse rupture strength whose evolution appears importantly well under Td. In the same manner, the increase in initial electrical conductivities begins well under Td. Under the threshold temperature, the relative increase in electrical conductivity is found to be independent of initial density of green compact pellets. This work highlights different evolutions in function of sintering temperature for the electrical conductivity and transverse rupture strength on the one hand, and for the densification and hardness on the other hand.

Keyword：

Ag-SnO₂; Pressureless sintering; Dilatometry; Mechanical properties; Electrical conductivity;

Received： 20 September 2016

1 Introduction

Ag-SnO₂ is contact material for the use in switches and circuit breakers.The control of electrical and mechanical properties of this material is an important issue.Ag-SnO₂was originally developed for the Ag-CdO replacement,in order to reduce human health risk and pollution.Silver is chosen because of its good electrical conductivity.The function of the metal oxide is to improve switching mechanical behavior.This,in combination with the finding of excellent low contact erosion at service life and good contact resistance values,gives a good substitute contact material of Ag-CdO [ 1, 2, 3, 4, 5, 6, 7, 8, 9] .Its high thermal stability reduces erosion by the electric arc.The effect of the proportion of the metal oxide on the electrical tests is sometimes analyzed [ 10, 11, 12, 13, 14] ,but poor wettability of tin oxide in the silver (wetting angles measured in Refs [ 15, 16] ) generates a more difficult sintering and a tendency toward segregation of the surface oxide;the adding of a dopant(In₂O₃,Bi₂O₃,CuO...) reduces these drawbacks,giving better electrical performances [ 17, 18, 19, 20, 21, 22, 23] .

There are mainly three major families of processes to produce this material:the internal oxidation,the chemical precipitation [ 23, 24, 25, 26, 27] and the milling (reactive [ 28] or nonreactive [ 29] ).Here,Ag-SnO₂ contact material is manufactured by chemical precipitation.This process ensures homogeneous oxide dispersion within the silver matrix.The powder employed for this work is an industrial powder(Schneider Electric Industry).The composition of the powder is 88 wt%Ag and 12 wt%SnO₂,doped with traces of CuO.The addition of CuO increases the wettability of SnO₂ in the silver matrix

After that,powder is shaped into pellets by compression:the greater the compaction pressure is,the greater the density of the green pellet is.In the next step,the powders can be sintered by heat treatment,combined or not to pressure.Various methods exist for densification under pressure,whose advantages are described in detail in Ref. [ 30] .In particular,pulsed electrical current,producing spark plasma sintering process [ 25] ,permits rapid heating of samples.Pressureless sintering,as realized here,is produced generally by heating in furnaces.Eventually,a supplementary shaping stage is realized [ 10] .In this work,densification was performed in furnace of the dilatometer,in air without applied pressure.

2 Experimental

The powder was compacted into cylindrical pellets with initial diameter of D₀=24 mm and thickness of 4 mm.For a compacting pressure of 650 MPa,the relative density was 85%,and for 150 MPa the relative density was 60%.

The microstructures of the compacted powder were measured by scanning electron microscope (SEM,JEOL JSM-6460LV).Measurements of densities were deduced from measurements of the variation of the diameter of the pellets in a dilatometer (LINSEIS L76/10,pressureless or free sintering).Mass of pellets was measured with MET-TLER TOLEDO XS205 balance.Mitutoyo Vickers hardness testing machine (AVK-C2) was used in standard conditions.HV1 was measured.For each test,at least five measurements were taken and were averaged on a pellet.

Three-points bending tests were achieved on sintered Ag-SnO₂ parallelepipedic samples (b=2 mm for thickness and width) pellets;the device used was a micro-test stage (Deben,Suffolk,UK) equipped with a load cell.The sample was placed on two supporting pins spaced from a distance of L=18 mm.The third loading pin applies a force (F) which was measured in the middle of the sample.The displacement of the moving pin and the applied force were recorded until the fracture strength was reached.

Electrical conductivity was measured at room temperature,by using an eddy currents probe with the device FISCHERSCOPE MMS PC2.The values were the average of five different points of measurements on the surface of the pellets.

3 Results and discussion

3.1 Microstructure of composite

Figure 1 shows SEM images of Ag-SnO₂ powder;particles seem to have a dendritic shape,specific to those obtained by electrochemical methods [ 28] .The particle (grain) size is of the order of 1μm:grains form agglomerates of a few tens of microns in diameter.

Figure 2a,b corresponds exactly to the same locations marked precisely (same local coordinates linked to the pellets).No discernable evolution of the microstructures is seen in Fig.2a,b,and the same structures on both views can be found.In the same manner,Fig.2c,d shows that at exactly the same place (the same compact powder heated at600℃),no appreciable change in microstructure can be found.For the difference between 65%and 85%compact powder pellets,it is considered [ 31] that during compaction,higher stresses promote higher plastic deformation of the grains,resulting in two consequences:the closing of the pores and the enlargement of the contact areas between particles which are more important for 85%compact pellets.

3.2 Dilatometry and densification

Figure 3 represents the variation of the pellet diameter from dilatometric tests with temperature (T) and time (t),ΔD(T,t).Three successive steps are programmed for sintering in the dilatometer.The first one corresponds to the rise in temperature from room temperature to 900℃;then,a plateau at 900℃corresponds to the main part of densification (sintering densification);and finally,cooling to room temperature is the third step.

In a similar way to what was done by K.Maca et al. [ 32] ,αwas first determined as the coefficient of linear thermal expansion (CTE) and then the irreversible shrinkage of the pellet,allowing calculation of the densification.

During heating,the diameter increases linearly with temperature[Eq.(1)]until a threshold temperature (T_d)from which densification begins caused by sintering mechanisms.This temperature is estimated at T_d=700℃,for pellets with initial relative density of85%(650=±50℃depending on the method of determination and on the initial relative density).

Then,with T_d the maximum temperature for non-densifying temperature,αvalue is deduced:

The slope of the linear variations,ΔD_therm(T),is determined,as shown in Fig.3.It is checked that the slope during cooling,determined in a similar manner,is the same as that during heating.

An average value ofαis determined as 17×10^-6 K^-1,which is lower than the CTE of silver equal to21.4×10^-6 K^-1 (average value between 293 and 800 K [ 33] ).The effect of SnO₂ (CTE of 5.6×10^-6 K^-1 [ 34] ) is a decrease in the effective CTE.Similar effect of ceramics in silver matrix composites was also observed by Chang et al. [ 35] .Calculation of theoretical CTE obtained by a volumetric rule of mixing gives a value of 18×10^-6 K^-1.For the different initial densities of Ag-SnO₂ used here,there is not any difference in CTE.

Fig.1 SEM images of green powder with different magnifications [31]

Fig.2 SEM images of compacted powder with initial relative density of 60%(a initial and c heated at 600℃) and initial relative density of85%(b initial and d heated at 600℃)

For temperature above T_d,the variation of diameter as a function of temperature and time is:

whereε_therm(T,t) is the thermal deformation andε_sint T,t) is the negative sintering deformation [ 32] .Considering isotropic thermal expansion and isotropic sintering shrinkage,the calculation of the density p(T,t) and the relative density d(T,t) was done using following relations [ 32] :

whereρ_green(T₀) is the initial density,ρ_d(T₀) is the density in the fully dense state,and d₀ is the initial relative density.They are related toρ_green(T₀)=ρ_d(T₀)d₀ withρ_d(T₀)=9893 kg m^-3.ρ_d(T₀) is deduced from the proportion of Ag(88%) and SnO₂ (12%).

Fig.3 Dilatometric measurement of variation of pellet diameter as a function of temperature

In Eq.(4),the relative density with temperature and time,d(T,t),is deduced from the value ofΔD_sint(T,t)(resulting from the measured valueΔD(T,t) and from thermal dilatationΔD_therm (T)).At room temperature,the measured densities by Archimede’s principle when it is possible(relative density greater than 92%) are in good accordance with the final densities obtained from dilatometric measurements deduced from Eq.(4).

Isothermal dilatometric tests were conducted at different temperatures (750,850 and 900℃).From these tests,it is possible to deduce the evolution of the kinetics sintering[relative density,d(T,t)]for each temperature,as shown in Fig.4.The changes in final relative density (after 2-h sintering) with the sintering temperature are reported in Fig.5.For the temperature of fusion of silver,961℃,the relative density is 100%(extrapolated value).

Fig.4 Isothermal sintering kinetics at temperatures of 750,850 and900℃

Fig.5 Relative density as a function of sintering temperature with2-h isothermal sintering and initial relative density of 85%

3.3 Hardness and transverse rupture strength

Figure 6 shows the evolution of hardness at room temperature with the maximum temperature at which the sample was heated for 2 h.One can see that the hardness increases beyond the temperature of densification.The increase in hardness depends on the maximum temperature at which the samples were heated;at 930℃,the hardness increases by 38%,from initial value of HV 63.Similar values were found by Pandey et al. [ 26] for sintered samples.

The indenter,whose imprint is of the order of 90μm(several times of the average size of grains and aggregates of grains,20μm [ 31] ),is subjected to a greater material resistance,then greater hardness,when density increases with sintering temperature.

From the failure load (F),the transverse rupture strength(TRS) is determined by the following relation.

Fig.6 Hardness at room temperature as a function of sintering temperature with 2-h isothermal sintering and initial relative density of 85%

The TRS increases with the maximum temperature reached (Fig.7).No effect of heating time (from 15 min to2 h) is found on the value of TRS.

Contrarily to the hardness,TRS grows up with temperatures below T_d.More than for hardness,the TRS involves shearing and cohesion between the grains.Then,the increase in TRS might be interpreted as change in the cohesion between the grains,growing with heating temperature,as the consequence of better contact,obtained by bonding diffusion [ 31] without decreasing the volume of the granular compact.

3.4 Electrical conductivity

Initial electrical conductivities (σ₀) for green samples are6.1 and 18.2 MS·m^-1,for the sample with relative density of 60%and 85%,respectively.The higher the relative density is,the higher the electrical conductivity is.It was shown [ 31] that the increase in the electrical conductivities of compact powder with relative density follows an empirical correlation proposed by Koh and Fortini [ 36] .

Figure 8 shows the relative variation of the normalized(by initial conductivity,σ₀) electrical conductivities with the maximum temperature at which the pellets were heated.Measurements were done at room temperature.The variation in electrical conductivity occurs under the threshold temperature (T_d,around 600℃,for 60%initial relative density).

But SEM results indicate no change in morphology or appearance (Fig.2).Some authors [ 25] ,in particular,exhibit the growth of necks between the powder particles at initial stage of spark plasma sintering with Ag powders sphere like.Here,because of the important dispersion of particles size and shape (Figs.1,2),it is not possible to investigate the evolutions of sizes of the contact areas between the particles through the formation of necks.Moreover,the interfaces between the particles with microcontacts are not accessible through SEM observation(Fig.2b,d).It is assumed that the contact between the grains is rendered more perfect due to thermally activated bonding diffusion.This bonding diffusion insures an intimate contact between the grains,without decreasing the volume of the granular compact.The increase in electrical conductivity with the increase in TRS does not depend on heating time (from 15 min to 2 h).At 600℃,the relative increase in conductivity is 43%.

Fig.7 Transverse rupture strength at room temperature as a function of sintering temperature with isothermal sintering and initial relative density of 85% [31]

Fig.8 Normalized electrical conductivity (σ/σ₀) at room temperature as a function of heating temperature,under threshold temperature of densification (d₀=60%,85%)

A decrease in TRS and electrical conductivities at a temperature around 300℃can be observed from Figs.7,8.This could be due to a physicochemical reaction;further investigations to elucidate the mechanism must be conducted.

At temperatures above the free densification threshold,the increase in electrical conductivity is more important(Fig.9).In addition to improved contacts quality,the disappearance of porosity (very low conductivity of airfilled pores) and the development of contact during densification may explain an additional increase in electrical conductivity of the densified material.Much better established,than the increase in electrical conductivity with the cohesion of the grains before the densification,is the increase in the conductivity with densification (or decrease with porosity),permitting the modeling of the phenomenon,in particular for metal-based densified powders [ 37, 38] .

Fig.9 Electrical conductivity at room temperature as a function of heating temperature for 20 min with initial relative density of 60%

4 Conclusion

In this study,measurements of thermal dilatation,sintering kinetics,relative densities,hardness,transverse rupture strength and electrical conductivity were investigated concerning Ag-SnO₂.Porous pellets with 85%initial relative density were used,mainly,and heated at different temperatures.Measurements in dilatometer give the evolution of isothermal free sintering in function of time.A threshold temperature of free densification(T_d=650±50℃) is determined.An average coefficient of thermal expansion of 17×10^-6 K^-1 is found,independent of the porosity.Experimental results show that the variation of hardness depends on the temperature above the threshold of densification (T_d).

By against the increase in transverse rupture strength and electrical conductivity occurs at temperatures well below the beginning of densification.After heating at the threshold temperature,the electrical conductivity measured at room temperature increases and is found to be 43%above the initial value.Observations show no change in the morphology or structure of the porous material when the conductivity variations occur.This might be the indication that only the quality of contact between the grains changes.The relative change in electric conductivity is found to be independent of the initial density of green pellets and heating time.

参考文献

[1] Wojtasik K, Missol W. PM helps develop cadmium-free electrical contacts. Met Powder Rep. 2004;59(7):34.

[2] Nilsson O, Hauner F, Jeannot D. Replacement of AgCdO by AgSnO_2 in DC contactors. In:Proceedings of the 50th IEEE Holm Conference Electrical Contacts and the 22nd International Conference on Electrical Contacts, Seattle; 2004. 70.

[3] Gengenbach B, Mayer U, Michal R, Saeger K. Investigation on the switching behavior of AgSn02 materials in commercial contactors. IEEE Trans Compon Hybrids Manuf Technol. 1985;8:58.

[4] Wingert PC, Leung CH. The development of silver-based cadmium-free contact materials. IEEE Trans Compon Hybrids Manuf Technol. 1989; 12:16.

[5] Behrens V, Honig T, Andreas Kraus A, Michal R, Saeger K,Schmidberger R, Staneff T. An advanced silver/tin oxide contact material. IEEE Trans Compon Hybrids Manuf Technol. 1994;17(24):B3.

[6] Wingert PC, Leung CH. Comparison of the inherent arc erosion behaviors of silver-cadmium oxide and silver-tin oxide contact materials. IEEE Trans Compon Hybrids Manuf Technol. 1987;10(1):56.

[7] McDonnell D, Gardener J, Gondusky J. Comparison of the switching behavior of internally oxidized and powder metallurgical silver metal oxide contact materials. In:Proceedings of the 39th IEEE Holm Conference on Electrical Contacts, Pittsburgh; 1993. 37.

[8] Behrens V, Honig T, Kraus A, Michal R, Saeger KE, Schmidberger R, Staneff T. An advanced silver/tin oxide contact material. In:Proceedings of the 39th IEEE Holm Conference on Electrical Contacts, Pittsburgh; 1993. 19.

[9] Nath D. Comparative behavior of silver tin oxide and silver cadmium oxide contact materials in commercially available contactors. In:Proceedings of the 36th IEEE Holm Conference on Electrical Contacts, Montreal; 1990. 126.

[10] Talijan N, Cosovi V, Staji-Trosi J, Gruji A, Zivkovic D, Romhanji E. Microstructure and properties of silver based cadmium free electrical contact materials. J Min Metall. 2007;43B(2):171.

[11] R(o|¨)hberg J, Honig T, Witulski N, Finkbeiner M, Behrens V.Performance of different silver/tin oxide contact materials for applications in low voltage circuit breakers. In:Proceedings of the 55th IEEE Holm Conference on Electrical Contacts, Vancouver; 2009. 187.

[12] Ivanov V, Nikolaeva S, Sidorak A, Shubin A, Sidorak V. Ag/ZnO and Ag/Sn02, Electrocontact materials obtained from fine-grained coprecipitated powder mixture. J Sib Fed Univ Chem. 2012;5(2):131.

[13] Witter G, Chen Z. A comparison of silver tin indium oxide contact materials using a new model switch that simulates operation of an automotive relay. In:Proceedings of the 50th IEEE Holm Conference on Electrical Contacts and the 22nd International Conference on Electrical Contacts, Seattle; 2004.382.

[14] Chen ZK, Witter GJ. A study of dynamic welding of electrical contacts with emphasis on the effects of oxide content for silver tin indium oxide contacts. In:Proceedings of the 56th IEEE Holm Conference on Electrical Contacts, Charleston; 2010. 1.

[15] Jeannot D, Pinard J, Ramoni P, Jost EM. Physical and chemical properties of metal oxide additions to Ag-SnO_2 contact materials and predictions of electrical performance. IEEE Trans Compon Hybrids Manuf Technol. 1994;17:17.

[16] Wang H. Influence of rare earth on the wetting ability of AgSnO_2 contact material. Rare Met Mater Eng. 2014;43(8):1846.

[17] Rong MZ, Wang QP. Effects of additives on the AgSn02 contacts erosion behavior. In:Proceedings of the 39th IEEE Holm Conference on Electrical Contacts, Pittsburgh; 1993. 33.

[18] Leung C, Streicher E, Fitzgerald D, Cook J. Contact erosion of Ag/Sn02/In2O_3 made by internal oxidation and powder metallurgy. In:Proceedings of the 52nd IEEE Holm Conference on Electrical Contacts, Montreal; 2006. 143.

[19] Gardener J, McDonnell D. The impact of process changes on the physical properties and formability of silver metal oxide contact materials. In:Proceedings of the 41st IEEE Holm Conference on Electrical Contacts, Montreal; 1995. 373.

[20] Chen ZK, Witter GJ. Comparison in performance for silver-tin-indium oxide materials made by internal oxidation and powder metallurgy. In:Proceedings of the 55th IEEE Holm Conference on Electrical Contacts, Vancouver; 2009. 182.

[21] Wang J, Wen M, Wang B, Lu J. Study on a new contact material Ag/Sn02-La2O3-Bi2O3. In:Proceedings of the 47th IEEE Holm Conference on Electrical Contacts, Montreal; 2001. 94.

[22] Wang H, Wang J, Wen M, Zhao J, Liu G. Preparation of Ag/Sn02+La2O3+Bi2O3 contact material electrical contacts. In:Proceedings of the 52nd IEEE Holm Conference on Electrical Contacts, Montreal; 2006. 131.

[23] Mutzel T, Niederreuther R. Advanced silver-tin oxide contact materials for relay application. In:Proceedings of the 26th Electrical Contacts, Seattle; 2012. 194.

[24] Xiuqing Q, Qianhong S, Lingjie Z, Lawson C, Xianping F, Hui Y. A novel method for the preparation of Ag/SnO_2 electrical contact material. Rare Met Mater Eng. 2014;43(11):2614.

[25] Xiong Q, Wang S, Xie M, Chen Y, Zhang J, Wang S. AgSn02electrical contact material prepared by spark plasma sintering.Precious Met. 2013;34(4):12.

[26] Pandey A, Verma P, Pandey OP. Comparison of properties of silver-tin oxide electrical contact materials through different processing routes. Indian J Eng Mater Sci. 2008;15(3):236.

[27] Lungu M, Gavriliu S, Canta T, Lucaci M, Enescu E. Ag-Sn02sintered electrical contacts with ultrafine and uniformlydispersed microstructure. J Optoelectron Adv Mater. 2006;8(2):576.

[28] Lorrain N, Chaffron L, Carry C, Delcroix P, Le Caer G. Kinetics and formation mechanisms of the nanocomposite powder Ag-Sn02 prepared by reactive milling. Mater Sci Eng A. 2004;367:1.

[29] Zoz H, Ren H, Spath S. Improved Ag-Sn02 electrical contact material produced by mechanical alloying. Metallurgy. 1999;53(7-8):423.

[30] Orru R, Licheri R, Locchi AM, Cincotti A, Cao GC. Consolidation/synthesis of materials by electric current activated/assisted sintering. Mater Sci Eng R. 2009;63:127.

[31] Brisson E, Carre P, Desplats H, Rogeon P, Keryvin V, Bonhomme A. Effective thermal and electrical conductivities of AgSn02 during sintering. Part I:experimental characterization and mechanisms. Metall Mater Trans A. 2016;47(12):6304.

[32] Maca K, Pouchly V, Boccaccini AR. Sintering densification curve—a practical approach for its construction from dilatometric shrinkage data. Sci Sinter. 2008;40:117.

[33] Touloukian YS, Kirby RK, Taylor RE, Desai PD. Thermophysical properties of matter, vol. 12. New York:IFI Plenum;1975. 298.

[34] Samsonov GV. In:Samsonov GV, editor. The Oxide Handbook.2nd ed. New York:IFI/Plenum; 1982. 126.

[35] Chang SY, Lin SJ, Flemings MC. Thermal expansion behavior of silver matrix composites. Metall Mater Trans A. 2000;31:291.

[36] Koh JC, Fortini A. Prediction of thermal conductivity and electrical resistivity of porous metallic materials. Int J Heat Mass Transf. 1973;16:2013.

[37] Montes JM, Cuevas FG, Cintas J. Porosity effect on the electrical conductivity of sintered powder compacts. Appl Phys A.2008;92:375.

[38] Argento C, Bouvard D. Modeling the effective conductivity of random packing of spheres through densification. Int J Heat Mass Transf. 1996;39:1343.