Electronic Specific Heat of Dilute Alloys: Fe(Ti), Fe(V), Fe(Cr), Fe(Mn), Fe(Co), Fe(Ni), Fe(Al), Fe(Si), Fe(Mo), Fe(W), and Ag(Au)

Abstract
A method of measuring simultaneously the low-temperature specific heat of three samples, typically one pure-metal sample and two alloys with 1 and 2% solute, is described. The procedure is used to determine the variation of the electronic-specific-heat coefficient γ (the coefficient of the term linear in temperature) with concentration c for various solutes (Al, Si, Ti, V, Cr, Mn, Co, Ni, Mo, and W) in Fe and for Au in Ag. The simultaneous calorimetry gives an increase in relative accuracy sufficient to enable determination of variations of (1γ)dγdc to within ±10% for dilute alloys of Fe. For Ag(Au) alloys the variation of γ with concentration is less than can be detected in these experiments. A linear decrease of γ with increasing c is found for six of the iron-based alloy systems. For these the values of (1γ)dγdc are: Fe(Ti), -1.0; Fe(V), -2.2; Fe(Cr), -2.0; Fe(Mo), -0.8; Fe(W), -2.4; and Fe(Co), -0.6. A nonlinear increase of γ with c is found for the remaining. The average values of (1γ)dγdc are: Fe(Al), +1.2; Fe(Si), +0.6; Fe(Mn), +2.0; and Fe(Ni), +2.6. From experiments on two-phase alloys of Fe-Fe2Nb, γ for Fe2Nb is estimated to be between 9 and 10 mJ/deg2 mole. There appears to be a correlation of (1γ)dγdc with certain aspects of Fe-alloy phase diagrams, namely, the variations of Curie temperature and the αγ transformation temperature with concentration of solute.