• PDF / 313,301 Bytes
• 6 Pages / 420.48 x 639 pts Page_size
• 12 Downloads / 257 Views

DOWNLOAD

REPORT

---

IAN L. SPAIN AND KENNETH J.

VOLIN

Department of Physics, Colorado State University, Fort Collins, CO 80523 ABSTRACT Calculations of the magnetoresistance of graphite acceptor compounds are made using a tight binding model for the carrier dispersion proposed by Blinowski et al, and measured values of the zero-field resistivity. It is shown that, if a reasonable physical model is used for the mobilities, the magnetoresistance cannot be fitted with two- or three-carrier models. Suggestions for the origin of the magnetoresistance are made. INTRODUCTION Electronic properties of acceptor compounds of graphite have been studied extensively since the discovery of electrical conductivity comparable to copper [1-3]. Although subsequent studies have shown that the highest roomtemperature conductivity is only ^2/3 that of copper [4-6], the specific conductivity (conductivity/unit mass) is still about three times higher. Highest conductivity is found for third or fourth stage compounds, with formula typified by C2 4A, or C3 6 A (A = acceptor molecule). If each acceptor molecule accepted unit electronic charge, the hole density, p, would be only a fraction of that of Cu. Experiments have indicated that the density is even lower than this, since there is only fractional charge transfer per acceptor. It is convenient to define fractional charge transfer, f:

f

[A]

[A]-

p [A]

(1)

where [A] is the concentration of acceptor molecules (ionized or neutral) and [A ] is the concentration of negatively charged molecules. Table I lists results for 1st, 2nd and 3rd stage AsF 5 intercalated graphite as an example. It can be seen that estimates differ widely. TABLE I

Comparison of values of the fractional charge transfer, f, for CnAsF5 Technique X-ray absorption (EXAFS) Spin susceptibility Optical reflectivity Magnetoresistance

Magneto-oscillations X-ray photoelectron spectroscopy X-ray diffraction

Mat. Res. Soc. Symp. Proc. Vol. 20 (1983)

Reference

Stage 1

7 8 9 10

0.67 0.24 0.05 0.37

Stage 2 0.67 0.48 0.11 0.43

Stage 3 -

0.14 -

-

3

0.15

11

0.37

0.41

0.45

12 13

-

0.22 0.42

0.41 -

14 11

-

0.3 0.3-0.55

-

15

-

0.45

-

Elsevier Science Publishing Co.,

Inc.

174

One of the techniques of potential use for determining f utilizes resistivity and magnetoresistance data. The conductivity can be written a = EPiepi = pep (2) where the summation is over groups of carriers of mobility pi and density pi (p = Epi). The mobility can be estimated in principle from the magnetoresistance 2 11 2 = (/oB )

p"= SP/B)

-*O

(3)

Unfortunately, the coefficient 8 may differ appreciably from unity and depends on details of the electronic structure. Wu [16] used these equations to fit magnetoresistance data for a two-band model. He adjusted values of pi and pi to give a best fit to experimental data. The purpose of the present paper is to use recent models for the dispersion relationships E(e) for acceptor compounds to analyze resistivity and magnetoresistance data of acceptor compounds CnAsF5 and CnSbF5. MODELS FOR THE ELECTRONIC