Physica B 156 & 157 (1989) 53-55 North-Holland, Amsterdam

MARTENSITIC TRANSFORMATION FROM BCC TO 9R AND FCC IN METALLIC LITHIUM

H.G. SMITH’, R. BERLINER’ and J.D. JORGENSEN3 ‘Solid State Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831, USA ‘Missouri Research Reactor Facility, University of Missouri, Columbia, MO 65211, USA ‘Materials Science Division, Argonne National Laboratory, Argonne, IL 60439, USA

High resolution neutron diffraction experiments have been performed on polycrystalline lithium-7 metal over a temperature and pressure range of 45 K to 300 K and 1 bar to 6.5 kbar. The partial transition from the bee phase to the 9R (Sm-type) phase occurred at approximately 90 K at 6.5 kbar. On warming at 6.5 kbar, the 9R phase transformed to the fee phase over a narrow temperature range near 140 K.

Recent neutron scattering experiments on both polycrystalline [l] and single crystal [2] samples revealed that the low temperature (~75 K) form of metallic lithium is a combina- tion of bee and 9R (Sm-type) material coexisting uniformly throughout the samples. The ideal 9R structure [3,4] is a close-packed lattice consisting of hcp and fee types of sites. Although Barret [5] had interpreted the low temperature X-ray re- sults in terms of an hcp lattice, he showed that Li cold-worked at low temperatures became fee. It has been suggested that the thermodynamically stable form is actually fee with a transition tem- perature of about 110 K. X-ray diffraction studies by Frolov and Rodionov [6] at room temperature to a pressure of 10 000 kg/cm* re- ported a transition to the fee phase at 3000 kg/ cm* (2.94 kbar). Olinger and Shanar [7] in their X-ray studies under pressure between 30- 100 kbar using a diamond anvil apparatus did not observe a phase transition until 70 kbar was reached (also fee).

The sample used in the present experiment was an ingot of 99.99% ‘Li metal 2” in length and etched down to 0.4” in diameter in methanol. It was encased in a vanadium sheath in a He atmosphere and inserted into an Al pressure vessel and pressurized hydrostatically with He gas on the Special Environment Powder Diffractometer (SEPD) at the Intense Pulsed Neutron Source (IPNS) at the Argonne National

Laboratory. Cadmium masks were used to re- move diffraction lines due to the Al pressure vessel.

The initial powder diffraction scans were made at room temperature at atmospheric pressure and at 2.9, 4.5, and 6.5 kbar under hydrostatic conditions. There was no evidence of any trans- formation even though the volume change at 6.5 kbar was twice as large as the volume change from room temperature to 75 K at atmospheric pressure. The sample was then cooled in stages down to 70 K, keeping the pressure as nearly constant as possible at 6.5 kbar. At 90 K a partial transformation to the 9R structure had taken place. The percentage transformed increased to about 50% at 70 K. This indicates T, increased about 15 or 20 K above the atmospheric pressure T,, or roughly 3 K/kbar of applied pressure. This ratio is similar to that obtained by Olinger and Shaner [7] in their room temperature X-ray diffraction studies of Li under pressure in a diamond anvil apparatus. However, they ob- served a change ‘to the fee structure in their experiments.

Upon warming in stages to 140 K at maximum pressure, the 9R structure had disappeared, but a small amount of the fee phase appeared, (ap- proximately 20%), and then disappeared on fur- ther warming to 175 K, leaving only the bee phase.

Several diffraction patterns are shown in fig. 1:

0921-4526/89/$03.50 @ Elsevier Science Publishers B.V. (North-Holland Physics Publishing Division)

54 H.G. Smith et al. I Martensitic transformation in lithium

I

(C) d SPACING (A)

x 8 !

P 1~

SE-

8”

Y 3300 5000 7000 9000 11000 13000 15000

Time of Flight ,u set

Fig. 1. (a) Run #2410 at T= 100 K, P = 1 bar, (b) Run

#2422 at T = 140 K, P = 6.5 kbar, and (c) Run #2418 at

T = 70 K, P = 6.5 kbar.

(a) Run #2410 at T = 300 K, P = 1 bar; (b) Run #2422 at T = 140 K, P = 6.5 kbar (after cooling to 70K), and (c) Run #2418 at T= 70K, P= 6.5 kbar. The initial scan, #2410, showed a small amount of preferred orientation in the bee phase, but this increased somewhat with increas- ing pressure and decreasing temperature, and

was largest after the 9R to fee transformation took place, as shown in fig. l(b) by the differ- ence pattern. A Rietveld refinement of Run #2422 indicated about 20% of the sample was fee and the remainder bee. The lattice constants were determined to be a(bcc) = 3.4361(l) A and a(fcc) = 4.3271(5) 8, with nearly identical atomic volumes, V(bcc) = 20.285 A’ and V(fcc) =

20.255 A”. Because some of the 9R peaks are shifted

from their ideal positions, indicative of severe stacking fault behaviour, it was not possible to do a Rietveld least squares refinement on this phase (fig. l(c)). About 50% of the bee phase has transformed to the 9R phase.

The results of these experiments are somewhat at variance with the X-ray studies of Frolov and Rodionov [6] in which a bee+ fee transition occurred at 2.94 kbar at room temperature. The data were obtained by photographic methods, but the techniques used are uncertain; however, it is likely that the methods used were non- hydrostatic, in contrast to the hydrostatic condi- tions in the He gas pressure system used in the neutron studies. It is possible that the non- hydrostatic conditions permitted the phase change to the fee structure at the low pressures used in the above X-ray studies. And it is not inconceivable that the 9R-+fcc partial trans- formation observed in the neutron scattering studies was due to non-hydrostatic conditions at the 9R-bee interfaces [8], even though hydro- static conditions existed at the surface of the sample.

The large discrepancy in transformation pressure between the two X-ray studies-the Olinger and Shaner experiments [7] which re- ported an fee transition at 70 kbar at room tem- perature and the low pressure experiments dis- cussed above - is not resolved, but may be due to the methods of applying pressure. Clearly much work needs to be done to fill the gap between the low and high pressure regions.

In summary, the application of modest pressure increased the bee to 9R transition tem- perature significantly and revealed a partial tran- sition from the 9R structure to the fee structure which was stable over a narrow temperature

H. G. Smith et al. I Marten&tic transformation in lithium

range at the highest pressure attained in this References experiment.

55

Acknowledgements

Two of us (HGS and RB) wish to thank the IPNS staff and the Argonne National Laboratory for their kind hospitality during the course of this experiment. This research was sponsored in part by the Division of Materials Sciences, U.S. De- partment of Energy under Contract No. DE- AC05840R21400 with Martin Marietta Energy Systems, Inc. and under Contract #W-31-109- ENG-38.

111

PI [31

[41 [51 PI

H.G. Smith, Phys. Rev. Lett. 58 (1987) 1228. A.H. Daane, R.E. Rundle, H.G. Smith and F.H. Sped- ding, Acta Cryst. 7 (1954) 532. A.W. Overhausser, Phys. Rev. Lett. 53 (1984) 64. C.S. Barret, Acta Cryst. 9 (1956) 671. A.P. Frolov and K.P. Rodionov, Sov. Phys. Solid State 16 (1975) 2297.

[71 Bart Olinger and J.W. Shaner, Science 219 (1983) 1071.

PI A.L. Ruoff, private communication.

R. Berliner and S.A. Werner, Phys. Rev. B 34 (1986) 3586.