862
MAGNETIC STRUCTURE OF PrB6
Vol. 36, No. 10
stat using a differential heating technique with
a calibrated germanium thermometer in a man-
ncr similar to that discussed earlier.4 No cx-
change gas w as used. T he precision of the
the wavevectors associated with these intensities~
These measurements were made on a triple axis
instrument. 3XE, at MURR w hich w as operated
in the constant energy elastic made, AE = 0.
M onochrom ator and analyzer cry stals w ere
both Cu(200) and yielded a beam of 1.066
neutrons on the sam ple w ith a flux of 6 x iAo6
n e u tro n s/c m 2 /se c .
m easu rem en ts,
based on run to run rep eata-
bility for pure copp er, is estim ated to be 0. 5%.
Due to the sm all size of the sam ple the accu-
racy of the m easu rem en ts repo rted h e re is e sti-
m tted to be n ear ± 2%.
Ill. R esults
B. E lectrical R esistivity and T herm al
E xpansion
A . Specific H eat
The e le c tric a l resistiv ity w as m easu red by
conventional 4-probe m ethod at 220 H z. E lec-
F ig u re 1 shows the m easured specific heat
of P r3 6 as a function of tem peratu re in the
range 1.7 K to 15 K. The prom inent features
a re the la rg e peak at 6.89 ± 0 .0 5 K for w hich
a
tric a l contact to the single cry stal w as m ade via
spot-w elded 0.002
of the low er anom aly in the resistiv ity v aried
from cry stal to cry stal. We believe th is v aria-
Pt w ire.
The tem p eratu re
the peak value of
C w as 81.6 Jo ule/m o le-K and
a m uch sm aller, but n ev erth eless d istin ct, peak
at 4 .2 2 ± 0 .0 5 K. The larg er peak can be id e n ti-
fied w ith the N ~ e l tem peratu re and w as rep orted
tion is due to ex trem e stra in sensitiv ity of the
low er transition, and this sensitiv ity req u ires
that care be taken to attach the leads using as
sm all a disch arge as possible.
T he the rm al expansion m easu rem en ts w ere
done using a capacitance d ilato m eter d escrib ed
e a rlie r by L ee.
I
H ow ever, th is peak is found
to be m uch higher and sh arp er in the present
w ork (peak height 81.6 Jfm ole-K com pared to
30 J/m o le-K ).
The peak at 4.22 K appears to
elsew h ere.
5
be c o rre la te d with the anom alies seen in the
the rn ial expansion and the electrical resistiv ity .
An attem pt was m ade to estim ate the e n -
C. N eutron S cattering M easurem ents
tro py change for the tran sitio n a t 6. 9 K. A plot
of C /T vs T w as extrapolated to give zero m a g -
netic contribution at 20 K and the sm all peak a t
N eutron scattering m easu rem en ts w ere
m ade on a larg e polycrystalline sam ple and on
a sm all single cry stal having a m ass of IS m g.
The polycrystallinc m aterial was p rep ared by
4. 2 K w as elim inated.
G raphical integration of
R ln(3. 4 ± 0.2).
the data then gave the re su lt ~S
the reduction of P r203 by B4C at about 1770
under high vacuum . In all cases, both w ith
K
A dditional u n certain ties in th is resu lt rem ain
because of unknow n o r poorly d eterm ined c o r-
rectio n s for lattice vibrations, electro n ic effects
poL ycrystalline and single cry stal sam p les, the
boron used in the prep arations w as highly en-
riched in the isotope I 111 (98. 5%) in o rd er to
m inim ize the effect of absorption of the rm al
neutrons by the isotope 10i~ .
and Schottky anom alies.
The value found here
for ~S is, how ever, totally consistent with the
expected value for a r5 triplet ground state.
Pow der sam p les w ere encapsulatec in a thin
w all alum inum sam ple holder suitably m ounted
B. E lectrical R esistivity and T herm al
E xpansion
in a helium cry o stat.
(O ur ex p erim en tal c o n d i-
tions requ ired that m easu rem en ts above 4. 2 K
and m easu rem en ts below 4. 2 K be perform ed in
T he electrical resistiv ity at low tem p era-
ture of a PrB 6 single cry stal is show n in Fig
-
d ifferen t cry o stats.
)
D iffraction data w as oh-
ore 2. A part from the sharp anom aly a t a p p ro x -
im ately 6. 95 K, th ere is a clear, sharp d ro p in
the resistiv ity a t approxim ately 4. 15 K. D uring
the m easu rem en t for w hich the data is show n,
the cooling of the sam ple w as stopped at 4.02 K,
and the sam ple w as allow ed to w arm to 4. 35 K,
before continuing cooling below 4 K. T here is
clear h y ste re sis present, w hich can be seen in
tam ed on ZXD, a double axis instrum ent at
M URR (M issouri U niversity R esearch R eactor),
using neutrons of w avelength 1. 293 A w ith
flux on sam ple of 2 x 106 n cu tro n s/cm 2 /sec.
D uring the accum ulation of data the sam ple w as
rotated in o rd e r to m inim ize the effects of pro-
ferred orientation,
a
G uided by the resu lts of the pow der data,
diffraction scans w ere perform ed on a sm all
single cry stal in o rd er to determ ine uniquely
the w ave vectors associated w ith the m agnetic
the figure.
Above 4. 30 K, the w arm ing and
cooling data coincide w ithin experim ental a c c u -
racy .
T he th erm al expansion data shown in F ig-
ure 3 v ery clearly shows two anom alies, peak-
ing at 6. 95 K and 3. 95 K, resp ectiv ely .
in ten sities.
The effects of sam ple absorp tio n .
extinction and problem s we encountered in o n -
enting the crystal in a sub 4. 2 K environm ent,
m ade it d ifficu lt to obtain m eaningful relative
intensity data for the single cry stal; how ever, it
w as possible to m ake p recise m easu rem en ts of
C. N eutron D iffraction
N eutron scattering m easu rem en ts m ade in