Self-propagating metathesis routes to metastable group 4 phosphides

Article abstract of DOI:10.1021/ic000057m

Group 4 phosphides, which are typically prepared at high temperatures (>800 °C) over several days, are synthesized in self-propagating metathesis (exchange) reactions in seconds. These reactions produce cubic forms of zirconium phosphide (ZrP) and hafnium phosphide (HfP) which are normally made at temperatures greater than 1425 °C and 1600 °C, respectively. To test whether the high temperatures reached in the metathesis reactions are responsible for the formation of the cubic phases, inert salts are added to lower the maximum reaction temperatures. The lower temperature reactions still result in cubic phosphides, although smaller crystallites form. Further experiments with phosphorus addition indicate that the phosphorus content is not responsible for cubic phase formation. Templating is ruled out using lattice mismatched KCl and hexagonal ZnS as additives. Therefore, the direct synthesis of the high-temperature cubic phase in metathesis reactions appears to be caused by nucleation of the metastable cubic form that is then trapped by rapid cooling. Heating the cubic phase of either ZrP or HfP to 1000 °C for 18 h, or carrying out metathesis reactions in sealed ampules at 1000 °C, results only in the hexagonal phase.

Full text of DOI:10.1021/ic000057m

Inorg. Chem. 2000, 39, 3243-3246
Self-Propagating Metathesis Routes to Metastable Group 4 Phosphides
Robert F. Jarvis, Jr., Richard M. Jacubinas, and Richard B. Kaner*
3243
Department of Chemistry and Biochemistry and Exotic Materials Institute,
University of California, Los Angeles, Los Angeles, California 90095-1569
ReceiVed January 14, 2000
Group 4 phosphides, which are typically prepared at high temperatures (>800 °C) over several days, are synthesized
in self-propagating metathesis (exchange) reactions in seconds. These reactions produce cubic forms of zirconium
phosphide (ZrP) and hafnium phosphide (HfP) which are normally made at temperatures greater than 1425 °C
and 1600 °C, respectively. To test whether the high temperatures reached in the metathesis reactions are responsible
for the formation of the cubic phases, inert salts are added to lower the maximum reaction temperatures. The
lower temperature reactions still result in cubic phosphides, although smaller crystallites form. Further experiments
with phosphorus addition indicate that the phosphorus content is not responsible for cubic phase formation.
Templating is ruled out using lattice mismatched KCl and hexagonal ZnS as additives. Therefore, the direct
synthesis of the high-temperature cubic phase in metathesis reactions appears to be caused by nucleation of the
metastable cubic form that is then trapped by rapid cooling. Heating the cubic phase of either ZrP or HfP to 1000
°C for 18 h, or carrying out metathesis reactions in sealed ampules at 1000 °C, results only in the hexagonal
phase.
Introduction
thermodynamic, hexagonal phase is obtained. Other methods
must be used to synthesize the cubic phase. Motojima et al.
The group 4 phosphides are known for their hardness and
chemical inertness even at elevated temperatures.1-6 When
zirconium phosphide (ZrP) and hafnium phosphide (HfP) are
heated above 1425 °C and 1600 °C, respectively, they each
undergo a hexagonal (h) to cubic (c) phase transition as
demonstrated by Irani and Gingerich.7 This transition may
involve loss of phosphorus.7 Hexagonal ZrP and HfP are
isostructural with h-TiP, each having a superstructure of NiAs
that can be thought of as interpenetrating slabs of NaCl-like
(ABC) and NiAs-like (BCB) sandwiches.8 Cubic ZrP and HfP
possess NaCl-type lattices. To observe the cubic phase, hex-
agonal metal phosphide samples are heated above their transition
temperatures and then quenched to room temperature.7
Other researchers have synthesized group 4 phosphides by a
variety of methods including reactions of metals with phospho-
rus,9,10 metals or metal halides with Ca3P2,3 metals with
PH3,4,5,10,11 and ZrCl4 with PCl3,6 or reduction of metal oxides
with aluminum in the presence of phosphorus and iodine.12
These synthetic methods have three points in common: (1) they
require high temperatures, usually greater than 800 °C; (2) they
generally need long reaction times of several days; and (3) the
extensive heating at elevated temperatures ensures that only the
have shown that the presence of Au, Pd, or Pt in gas-phase
reactions of ZrCl4 and PCl3 helps to control nucleation and yields
whiskers of c-ZrP.13
Solid-state metathesis reactions are capable of producing in
seconds materials that normally take days to form from the
elements.14-20 Metal halides and alkali metal chalcogenides or
pnictides react to form metal chalcogenides or pnictides and an
alkali metal halide salt that is easily washed away with alcohol
or water. The reaction, depending on the system, is initiated by
either stirring the reactants together or igniting them using a
resistively heated wire. The self-propagating reaction that ensues
can reach its peak temperature in less than 1 s, and then it rapidly
cools.14 This paper reports the application of solid-state me-
tathesis reactions using sodium phosphide (Na3P) and metal
tetrahalides (ZrCl4 or HfCl4) to produce the high-temperature
cubic phase of ZrP or HfP in seconds. An earlier survey of
metathesis reactions conducted in sealed tubes produced phos-
phides, indicating that cubic ZrP and HfP could form, but did
not explain why.20 In this paper we examine the process leading
to high-temperature metastable cubic ZrP and HfP formation.
Experimental Section
Starting Materials. All handling of air-sensitive elements and
compounds was carried out in a helium-filled drybox (Vacuum
Atmospheres) with a drytrain (Mo-40), which kept the oxygen and
(1) Gingerich, K. A. Nature 1963, 200, 877.
(2) Strotzer, E. F.; Biltz, W.; Meisel, K. Z. Anorg. Chem. 1938, 239, 216.
(3) Ripley, R. L. J. Less-Common Met. 1962, 4, 496.
(4) Samsonov, G. V.; Vereikina, L. L.; Popova, O. I. Tr. Semin.
Zharostoikim Mater., Akad. Nauk Ukr. SSR., Byull. Inst. Metallokeram.,
Spets. SplaVoV, KieV 1960, 6, 75.
(5) Blitz, W.; Rink, A.; Wiechmann, F. Z. Anorg. Chem. 1938, 238, 395.
(6) Motojima, S.; Wakamatsu, T.; Sugiyama, K. J. Less-Common Metals
1981, 82, 379.
(13) Motojima, S.; Takahashi, Y.; Sugiyama, K. J. Cryst. Growth 1975,
30, 1.
(14) Bonneau, P. R.; Jarvis, R. F., Jr.; Kaner, R. B. Nature 1991, 349,
510.
(7) Irani, K. S.; Gingerich, K. A. J. Phys. Chem. Solids 1963, 24, 1153.
(8) Snell, P.-O. Acta Chem. Scand. 1967, 27, 1773.
(9) Lundstro¨m, T.; Snell, P.-O. Acta Chem. Scand. 1967, 21, 1343.
(10) Sho¨nberg, N. Acta Chem. Scand. 1954, 8, 226.
(11) Vereikina, L. L.; Samsonov, G. V. Zh. Neorg. Khim. 1960, 5, 1888.
(12) Martin, J.; Gruehn, R. Solid State Ionics 1990, 43, 19.
(15) Wiley, J. B.; Kaner, R. B. Science 1992, 255, 1093.
(16) Rowley, A. T.; Parkin, I. P. J. Mater. Chem. 1993, 3, 689.
(17) Parkin, I. P. Chem. Soc. ReV. 1996, 25, 199.
(18) Gillan, E. G.; Kaner, R. B. Chem. Mater. 1996, 8, 333.
(19) Bonneau, P. R.; Wiley, J. B.; Kaner, R. B. Inorg. Synth. 1995, 30, 33.
(20) Hector, A. L.; Parkin, I. P. J. Mater. Chem. 1994, 4, 229.
10.1021/ic000057m CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/06/2000

3244 Inorganic Chemistry, Vol. 39, No. 15, 2000
Jarvis et al.
Table 1. Summary of Group 4 Phosphide Metathesis Reactions
X-ray
reaction
conditions
diffraction
figure
reactants
products
h-TiP
c-ZrP
c-HfP + h-HfP
c-ZrP
TiI4 + 4/3 Na3P
ZrCl4 + 4/3 Na3P
HfCl4 + 4/3 Na3P
ZrCl4 + 4/3 Na3P
ignited
1a
1b
1c
2a
2b
ignited
ignited
ignited
ZrCl4 + 4/3 Na3P + ignited
3 mol equiv NaCl
c-ZrP
ZrCl4 + 4/3 Na3P + ignited
6 mol equiv NaCl
c-ZrP
c-ZrP
c-ZrP
c-ZrP
2c
ZrCl4 + 4/3 Na3P + ignited
3 mol equiv KCl
ZrCl4 + 4/3 Na3P + ignited
3 mol equiv ZnS
ZrCl4 + 4/3 Li3P +
6 mol equiv NaCl
TiI4 + 4/3 Na3P
ignited
1000 °C, 18 h
1000 °C, 18 h
1000 °C, 18 h
stirred
h-TiP
h-ZrP
h-HfP
HfCl4 + Na3P
amorphous
3a
3b
3c
4a
4b
4c
4d
ZrCl4 + 4/3 Li3P
HfCl4 + 4/3 Na3P
HfCl4 + 4/3 Na3P
HfCl4 + 4/3 Na3P
HfCl4 + 4/3 Na3P
HfCl4 + 4/3 Na3P
Figure 1. Powder X-ray diffraction patterns of ignition metathesis
reactions between Na3P and TiI4, ZrCl4, or HfCl4 forming (a) h-TiP,
(b) c-ZrP, and (c) c-HfP + h-HfP, respectively. The Miller indices for
h-TiP and c-ZrP are given in (a) and (b), respectively, and * ) c-HfP
in (c).
ground
stirred, ignited c-HfP + h-HfP
ground, ignited c-HfP +
trace h-HfP
Results
moisture contents to <1 ppm. Na3P was synthesized by reacting a 3:1
ratio of sodium and phosphorus in an evacuated Pyrex tube that was
heated for 30 min at 170 °C, 30 min at 350 °C, and 5 h at 480 °C. Li3P
was synthesized by combining lithium and phosphorus in a tantalum
boat inside an evacuated Pyrex tube. The tube was then placed in a
furnace set at 150 °C and the temperature was raised to 350 °C over a
period of 5 h, followed by a final heating at 400 °C for 4 h. KCl
(Aldrich, 99.99+%), NaCl (Fisher, certified ACS), and ZnS (Johnson
Matthey, 99.999%) were used as received. Reagent grade TiI4 (Alfa,
99%), TiCl3 (Fluka, >97%), ZrCl4 (Aldrich, 99.9%), and HfCl4
(Johnson Matthey, 99%) were purified using vapor transport in a 300
°C to room-temperature gradient. This purification allowed separation
of the tetrahalide from lower halides and oxyhalides as described by
Scha¨fer.21
Solid-state metathesis reactions forming group 4 phosphides
were ignited with a hot wire or carried out in sealed quartz tubes.
These reactions follow the form shown in Eq 1 where M ) Ti,
Zr, or Hf and X ) halogen.
MX4 + 4/3Na3P f MP + 4NaX + 1/3P
(1)
Both the Na3P and the metal halide starting reagent are solids.
Ignition with a hot wire initiates a fast, self-propagating reaction
which is followed by rapid cooling over several minutes. A
summary of the group 4 phosphide metathesis reactions carried
out in this study is presented in Table 1.
TiI4, ZrCl4, and HfCl4 were reacted with Na3P in ignition
reactions forming h-TiP, c-ZrP, and a mixture of c-HfP and
h-HfP, respectively (Figure 1). Adding 3 and 6 mol equiv
(approximately 0.5 and 1 wt equiv) respectively, of NaCl to
the ZrP ignition reactions caused line broadening in the X-ray
patterns, but the resulting products were still c-ZrP (Figure 2).
Using the Scherrer equation,22 the ZrP crystallite size was
calculated to be 760 Å for the regular ignition reaction (Figure
2a), 250 Å when 3 mol of NaCl was added (Figure 2b), and
140 Å when 6 mol of NaCl was added to the reaction (Figure
2c). The addition of either 3 mol of KCl or 3 mol of ZnS to the
reaction between ZrCl4 and Na3P still resulted in formation of
c-ZrP exclusively. The average crystallite size was calculated
to be 320 Å with KCl addition and 290 Å with ZnS addition.
Reactions between Li3P and ZrCl4 also led only to cubic ZrP.
Metal halides were also reacted with Na3P in sealed tubes at
1000 °C for 18 h producing only crystalline hexagonal
products: h-TiP, h-ZrP, and h-HfP (Figure 3). In addition, when
c-ZrP from an ignition reaction was heated in a sealed quartz
tube for 18 h at 1000 °C, X-ray diffraction showed that it had
converted completely to h-ZrP.
Metathesis Reactions. Sodium phosphide and the appropriate metal
halide, salt-balanced on a molar scale, were each ground separately
with a mortar and pestle and then either stirred or ground together before
reaction. Ignition reactions were carried out in a stainless steel reaction
vessel (50-mm height, 48-mm o.d., 35-mm i.d.)19 typically using 0.5 g
of total starting reagents. When red phosphorus was added to the
reaction it was ground in with the Na3P. Half of the NaCl (or KCl or
ZnS) added as a heat sink was ground with the Na3P and half with the
metal halide. A resistively heated Nichrome wire was used to initiate
the reactions. Sealed-tube reactions were run in evacuated quartz tubes
for 18 h at 1000 °C using 1.0 g of starting reagents and 2.0 g of NaCl
as an inert additive to prevent ignition of the reactants. Products of all
reactions were washed with methanol, water, and ethyl ether before
being dried under dynamic vacuum for at least 1 h.
CAUTION! Unreacted Na3P reacts Violently with water and can
ignite methanol.
Characterization. Powder X-ray diffraction patterns were taken
using a Crystal Logic diffractometer with Ni-filtered Cu KR radiation.
X-ray patterns were scanned with 0.1° 2θ steps at 2 s/step. Lattice
constants were calculated using linear regression on patterns taken with
0.02° 2θ steps. Line-broadening analysis was performed on the (200),
(220), and (311) peaks of ZrP using the Scherrer equation to calculate
the average crystallite size.22 Silicon was used as an internal standard
for both lattice-constant and line-broadening studies. Polypropylene film
was used to cover air-sensitive X-ray samples.
The reagent mixture preparation affects the products formed
in a reaction (Figure 4). The X-ray pattern of crystalline HfCl4
and Na3P stirred together is shown in Figure 4a (note: no
reaction has taken place). Grinding HfCl4 and Na3P together
leads to a loss in crystallinity (Figure 4b). Running an HfP
ignition reaction after the reagents were stirred together, after
being ground individually, formed an approximately equal
(21) Scha¨fer, H. Chemical Transport Reactions; Academic Press: New
York, 1964.
(22) Cullity, B. O. Elements of X-ray Diffraction; Addison-Wesley:
Reading, MA, 1956.

Self-Propagating Group 4 Phosphides Metathesis
Inorganic Chemistry, Vol. 39, No. 15, 2000 3245
Figure 2. Expanded powder X-ray diffraction patterns of the (111)
and (200) peaks of c-ZrP formed in ignition metathesis reactions with
the addition of (a) no NaCl, (b) 3 mol equiv of NaCl (∼0.5 wt equiv),
and (c) 6 mol equiv of NaCl (∼1 wt equiv) to the starting reactants.
Figure 4. Powder X-ray diffraction patterns of HfCl4 and Na3P (a)
stirred together, (b) ground together, (c) stirred together and ignited,
and (d) ground together and ignited.
Table 2. Selected Group 4 Phosphide Lattice Parameters
product
lattice constant (Å)
literature values (Å)
ignited h-TiP
a ) 3.498(1),
c ) 11.70 (1)
a ) 5.275(1)
a ) 3.651(2),
c ) 12.38(2)
a ) 5.213(2)
a ) 3.501(1),
c ) 11.69(1)
a ) 3.691(1),
c ) 12.55(1)
a ) 3.650(4),
c ) 12.39(2)
a ) 3.499(1),
c ) 11.700(6)8
a ) 5.2723
ignited c-ZrP
ignited h-HfP
a ) 3.650,
c ) 12.37 24
ignited c-HfP
sealed tube h-TiP
a ) 3.499(1),
c ) 11.700(6)8
a ) 3.677,
sealed tube h-ZrP
sealed tube h-HfP
c ) 12.5223
a ) 3.650,
c ) 12.3724
tography of an ignition reaction forming MoS2 from MoCl5 and
Na2S has shown that the peak temperature of a reaction can be
reached in approximately one-third of a second, followed by
rapid cooling of the products.14 It is likely that the analogous
short reaction times and rapid cooling are responsible for
quenching the phosphide reactions, thus preserving the cubic
phase. Sealed tube metathesis reactions carried out at 550 °C
for 4 h also produce the metastable cubic phase.20 This is likely
to be due to the reaction igniting to create the cubic phase and
the temperature (550 °C) and time (4 h) being insufficient to
cause conversion to the thermodynamically stable hexagonal
form.
Figure 3. Powder X-ray diffraction patterns of sealed-tube metathesis
reactions forming (a) h-TiP, (b) h-ZrP, and (c) h-HfP.
mixture of c-HfP and h-HfP (Figure 4c). However, if the
reagents were ground together in a mortar and pestle after
individual grinding, an ignition reaction then formed mainly
c-HfP with only a trace amount of h-HfP (Figure 4d).
A summary of all of the metathesis reactions performed is
provided in Table 1. Most products were found, by X-ray
diffraction, to be pure crystalline phosphides, and their lattice
constants were in good agreement with values published in the
literature,8,24,25 as presented in Table 2.
The process by which these high-temperature cubic phases
form is a question that needs to be addressed. Four possible
explanations are considered. First and simplest, these reactions
could be exothermic enough to reach the high temperatures at
which the phase transition occurs. Second, the short reaction
times in which the ignition reactions take place may prevent
sufficient phosphorus from diffusing into the zirconium, thereby
forming a phosphorus deficient cubic compound.7 Third, the
byproduct salt could assist in the crystallization of the cubic
phases via templating. Finally, the conditions present in the
ignition reaction may cause the cubic phase to nucleate first,
followed by rapid quenching of the system. These four pos-
sibilities were investigated to determine which most likely causes
the formation of metastable c-ZrP and c-HfP.
Discussion
As with other literature syntheses of group 4 phosphides,
sealed-tube metathesis reactions run at 1000 °C produce only
the low-temperature hexagonal phase as expected (Figure 3).
However, ignition reactions can form the cubic phase due to
the manner in which these reactions take place (Figure 1). These
highly exothermic reactions are very rapid. High-speed pho-
(23) Scha¨fer, H. Acta Chem. Scand. 1954, 8, 226.
(24) Jeitschko, N. Monatsh. Chem. 1962, 93, 1107.
(25) Unfortunately, an exact calculation could not be performed for the
reactions given in eq 1 because no thermodynamic data are available
for TiP, ZrP, or HfP.
Using the calculated heat of reaction, heat capacities, heats
of fusion, and heats of vaporization of the products, a theoretical
maximum reaction temperature can, in principle, be calculated

3246 Inorganic Chemistry, Vol. 39, No. 15, 2000
Jarvis et al.
for any metathesis reaction.14 Reactions involving group 4
halides are typically predicted to reach the boiling point of the
salt produced in the reaction.25 In this case, the boiling point of
the byproduct salt, NaCl, is 1413 °C.26 This is very close to the
phase transition temperature reported for ZrP, which is 1425
°C. To see if the maximum reaction temperature causes the
formation of the cubic phase, NaCl was added to the system as
a heat sink. Results of the salt additions are shown in Figure 2.
Energy given off in the reaction now has to heat both the
products and the heat sink, thereby lowering the maximum
temperature reached in the reaction.14 Because there is now less
heat present to sinter the ZrP crystallites to larger sizes, the
ZrP X-ray peaks have broadened. Note, however, that the
product is still c-ZrP. Therefore, it can be concluded that the
maximum reaction temperature is not responsible for the
formation of the cubic phase.
Because cubic phase formation is not due to the maximum
temperature, phosphorus deficiency was investigated next. Irani
and Gingerich showed that their c-ZrP was relatively stable on
heating in the absence of phosphorus, but converted to h-ZrP
upon heating in the presence of phosphorus. This suggested that
ignition metathesis reactions might not allow for sufficient
diffusion of phosphorus to the zirconium resulting in cubic ZrP.
However, when 0.25 g of red phosphorus was added to a 0.5 g
ZrP ignition reaction, only c-ZrP was produced. Adding the extra
phosphorus should alleviate the problem of insufficient phos-
phorus diffusion in the reaction, if it exists. In the presence of
sufficient phosphorus the system still nucleates the high-
temperature cubic phase. It can be concluded that, although
phosphorus deficiency may be associated with c-ZrP, it is not
responsible for its formation in metathesis reactions.
then stirred together and ignited, approximately equal amounts
of c-HfP and h-HfP are observed in the product (Figure 4c).
However, if the starting materials are ground separately, then
ground vigorously together and ignited, the product is almost
exclusively c-HfP (Figure 4d). The explanation for these results
becomes clearer upon closer examination of the reactants before
ignition. An X-ray pattern taken of the reactants just after stirring
shows only peaks for HfCl4 and Na3P (Figure 4a). But the
pattern of the reactants ground together is different in two
obvious ways (Figure 4b). First, the diffraction peaks of the
reactants are greatly reduced in intensity if not altogether
eliminated. Second, two diffraction peaks emerge at ∼26° and
∼32° 2θ that can be assigned to the (111) and (200) lines of
NaCl. From these two observations it can be concluded that
the starting material lattices begin breaking down and the
product salt, NaCl, starts to form. The formation of NaCl during
the grinding stage lowers the amount of heat given off during
ignition by forming some product ahead of time. Less heat given
off with the ignition reaction means that the product is heated
less and quenched sooner. This is why more c-HfP is observed
and the diffraction peaks are broadened when the precursors
are ground separately and then ground vigorously together and
ignited.
The compound TiP has been predicted to undergo a hexagonal
to cubic transition at a lower temperature than ZrP (1425 °C)
does.7 This study does not support this prediction. Both ignition
and sealed tube metathesis reactions between TiI4 and Na3P form
only h-TiP. No evidence was found in any of the TiP reactions
that a phase transition takes place in this system. Additionally,
a reaction between TiCl3 and Na3P produced exclusively
hexagonal TiP, thereby ruling out a templating effect between
the byproduct salt NaCl and cubic TiP.
A third possibility is that the byproduct salt is assisting in
the crystallization of the cubic phase via templating. Because
there is only a 6.6% lattice mismatch between the lattice
parameters of the byproduct salt NaCl (a ) 5.64 Å)26 and the
product c-ZrP (a ) 5.27 Å)14 and the requirement for some
epitaxial growth is a mismatch of e15%, templating is a
possibility.27 When Li3P is substituted for Na3P to produce the
byproduct salt LiCl (a ) 5.14 Å),13 the result is still cubic ZrP;
this is consistent because the lattice mismatch is only 2.5%. To
further test the templating hypothesis, the salt KCl (a ) 6.29
Å)13 was added to the reaction. KCl has a 16% lattice mismatch
with c-ZrP and would therefore not be expected to assist in the
crystallization of the cubic phase. The reaction still yielded only
cubic ZrP. As a further test, hexagonal ZnS was added to the
ZrCl4/Na3P reaction. Hexagonal ZnS would certainly not be
expected to help form the cubic phase and might reasonably be
expected to assist in the formation of the hexagonal phase.
Interestingly, cubic ZrP was still the exclusive phosphide phase
and the crystallite size decreased to 290 Å as may be expected
for an inert additive. Hence, templating of the cubic phase can
be ruled out.
Conclusions
Solid-state metathesis reactions can be used to rapidly
synthesize metastable cubic phases of transition-metal phos-
phides. In contrast, sealed tube reactions between metal halides
and sodium phosphide, at 1000 °C, lead exclusively to the low-
temperature hexagonal phase. Salts were added to lower the
maximum reaction temperatures to check whether the high
temperatures reached in the rapid metathesis reactions were
responsible for cubic phase formation. The lower temperatures
produced by salt addition resulted in lower crystallinity, but still
produced only the cubic phase. Because phosphorus deficiency
can lead to cubic phases, excess phosphorus was added to the
ignited metathesis reactions; however, the result was still
exclusively the cubic phase. To test a possible templating effect,
compounds with lattice mismatches greater than the 6.6%
mismatch of NaCl were added to the metathesis reactions. The
addition of KCl with a 16% lattice mismatch still resulted in
cubic ZrP. Even the addition of hexagonal ZnS led only to cubic
ZrP formation, essentially ruling out a templating effect. Hence,
it appears most likely that the high-temperature cubic phases
of ZrP and HfP are nucleated first and then quenched rapidly
before the cubic phase can convert to the low-temperature
hexagonal phase.
It can be concluded that the high-temperature phases are most
likely nucleating first and retained due to the rapid quenching
of the system because neither maximum reaction temperature,
phosphorus deficiency, nor a templating effect is responsible
for cubic phase formation. Further evidence to support this
conclusion comes from the HfP system when the reaction
conditions are varied. If HfP and Na3P are ground separately,
Acknowledgment. The authors thank Professor John B.
Wiley, Dr. Philippe R. Bonneau, and Dr. Randolph E. Treece
for their insightful discussions regarding this work. This research
was funded by the National Science Foundation and the
University of California Campus Laboratory Collaboration
Program.
(26) CRC Handbook of Chemistry and Physics; West, R. C., Ed.; CRC
Press: Boca Raton, FL, 1985.
(27) West, A. R. Solid-State Chemistry and Its Applications; John Wiley
and Sons: New York, 1984.
IC000057M