Rapid solid-state synthesis of tantalum, chromium, and molybdenum nitrides

Article abstract of DOI:10.1021/ic001265h

Solid-state metathesis (exchange) reactions can be used to synthesize many different transition-metal nitrides under ambient conditions including TiN, ZrN, and NbN. Typical metathesis reactions reach temperatures of greater than 1300 °C in a fraction of a second to produce these refractory materials in highly crystalline form. Likely due to the large amount of heat produced in these solid-state reactions, some transition-metal nitrides such as TaN, CrN, and γ-Mo₂N cannot easily be synthesized under ambient conditions. Here metathesis reactions are demonstrated to produce the cubic nitrides TaN, CrN, and γ-Mo₂N when sufficient pressure is applied before the reaction is initiated. By pressing a pellet of TaCl₅ and Li₃N with an embedded iron wire, crystalline cubic TaN forms under 45 kbar of pressure after a small current is used to initiate the chemical reaction. Crystalline cubic CrN is synthesized from CrCl₃ and Li₃N initiated under 49 kbar of pressure. Crystalline γ-Mo₂N is produced from MoCl₅ and Ca₃N₂ (since MOCl₅ and Li₃N self-detonate) initiated under 57 kbar of pressure. The addition of ammonium chloride to these metathesis reactions drastically lowers the pressure requirements for the synthesis of these cubic nitrides. For example, when 3 mol of NH₄Cl is added to CrCl₃ and Li₃N, crystalline CrN forms when the reaction is initiated with a resistively heated wire under ambient conditions. Cubic γ-Mo₂N also forms at ambient pressure when 3 mol of NH₄Cl is added to the reactants MoCl₅ and Ca₃N₂ and ignited with a resistively heated wire. A potential advantage of synthesizing γ-Mo₂N under ambient conditions is the possibility of forming high-surface-area materials, which could prove useful for catalysis. Nitrogen adsorption (BET) indicates a surface area of up to 30 m²/g using a Langmuir model for γ-Mo₂N produced by a metathesis reaction at ambient pressure. The enhanced surface area is confirmed using scanning electron microscopy.

Full text of DOI:10.1021/ic001265h

2
240
Inorg. Chem. 2001, 40, 2240-2245
Rapid Solid-State Synthesis of Tantalum, Chromium, and Molybdenum Nitrides
Jennifer L. O’Loughlin, Charles H. Wallace, Meredith S. Knox, and Richard B. Kaner*
Department of Chemistry and Biochemistry and Exotic Materials Institute, University of California,
Los Angeles, Los Angeles, California 90095-1569
ReceiVed September 26, 2000
Solid-state metathesis (exchange) reactions can be used to synthesize many different transition-metal nitrides
under ambient conditions including TiN, ZrN, and NbN. Typical metathesis reactions reach temperatures of greater
than 1300 °C in a fraction of a second to produce these refractory materials in highly crystalline form. Likely due
to the large amount of heat produced in these solid-state reactions, some transition-metal nitrides such as TaN,
CrN, and γ-Mo2N cannot easily be synthesized under ambient conditions. Here metathesis reactions are
demonstrated to produce the cubic nitrides TaN, CrN, and γ-Mo2N when sufficient pressure is applied before the
reaction is initiated. By pressing a pellet of TaCl5 and Li3N with an embedded iron wire, crystalline cubic TaN
forms under 45 kbar of pressure after a small current is used to initiate the chemical reaction. Crystalline cubic
CrN is synthesized from CrCl3 and Li3N initiated under 49 kbar of pressure. Crystalline γ-Mo2N is produced
from MoCl5 and Ca3N2 (since MoCl5 and Li3N self-detonate) initiated under 57 kbar of pressure. The addition of
ammonium chloride to these metathesis reactions drastically lowers the pressure requirements for the synthesis
of these cubic nitrides. For example, when 3 mol of NH4Cl is added to CrCl3 and Li3N, crystalline CrN forms
when the reaction is initiated with a resistively heated wire under ambient conditions. Cubic γ-Mo2N also forms
at ambient pressure when 3 mol of NH4Cl is added to the reactants MoCl5 and Ca3N2 and ignited with a resistively
heated wire. A potential advantage of synthesizing γ-Mo2N under ambient conditions is the possibility of forming
high-surface-area materials, which could prove useful for catalysis. Nitrogen adsorption (BET) indicates a surface
2
area of up to 30 m /g using a Langmuir model for γ-Mo2N produced by a metathesis reaction at ambient pressure.
The enhanced surface area is confirmed using scanning electron microscopy.
Introduction
the metal in the presence of nitrogen or ammonia at elevated
1
,15-17
temperatures.
high-temperature synthesis (SHS).
An alternate method is self-propagating
Cubic-phase transition-metal nitrides are of interest due to
their hardness (8-9 on Moh’s scale) and high melting points
1
8-21
The SHS method in-
1
volves the ignition of high-surface-area metal powders under
high nitrogen pressure or in the presence of sodium azide.
Although this method produces nitrides rapidly, unreacted or
undernitrided metals can be a problem. Another rapid method
for synthesizing transition-metal nitrides uses metal halides and
alkali-metal or alkaline-earth-metal nitrides in metathesis (ex-
(
2000-4000 °C). These nitrides are resistant to chemical attack
2
even at elevated temperatures. Potential applications include
3,4
protective coatings for cutting tools and ultra-high-vacuum
system components. Recently, molybdenum nitrides have been
5
found to be useful as heterogeneous catalysts in the synthesis
6
7-9
of ammonia and for hydrodenitrogenation and hydrodesulfur-
2
2-24
ization10-13
change) reactions.
These solid-state metathesis (SSM)
reactions. In addition, chromium nitride coated steels
1
4
reactions are self-propagating and driven by the formation of
for corrosion protection are now commercially available.
The cubic group 4 (Ti, Zr, Hf) and group 5 (V, Nb) transition-
metal mononitrides (MN1.0) are generally synthesized by heating
22-26
stable salt byproducts.
After initiation at room temperature,
these reactions often reach temperatures of >1300 °C in less
than 1 s and cool rapidly. This method is especially effective
at synthesizing the mononitrides of groups 4 (Ti, Zr, Hf) and 5
(
(
(
1) Toth, L. E. Transition metal carbides and nitrides; Academic Press:
New York, 1971.
2) Blocher, J. M., Jr. In High-temperature technology; Campbell, I. E.,
Ed.; Wiley: New York, 1956; pp 171-186.
3) Buhl, R.; Pulker, H. K.; Moll, E. Thin Solid Films (Switzerland) 1980,
(15) Blumenthal, W. B. The chemical behaVior of zirconium; Van Nos-
trand: Princeton, NJ, 1958.
8
0, 265-270.
(16) Fabbricatore, P.; Musenich, R.; Occhetto, M.; Parodi, R.; Pompa, P.
IEEE Trans. Magn. 1991, 27, 1291-1294.
(
(
(
(
(
4) Sundgren, J.-E. Thin Solid Films (Switzerland) 1985, 128, 21-44.
5) Moriyama, K. Mater. Res. Soc. Bull. 1990, 15, 32-4.
6) Volpe, L.; Boudart, M. J. Phys. Chem. 1986, 90, 4874-7.
7) Oyama, S. T. Catal. Today 1992, 15, 179-200.
8) Choi, J. G.; Brenner, J. R.; Colling, C. W.; Demczyk, B. G.; Dunning,
J. L.; Thompson, L. T. Catal. Today. 1992, 15, 201-222.
9) Schlatter, J. C.; Oyama, S. T.; Metcalfe, J. E., III; Lambert, J. M., Jr.
Ind. Eng. Chem. Res. 1988, 27, 1648-53.
(17) Brauer, V. G.; Esselborn, R. Anorg. Allg. Chem. 1961, 309, 151-70.
(18) Merzhanov, A. G.; Borovinskaya, I. P.; Volodin, Y. E. DANKAS 1972,
206, 905-8.
(19) Munir, Z. A.; Holt, J. B. J. Mater. Sci. 1987, 22, 710-714.
(20) Holt, J. B. (U.S. Department of Energy). Refractory Materials. U.S.
Patent 4,459,363, July 10, 1984.
(21) Holt, J. B. (U.S. Department of Energy). Synthesis of Refractory
Materials. U.S. Patent 4,446,242, May 1, 1984.
(22) Wiley, J. B.; Kaner, R. B. Science 1992, 255, 1093-1097.
(23) Gillan, E. G.; Kaner, R. B. Inorg. Chem. 1994, 33, 5693-5700.
(24) Fitzmaurice, J. C.; Hector, A. L.; Parkin, I. P. J. Chem. Soc., Dalton
Trans. 1993, 2435-2438.
(25) Parkin, I. P.; Hector, A. J. Mater. Sci. Lett. 1993, 12, 1856-1857.
(26) Hector, A. L.; Parkin, I. P. Polyhedron 1995, 14, 913-917.
(
(
(
(
(
10) Markel, E. J.; Vanzee, J. W. J. Catal. 1990, 126, 643-657.
11) Abe, H.; Cheung, T. K.; Bell, A. T. Catal. Lett. 1993, 21, 11-18.
12) Nagai, M.; Miyao, T.; Tuboi, T. Catal. Lett. 1993, 18, 9-14.
13) Liaw, S. J.; Raje, A.; Bi, X. X.; Eklund, P. C.; Graham, U. M.; Davis,
B. H. Energy Fuels 1995, 9, 921-927.
(
14) Su, Y. L.; Yao, S. H.; Leu, Z. L.; Wei, C. S.; Wu, C. T. Wear 1997,
213, 165-174.
1
0.1021/ic001265h CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/15/2001

Solid-State Synthesis of TaN, CrN, and MoN
V, Nb), most likely because they are thermodynamically stable
Inorganic Chemistry, Vol. 40, No. 10, 2001 2241
(
to have wider applicability, enabling the synthesis of the group
in this temperature range, as indicated in the appropriate phase
diagrams.
5 nitride TaN and the group 6 nitrides CrN and γ-Mo2N.
2
7,28
Experimental Section
The other cubic transition-metal nitride of group 5 (i.e., TaN)
and group 6 cubic nitrides (e.g., CrN and γ-Mo2N) are not as
easy to synthesize by solid-state metathesis reactions because
the temperature that the metathesis reactions reach is higher
Caution! Most solid-state metathesis reactions are highly exothermic,
and in some cases the precursors may spontaneously detonate when
mixed or ground together. Care should be taken to do reactions of this
type on a small scale first (less than 1 g of total reactant mixture) with
adequate safety precautions. These reactions may ignite when exposed
to small amounts (one drop) of a solvent such as water or methanol. It
is also important to calculate the pressure of the nitrogen gas byproduct
before performing any reactions in closed vessels, since high gas
pressures can lead to explosions.
2
7,28
than the decomposition temperatures of these nitrides.
Attempts to use metathesis reactions to synthesize group 6
nitrides have been reported recently in the literature.29,30 A new
phase of Mo2N with the Mo2C structure type was synthesized
by a metathesis reaction between MoCl5 and Ca3N2 performed
in a CaCl2 melt.29 Solid-state metathesis reactions between CrCl2
and Mg3N2, with the addition of MgCl2 as a diluent, can produce
CrN when heated at 350-500 °C.30
5 5
Reagents. The metal halides TaCl (Aldrich, 99.9%) and MoCl
(
Strem, 99.5%) were purified prior to use by vapor transport in sealed,
-4
evacuated (10 Torr) Pyrex tubes across a temperature gradient from
The cubic transition-metal nitride systems appear to require
higher nitrogen pressures than the reactant nitrides (i.e., Li3N,
Ca3N2, and/or NaN3) typically used in metathesis reactions can
provide on decomposition.31 The metathesis reaction between
tantalum pentachloride and lithium nitride to produce cubic TaN
illustrates this principle. In conventional methods, the synthesis
of cubic TaN requires temperatures >1700 °C and g160 bar
of nitrogen.32 The metathesis reaction between tantalum pen-
tachloride and lithium nitride
1
90 °C (for TaCl
purification enabled separation of the pentachloride from lower halides
and oxyhalides as described by Sch a¨ ffer.34 The compounds CrCl
(Aldrich, 99.9%), Li N (Cerac, 99.5%), and Ca (Cerac, 99.5%) were
used as received. Chromium iodide (CrI ) was synthesized by heating
Cr metal (Cerac, -325 mesh, 99.2%) and solid I (Strem) for 12 h in
a sealed, evacuated Pyrex tube across a temperature gradient from 170
to 450 °C. Ammonium chloride (NH Cl) (Baker Chemical Co.) was
5 5
) or 200 °C (for MoCl ) to room temperature. This
3
3
3 2
N
3
2
4
heated to 200 °C under dynamic vacuum with an inline cold trap to
remove any adsorbed water.
Synthesis. All reactions were performed in a helium-filled drybox.
The finely divided reactants were ground together (see specific reactions
below for the order of mixing) in an agate mortar and pestle. The
5
1
TaCl + / Li N f TaN + 5LiCl + / N
2
(1)
5
3
3
3
reactants were then transferred to a 45 mL stainless steel reaction vessel
under ambient conditions produces only the subnitride, Ta2N,
rather than the cubic mononitride, TaN. However, by adding
35
(non-air-tight) modeled after a bomb calorimeter. Reactions were
restricted to small scales (approximately 1 g of total reactants) since
high temperatures and high pressures are often generated. In the γ-Mo
1
2 mol of sodium azide (NaN3) to the above reaction, a
2
N
significant nitrogen overpressure is created, which enables the
synthesis, 4 mmol (∼1 g) of MoCl was used with a stoichiometric
5
2
3
cubic phase (TaN) to form along with hexagonal Ta2N.
3 2 4
amount of Ca N and 3 mol of NH Cl (referred to as the ambient-
Although the formation of high-pressure phases has been
observed in the products of ambient-pressure metathesis reac-
tions, these products have always been contaminated with a
significant fraction of subnitrides (i.e., Ta2N, Cr2N) and/or metal
pressure reaction). MoCl
before the addition of Ca
MoCl with Ca or Li N directly, as this mixture will spontaneously
5
was initially ground together with NH
4
Cl
3
N . Caution! Never grind freshly transported
2
5
N
3 2
3
detonate. The ambient-pressure reactions that produce TaN and CrN
were performed using the same procedure except that the metal halide
(Cr, Mo). This is observed for tantalum nitride (TaN with Ta2N
5 3 3 3 4
(TaCl , CrCl , or CrI ) was ground together with both Li N and NH -
impurities), chromium nitride (CrN with Cr2N and Cr impuri-
ties), and molybdenum nitride (γ-Mo2N with Mo impurities).
To synthesize the high-pressure and/or high-temperature
phases of chromium, molybdenum, and tantalum nitride directly,
at least one of two parameters must be altered to favor the cubic
phases. Specifically, either sufficient pressure needs to be
applied to the reactions or the temperature of the reaction needs
to be lowered to a regime where these compounds become
thermodynamically favorable. In this paper, we report the
successful synthesis of the cubic nitrides TaN, CrN, and γ-Mo2N
either using high pressures or by adding ammonium salts which
decrease the maximum temperature of the reaction and increase
the amount of active nitrogen species in the reaction. Solid-
state metathesis reactions under pressures of up to 50 kbar have
been reported recently using a modified Bridgman anvil cell
Cl. These reactions were initiated by a resistively heated Nichrome
wire (∼850 °C applied for <1 s) placed in the reactant powder mixture.
The products were removed from the drybox and washed with distilled
water to dissolve the salt byproducts and any unreacted starting
materials. The products were then isolated using vacuum filtration.
High-pressure reactions were performed in modified Bridgman anvil
cells, utilizing a method developed originally for the synthesis of gallium
nitride.33 The reactants (∼60 mg for a 0.025 mL cell) were ground to
a fine powder and put in a pellet press with a thin (0.15 mm diameter)
iron wire placed in the middle of the powder. This assembly was then
pressed into a dense pellet (∼4.8 mm diameter, ∼1.5 mm high). An
Inconel gasket was coated with an alumina/magnesia paste (dyed red
with ∼1% iron oxide, added for identification purposes), affixed to
the bottom anvil, and cured. The reactant pellet was then placed in the
gasket with the top anvil aligned over it. Leads made of tantalum or
copper were next put in contact with the tungsten carbide cores of the
anvils and then covered with poly(vinyl chloride) sheets. This apparatus
was kept in a helium-filled resealable plastic bag for transport to the
high-pressure press. The Bridgman anvils were then placed in a
hydraulic press that can generate up to 100 kbar of static pressure. The
pressure applied was monitored using a calibrated load cell that
measures applied force. After the pressure was increased, the reaction
was initiated by passing current through the metal leads, which
33
for the successful synthesis of GaN. This method is now shown
(
27) Moffatt, W. G. (Research and Development Center, General Electric
Co.). The handbook of binary phase diagrams; Genium Publishing
Corp.: Schenectady, NY, 1984.
(
(
(
(
(
28) Massalski, T. B.; Murray, J. L.; Bennett, L. H.; Baker, H. Binary alloy
phase diagrams; American Society for Metals: Metals Park, OH, 1986.
29) Marchand, R. J.; Gouin, X.; Tessier, F.; Laurent, Y. Mater. Res. Soc.
Symp. Proc. 1995, 368, 15-20.
30) Aguas, M. D.; Nartowski, A. M.; Parkin, I. P.; MacKenzie, M.; Craven,
A. J. J. Mater. Chem. 1998, 8, 1875-1880.
(33) Wallace, C. H.; Kim, S. H.; Rose, G. A.; Rao, L.; Heath, J. R.; Nicol,
M.; Kaner, R. B. Appl. Phys. Lett. 1998, 72, 596-598.
(34) Sch a¨ fer, H. Chemical transport reactions; Academic Press: New York,
1964.
(35) Shoemaker, D. P. Experiments in physical chemistry, 4th ed.; McGraw-
Hill: New York, 1981.
31) Agrafiotis, C. C.; Puszynski, J. A.; Hlavacek, V. Combust. Sci. Technol.
1
991, 76, 187-218.
32) Gatterer, J.; Dufek, G.; Ettmayer, P.; Kieffer, R. Monatsch. Chem.
975, 106, 1137-47.
1

2242 Inorganic Chemistry, Vol. 40, No. 10, 2001
O’Loughlin et al.
Figure 1. Powder X-ray diffraction patterns of washed products from
the SSM reaction of TaCl and Li N under the following conditions:
a) ignited at ambient pressure in a steel reactor; (b) ignited under 45
kbar of static pressure; (c) combined with NH Cl (in a 1:2:3 molar
ratio of TaCl :Li N:NH Cl) and ignited in a steel reactor under ambient
pressure. The Miller indices for cubic TaN are given in (c) and marked
with a dot in (a) and (b). The diffraction peaks marked with an asterisk
2
correspond to hexagonal Ta N.
Figure 2. Powder X-ray diffraction patterns of washed products from
the SSM reaction between CrCl and Li N: (a) ignited at ambient
3 3
pressure in a steel reactor; (b) ignited under 39 kbar of static pressure
in a modified Bridgman anvil cell; (c) ignited under 49 kbar of pressure.
The Miller indices for cubic CrN are given in (c) and indicated by a
dot in (b). The other phases are identified as follows: Cr N, asterisk;
2
Cr, open dot; alumina/magnesia paste used in high-pressure cell, plus
sign.
5
3
(
4
5
3
4
resistively heats the wire in the reactant pellet. After the reaction was
complete, the products were collected and washed with distilled water
to remove the salt byproducts. These reactions can be scaled up using
larger volume cells such as those found in a modified-belt-type
and at high pressure by self-propagating high-temperature
31
synthesis. The ambient-pressure reaction between TaCl5 and
5
/
3Li3N results in the synthesis of hexagonal Ta2N (Figure 1a).
3
6,37
Diffraction peaks of hexagonal Ta2N are marked with an asterisk
in Figure 1a. By applying 45 kbar of static pressure to the
reaction of TaCl5 and Li3N, the only crystalline product
produced is cubic TaN (Figure 1b). The diffraction peaks
marked with a dot in Figure 1b correspond to face-centered
cubic TaN (JCPDS no. 32-1283). The reaction of TaCl5, Li3N,
and NH4Cl (in a 1:2:3 molar ratio) results in the formation of
crystalline cubic TaN along with a trace of hexagonal Ta2N
apparatus.
Powder X-ray Diffraction. The washed products were examined
by powder X-ray diffraction (XRD) using a θ-2θ diffractometer
(Crystal Logic) with Cu KR radiation. Scans were taken in the range
of 2θ, 10-100° at 0.10 or 0.05 deg intervals, and with 3 s count times.
The lattice parameters were computed using least-squares refinement.
Electron Microscopy. Loose powders were affixed to graphite disks
with colloidal graphite paste, and particle morphologies and sizes were
analyzed by scanning electron microscopy (SEM; Cambridge Stereoscan
(Figure 1c).
2
50).
Surface Area Analysis. A 0.5-1 g sample of loose powder was
Chromium Nitride. The reaction of CrCl3 and Li3N at
placed in a glass tube with a frit seal cap and heated under dynamic
vacuum to 150 °C to remove any adsorbed gases and water. The tube
was then moved to a Micrometrics ASAP 2010 BET surface area
analyzer. The data were worked up using specialized software
ambient pressure produces the mixed products Cr and Cr2N
(Figure 2a). The major diffraction peaks of Cr N are marked
2
with an asterisk, while the remaining peaks belonging to Cr
metal are marked with an open dot. The peaks for Cr2N are
shifted to slightly higher 2θ values likely due to nitrogen
vacancies, as compared to a standard pattern (JCPDS no. 11-
(Micrometrics ASAP 2010). Surface areas were measured by the
pressure difference between the amount of nitrogen gas adsorbed and
desorbed from the powder surface utilizing both the BET and Langmuir
3
8
65). By applying 39 kbar of pressure to the reaction of CrCl3
methods of analysis.
and Li3N, some cubic CrN (denoted with a dot in Figure 2b) is
Results
formed in addition to the ambient-pressure products Cr and
Cr2N. The peak at 63.55 2θ (denoted by a plus sign) is due to
the magnesia/alumina paste used to affix the gasket to the
Bridgman anvil cell. Applying 49 kbar of pressure, followed
by ignition of the reactants, produces cubic CrN as the only
crystalline nitride product (Figure 2c). The ambient-pressure
reaction of CrCl3, Li3N, and NH4Cl (in a 1:1:3 molar ratio)
results in the formation of Cr, Cr2N, and CrN. Reactants using
molar ratios of 1:1:y where y ranges from 0.5 to 5 as well as
Tantalum Nitride. The synthesis of tantalum nitrides has
been explored at ambient pressure by solid-state metathesis
2
3
(
(
(
36) Bundy, F. P. In High-pressure technology; Paauwe, J., Spain, I. L.,
Eds.; Marcel Dekker: New York, 1977; Vol. 2, p 325.
37) Sherman, W. F.; Stadtmuller, A. A. Experimental techniques in high-
pressure research; Wiley: Chichester, West Sussex, New York, 1987.
38) Webb, P. A.; Orr, C. Analytical methods in fine particle technology;
Micro-metrics Instrument Corp.: Norcross, GA, 1997.

Solid-State Synthesis of TaN, CrN, and MoN
Inorganic Chemistry, Vol. 40, No. 10, 2001 2243
Figure 3. Powder X-ray diffraction patterns of washed products from
the SSM reaction between MoCl and Ca N under the following
5 3 2
conditions: (a) reactants ignited as a dense pellet; (b) reactants held
under constant-volume conditions; (c) reactants put under 57 kbar of
2
static pressure. The Miller indices for γ-Mo N are given in (c), and
Figure 4. Powder X-ray diffraction patterns of washed products from
the ambient pressure reaction between MoCl
mol of NH
Cl added to ∼1 g of total reactants, (b) product of washing
a) in a 0.2 M HCl solution, (c) 3 mol of NH
Cl aded to ∼2 g of total
reactants, and (d) 3.5 mol of NH
Cl added to ∼2 g of total reactants.
The Miller indices for cubic γ-Mo N are given in (a). The diffraction
5 3 2
and Ca N with (a) 3
4
(
4
4
the diffraction peaks are indicated by a dot in (a) and (b). The peaks
marked with an open dot correspond to Mo metal.
2
4
peaks marked by an asterisk correspond to CaMoO , and those marked
by an open dot correspond to Mo metal.
1:0.5:0.5 and 1:0.75:0.25 have been carried out. All of these
ratios result in similar product mixtures (Cr, Cr2N, CrN).
The ambient-pressure reaction of CrI3 and Li3N results in
highly crytalline Cr2N without any Cr metal impurity. The peaks
corresponding to Cr2N are again shifted to slightly higher angles
compared to the standard pattern. By adding 3 mol of NH4Cl
to the ambient pressure reaction of CrI3 and Li3N, highly
crystalline CrN with some Cr2N results.
Molybdenum Nitride. The reaction between MoCl5 and Li3N
is self-detonating; i.e., it goes off spontaneously upon grinding
the two powders together in a mortar and pestle. An alternative
nitrogen source, Ca3N2, was therefore used to react with MoCl5.
The ambient-pressure reaction of MoCl5 and Ca3N2
applied to these reactants using a Bridgman anvil cell, crystalline
cubic γ-Mo2N forms (Figure 3c). The addition of 3 mol of NH4-
Cl to the reaction between MoCl5 and Ca3N2 (eq 2) at ambient
pressure also produces highly crystalline γ-Mo2N with a small
amount of Mo metal and CaMoO4 (Figure 4a). These impurities
can readily be removed by washing with 0.2 M hydrochloric
acid (Figure 4b).
The mixing order of the reactants is important in the synthesis
of γ-Mo2N. If untransported MoCl5, which contains lower
oxidation state molybdenum chlorides and oxychlorides, is used
in the reaction with Ca3N2 and NH4Cl, the MoCl5 and Ca3N2
can be safely ground together. However, if pure, vapor-
transported MoCl5 is used, it must first be mixed with NH4Cl
before Ca3N2 is added to avoid spontaneous detonation. The
above ambient-pressure synthesis is the result of reactions
carried out on a small scale (<1 g of total reactants). When the
6
3
7
/5
MoCl + Ca N f / γ-Mo N + 3CaCl + / N (2)
5 3 2 5 2 2 10 2
produces highly crystalline Mo metal and no crystalline nitride
products. By pressing the reactant mixture into a pellet before
detonation, the reaction produces an only slightly crystalline
product indexed to γ-Mo2N (JCPDS no. 25-1366) along with
crystalline Mo metal (Figure 3a). The γ-Mo2N product is
denoted with an asterisk in the XRD pattern, while Mo metal
is denoted with an open dot. By performing the same reaction
in a piston cylinder cell,39 which provides pseudo-constant-
volume conditions, the crystallinity of the γ-Mo2N produced is
improved, but is still far less crystalline than that of the Mo
metal contaminant (Figure 3b). When 57 kbar of pressure is
reaction is scaled up (g1 g of total reactants), the γ-Mo N
2
formed is less crystalline compared to that in the small-scale
reaction and crystalline Mo metal is found (Figure 4c). Addition
of an extra 0.5 mol of NH Cl (ratio 1.2:1:3.5) to a ∼2 g reaction
4
results in products comparable to those of the small-scale
reaction (Figure 4d).
Surface Area. Cubic molybdenum nitride (γ-Mo2N) powder
synthesized from a small-scale ambient-pressure reaction among
MoCl5, Ca3N2, and 3 mol of NH4Cl (Figure 3c) has a surface
2
area of 22 m /g calculated using the BET modeling system and
2
3
0 m /g using a Langmuir model. The surface area of the
(
39) Holm, S. R. Rapid synthetic routes to refractory intermetallics,
transition metal carbides, solid solutions, and hybrid materials;
University of California, Los Angeles: Los Angeles, 1997; p 163.
γ-Mo2N produced by a large-scale reaction (∼2 g of reactants,
Figure 3d) of MoCl5, Ca3N2, and 3.5 mol of NH4Cl was slightly

2244 Inorganic Chemistry, Vol. 40, No. 10, 2001
O’Loughlin et al.
Figure 5. Scanning electron microscopy images of (a) a small-scale reaction (∼1 g) between MoCl
the reaction of MoCl and Ca in a piston cylinder cell, (c) a large-scale reaction of MoCl and Ca
5
and Ca
3
N
2
with 3 mol of NH
4
Cl added, (b)
5
N
3 2
5
3
N
2
with 3.5 mol of NH
4
Cl added, and (d) an
expanded view of the center particle in (b).
2
group 4 and 5 nitrides;28 (3) since metathesis reactions take place
in <1 s, the nitrogen gas produced by reactant decomposition
may not be sufficient to form the product nitride. It appears
that, to incorporate a stoichiometric amount of nitrogen, a
significant nitrogen overpressure must be maintained during the
lower, 16 and 22 m /g using the BET and Langmuir models,
respectively. SEM images of the washed powders are shown
in Figure 5. As can be seen in Figure 5a, these SEM images
indicate that the powder synthesized from MoCl5, Ca3N2, and
mol of NH4Cl on a small scale (∼1 g of total reactants)
3
consists of spherical particles with average diameters of 5-10
µm. Figure 5c shows that the γ-Mo2N powder synthesized on
a larger scale (∼2 g of total reactants) consists of much larger,
nonspherical particles 30-50 µm in diameter. Parts b and d of
Figure 5 are images of the γ-Mo2N powder synthesized from
the reaction between MoCl5 and Ca3N2 in a piston cylinder cell
under pseudo-constant-volume conditions. The particle size of
the γ-Mo2N in Figure 5b is much larger than that synthesized
at ambient pressure. The γ-Mo2N consists mainly of spherical
particles approximately 150 µm in diameter. Figure 5d shows
a more detailed view of the particles found in the center of
Figure 5b.
reaction. In the synthesis of TaN, CrN, and γ-Mo N, the non-
2
air-tight steel reactor is inadequate to contain the nitrogen gas
produced from the solid-state metathesis reactions. As is found
in the case of γ-Mo N, by pressing the reactant mixture into a
2
fully dense pellet, the amount of nitride formed in the reaction
increases. The pellet likely provides a larger barrier to diffusion
of gas out of the reactant mixture, thus increasing the amount
of nitrogen available for nitride formation. By applying sufficient
pressure (e.g., >40 kbar) to an SSM reaction in a Bridgman
anvil cell, this diffusion barrier can be greatly enhanced. This
is particularly important in the reaction between CrCl3 and Li3N
to produce CrN. From the phase diagram of chromium and
nitrogen, it is known that only in a small finite range (49.5-
Discussion
2
8
5
0.0 atom % nitrogen) will cubic CrN form. Thus, experi-
The nitrides reported here (TaN, CrN, and γ-Mo2N) have
proven to be difficult to synthesize in previous attempts using
solid-state metathesis reactions, likely due to the following: (1)
tantalum mononitride (TaN) requires higher nitrogen pressures
to form in comparison to its subnitride, Ta2N;15 (2) both CrN
and γ-Mo2N have a small stability range in comparison to the
mentally we observe that only at high pressures (g49 kbar),
where the nitrogen produced by Li3N is maintained in the
reactant mixture throughout the reaction, can cubic CrN be
synthesized by solid-state metathesis reactions. Similar high
pressures are needed for both the tantalum nitride (45 kbar) and
molybdenum nitride (57 kbar) systems.

Solid-State Synthesis of TaN, CrN, and MoN
Inorganic Chemistry, Vol. 40, No. 10, 2001 2245
The containment of the nitrogen byproduct is essential for
obtaining a stoichiometric ratio of transition metal to nitride.
This is a major factor influencing the synthesis of group 6
nitrides as well as tantalum nitride. By applying static pressure,
the nitrogen gas can be contained and the stoichiometry
maintained, making cubic nitride formation favorable.
The use of NH4Cl in solid-state metathesis reactions enables
a rapid, low-temperature route to bulk group 5 (tantalum) and
group 6 (chromium and molybdenum) nitrides. The nitride of
most interest may be γ-Mo2N. Since reports of the temperature-
programmed reaction of MoO3 and NH3 at high temperatures
2
45
resulting in high surface area (up 170 m /g) γ-Mo2N powders,
many efforts have been made to produce bulk, high-surface-
area molybdenum nitride powders for catalysis. The molybde-
num nitride powders synthesized here have considerably smaller
A result of applying pressure to solid-state metathesis
reactions is that the products have the property of well-sintered
materials. The SEM image of the product of the pseudo-
constant-volume reaction between MoCl5 and Ca3N2 (Figure
2
surface areas (up to 30 m /g, using BET with a Langmuir
model), which is consistent with γ-Mo2N synthesized by other
5d) shows this sintering effect. The material produced is highly
4
6
chemical methods. Recent studies on the catalytic properties
of γ-Mo2N powders indicate that both high- and medium-
surface-area powders demonstrate the same activity and selec-
crystalline but has a low surface area. Because some of these
nitrides, particularly γ-Mo2N, are potential catalytic materials,
low-pressure synthetic methods are desirable.
1
3
tivity as hydrodesulfurization catalysts. Thus, molybdenum
nitrides of medium surface area, such as those synthesized by
solid-state metathesis reactions, could be potentially useful as
hydrodesulfurization catalysts.
The addition of ammonium chloride to an SSM reaction has
4
0
been shown to provide a low-pressure route to other nitrides.
Ammonium chloride appears to have two functions, acting first
as a heat sink and then as a source of nitrogen. The practice of
using inert salts (i.e., LiCl, NaCl, etc.) as heat sinks in solid-
Conclusions
41-43
state metathesis reactions is well documented in the literature.
Highly crystalline cubic chromium, molybdenum, and tan-
talum nitrides have been synthesized by using solid-state
metathesis reactions under pressure. The application of static
pressures (>40 kbar) to solid-state metathesis reactions favors
the formation of nitrides that have small stability ranges and
require high nitrogen pressures to form. An alternative route to
synthesizing these materials is to add several moles of am-
monium chloride to the reactants. Ammonium chloride acts as
both a nitriding agent and a heat sink for the reactants. The use
of this salt is effective in synthesizing both cubic TaN and
γ-Mo2N. Materials synthesized under ambient conditions have
higher surface areas than those synthesized under high pressure,
as confirmed by scanning electron microscopy. Molybdenum
nitride, γ-Mo2N, synthesized under ambient conditions, is shown
However, ammonium chloride only functions as a heat sink until
the reaction temperature exceeds its sublimation point of 340
°
C. Upon sublimation, it decomposes to NH3 and HCl, providing
44
a local ammonia atmosphere for the reactants. Experimental
evidence indicates that the ammonia produced can have an active
role in the nitridation process. For example, when 2 mol of NH4-
Cl is added to an SSM reaction expected to form TaC
9
5
TaCl + CBr + / Mg f TaC + / MgCl + 2MgBr (3)
5
4
2
2
2
2
the only nonsalt product formed is partially crystalline TaN.
Therefore, as the reaction temperature increases above 340 °C,
it reaches a regime where the NH3 produced from NH4Cl
decomposition is capable of further acting as a nitriding agent.
Since the amount of ammonium chloride that is effective in
acting as a nitriding agent is unknown and may vary from one
reaction to another, this could explain why its use is not effective
in producing CrN. Cubic chromium nitride (CrN) has a very
small stability range, and if nitrogen is in excess, a mixture of
products will result. Therefore, our results appear to be
consistent with the phase diagram of CrN.28
2
to have a surface area of 30 m /g using the Langmuir model.
Materials synthesized under ambient conditions could be
potentially useful in catalysis.
Acknowledgment. We thank the NSF Center for High
Pressure Research (CHiPR) and Dr. Ivan Getting for consulta-
tion on the development and maintenance of our high-pressure
equipment, Dr. Steven Holm for the design of the piston cylinder
cell, and Dr. Bruce Dunn (UCLA Department of Materials
Science) for the use of his BET surface area analyzer, and Lisa
Viculis for designing the cover illustration. This work was
supported by National Science Foundation Grants DMR-
(
40) Wallace, C. H.; Reynolds, T. K.; Kaner, R. B. Chem. Mater. 1999,
1, 2299-2301.
41) Bonneau, P. R.; Jarvis, R. F.; Kaner, R. B. Nature 1991, 349, 510-
12.
1
9
704964 and DMR-0073581.
(
5
IC001265H
(
(
(
42) Gillan, E. G.; Kaner, R. B. Chem. Mater. 1996, 8, 333-343.
43) Parkin, I. P. Chem. Soc. ReV. 1996, 25, 199-207.
(45) Volpe, L.; Boudart, M. J. Solid State Chem. 1985, 59, 332-47.
(46) Sriram, M. A.; Kumta, P. N.; Ko, E. I. Chem. Mater. 1995, 7, 859-
864.
44) Kirk, R. E.; Grayson, M.; Eckroth, D.; Othmer, D. F. Encyclopedia
of chemical technology, 3rd ed.; Wiley: New York, 1978.