Low-Temperature Solution Route to Molybdenum Nitride

It should not be forgotten that single shell nanotube

sam-ples contain tubes of not only different diameters but also

different electronic properties. In any case, both

semicon-ducting and metallic nanotubes will have high surface

ener-gies and, therefore, even if different their wetting

proper-ties will probably fall within the intervals reported here for

g

C

and g

max

. When considering the capillarity of single shell

tubes with very small inner diameter (< 1 nm), the fact that

the inner and outer surfaces may have different energies

owing to the asymmetry of the electronic densities

[22]

should be taken into account. To gain proper insight into

such issues, it will be necessary to have samples with just

one type of well-defined single shell nanotube, which are

currently not available.

Experimental

Raw nanotubes were sonicated for 1 h in carbon disulfide (25 mL per 10 mg of nanotubes), filtered, and dried before wetting experiments. Pure samples were prepared by following the purification procedure described elsewhere [14]. The annealed nanotubes were first purified and then heated under vacuum (2 ´ 10±6_{torr) at 100 C for 10 min, followed by 15 min at}

500 C, and eventually the temperature was raised to 900 C before natural cooling. The powders were ground in a mortar and divided in small quartz tubes containing 2±3 mg of nanotubes each. The samples were degassed un-der vacuum (5 ´ 10±6_{torr) and heated for an hour (at 500 C for the raw and}

annealed, 120 C for the purified nanotubes) before being transferred under argon to a glove box. The chemical, (the purest quality available from Al-drich Chemical Company Inc., 99.95±99.99 %) to be melted was added on top of the packed nanotubes in an approximately 1:1 volume ratio. The quartz tube was then degassed for another hour at the same temperature as described earlier (or below the melting point of the chemical, depending on its nature) and sealed under vacuum (2±5 ´ 10±6_torr).

The sealed quartz tubes were heated from room temperature to 50 C above the melting point of the tested compounds at about 1 C/s and left at the final temperature for 1±4 h.

Received: April 24, 1998 Final version: June 29, 1998 ±

[1] P. M. Ajayan, T. W. Ebbesen, Rep. Prog. Phys., 1997, 60, 1025. [2] P. M. Ajayan, S. Iijima, Nature,1993 361, 333.

[3] E. Dujardin, T. W. Ebbesen, H. Hiura, K. Tanigaki, Science, 1994, 265, 1850.

[4] S. C. Tsang, Y. K. Chen, P. J. F. Harris, M. L. H. Green, Nature, 1994, 372, 159.

[5] D. Ugarte, A. Châtelain, W. A. de Heer, Science, 1996, 274, 1897. [6] M. H. L. Green et al., Chem. Commun., 1998, 347.

[7] P. M. Ajayan, O. Stephan, P. Redlich, C. Colliex, Nature, 1995, 375, 564.

[8] T. W. Ebbesen, H. Hiura, M. E. Bisher, M. M. J. Treacy, J. L. Shreeve-Keyer, R. C. Haushalter, Adv. Mater., 1996, 8, 155.

[9] S. C. Tsang, Y. K. Chen, P. J. F. Harris, M. L. H. Green, Nature, 1994, 372, 159.

[10] J. Israelachvili, Intermolecular and Surface Forces, 2nd edition, Aca-demic Press, London, 1992, 317.

[11] Handbook of Physics and Chemistry (Ed: R. C. Weast), CRC Press, West Palm Beach, FL, 58th edition, 1977±78.

[12] Y. Miyamoto, A. Rubio, X. Blase, M. L. Cohen, S. Louie, Phys. Rev. Lett., 1995, 74, 2993 (the value of surface tension of potassium (g = 395 mN/m) used in this reference corresponds to a very old study, more recent measurements indicate a value of about 100±120 mN/m (see reference 11); A. Rubio, Y. Miyamoto, X. Blase, M. L. Cohen, S. Louie, Phys. Rev. B, 1996, 53, 4023.

[13] T. Guo, P. Nikolaev, A. Thess, D. T. Colbert, R. E. Smalley, Chem. Phys. Lett., 1995, 243, 49.

[14] E. Dujardin, T. W. Ebbesen, A. Krishnan, M. M. J. Treacy, Adv. Ma-ter., 1998, 10, 611.

[15] A. M. Schwartz, C. A. Rader, E. Huey, in Contact Angle, Wettability and Adhesion, Advances in Chemistry Series 43, ACS, Washington, 1963, p. 250.

[16] W. A. Zisman, in Contact Angle, Wettability and Adhesion, Advances in Chemistry Series 43, ACS, Washington, 1963, p. 1.

[17] Notice that by convention the units of gCare in energy per unit area

since it is a characteristic of the solid surface, while the surface tension of a liquid is given in units of force per unit length.

[18] J. C. Charlier, X. Gonze, J. P. Michenaud, Europhys. Lett., 1995, 29, 43.

[19] V. N. Bogomolov, Sov. Phys. Tech. Phys., 1992, 37, 79. [20] S. K. Rhee, J. Am. Ceram. Soc., 1971, 54, 376. [21] V. A. Ogarev, Kolloidnyi Zhurnal, 1978, 40, 153.

[22] X. Blase, L. X. Benedict, E. L. Shirley, S. G. Louie, Phys. Rev. Lett., 1994, 72, 1878.

[23] H. M. Princen, Surface and Colloid Science (Ed: E. Matijevic), vol. II, Interscience, New York, 1969.

[24] B. J. Carroll, J. Colloid Interface Sci., 1976, 57, 488.

Low-Temperature Solution Route to

Molybdenum Nitride**

By Hsin-Tien Chiu,* Shiow-Huey Chuang,

Gene-Hsiang Lee, and Shie-Ming Peng

Transition metal nitrides are technologically important

materials with many interesting properties.

[1±3]

_Frequently,

these materials are prepared at high temperatures by direct

nitridation or chemical vapor deposition (CVD)

[4]

employ-ing N

2

or NH

3

as the source of nitrogen atoms. Rapid

solid-state synthesis, a highly exothermic self-propagating

reac-tion, represents an alternative route.

[5±9]

_{Here, we wish to}

report the synthesis of molybdenum nitride powder,

[10±12]

an effective catalyst for desulfurization and

hydro-denitrogenation of hydrocarbons,

[13]

_{via a sol-gel type of}

solution process, employing a mixture of Na

2

MoO

4

, (Me

3

-Si)

2

NH, Me

3

SiCl, and NEt

3

in refluxing DME

(1,2-di-methoxyethane, boiling point 358 K). Although the growth

of metal nitride thin films by CVD at 473 K has been

re-ported,

[14,15]

_{this is the first time that a transition metal}

ni-tride has been prepared from solution at low temperature.

When Na

2

MoO

4

was reacted with Me

3

SiCl, (Me

3

Si)

2

NH

and NEt

3

in refluxing DME under N

2

or Ar atmosphere, a

relatively air-stable black powder 1 precipitated, while a

white solid, identified as Et

3

NHCl, deposited on the inner

surface of the condenser. (Me

3

Si)

2

O was detected as the

major by-product in the solution; no volatile Mo

by-prod-ucts were detected in the reaction mixture. The average

particle size of 1 was determined to be 15±30 nm using

±

[*] Prof. H. T. Chiu, Dr. S. H. Chuang Department of Applied Chemistry National Chiao Tung University Hsinchu, Taiwan 30050 (R.O.C.) G. H. Lee, S. M. Peng Department of Chemistry National Taiwan University Taipei, Taiwan 10764 (R.O.C.)

[**] We thank the National Science Council of Taiwan, the Republic of China (NSC-85-2113-M-009-012) for support.

STEM (scanning transmission electron microscopy,

Fig. 1a). ED (electron diffraction) (Fig. 1b) identifies

over-lapping diffraction signals, which are assigned to randomly

dispersed microcrystals of a cubic-phase material with a =

4.2 Š and residual NaCl. The lattice parameter of a = 4.2 Š

is close to that of g-Mo

2

N, 4.16 Š. Due to the small particle

size of 1, X-ray diffraction (XRD) only showed reflections

of residual NaCl, the reason being that in general X-rays,

with their wavelengths much longer than that of the

elec-tron beam, can only be used to show diffraction patterns of

crystalline particles with sizes greater than 200 nm.

[16]

After 1 was thermally treated at 873 K for 1 h, broad XRD

peaks corresponding to the g-Mo

2

N phase were observed.

A high-resolution XPS (X-ray photoelectron spectroscopy)

study of 1 showed signals corresponding to Mo(3d

5/2

),

Mo(3d

3/2

), Mo(3p

3/2

), Mo(3p

1/2

), and N(1s) electrons at

228.2, 231.4, 394.1, 411.8, and 397.3 eV, respectively. The

signals of C(1s) and O(1s) electrons were observed at 284.1

and 531.0 eV, respectively. They are assigned to

by-prod-ucts, which are yet to be fully separated. The N(1s) signal is

characteristic for a metal nitride material. These data were

compared with a commercial sample of Mo

2

N, which

showed corresponding elemental signals at 228.2, 231.3,

394.2, 411.7, and 397.2 eV. In Figure 2, high-resolution

spectra of Mo(3p

3/2

), Mo(3p

1/2

), and N(1s) electrons are

shown for 1 and the reference sample, thus confirming that

1 contains molybdenum nitride. After NaCl was separated

from 1 by dissolving the salt in water, the insoluble residue,

1a, was collected and dried in vacuo at room temperature.

XPS of 1a indicates that the molybdenum and nitrogen

sig-nals (Fig. 2) differ little from those of 1. Heating 1 in

vacu-um at 673 K, to remove traces of any volatile by-products,

produced a black powder 1b. As shown in Figure 2, the

high-resolution XPS signals of 1b show little difference

either. In addition, XPS signals of NaCl were observed for

1b. Neither 1a nor 1b showed C±H stretching signals in the

IR spectra.

Analyzing these data, where the relative concentrations

of Mo and N were determined from the XPS data by

inte-grating the corresponding signals after curve fitting (with a

commercial sample as a standard), we conclude that the

black powder 1 contains nanosized molybdenum nitride

Fig. 1. a) STEM bright-field image and b) ED (L = 80 cm, l = 0.0336 Š) of 1 (ultrasonically irradiated in ethanol).

Fig. 2. High-resolution XPS signals of Mo(3p3/2), Mo(3p1/2), and N(1s)

particles, MoN

x

(x = 0.4  0.1). A balanced equation is

pro-posed in Equation 1 to describe the overall reaction

stoi-chiometry. The evolution of dinitrogen molecules has been

proposed but has not yet been confirmed.

Na

2

MoO

4

+ 2(Me

3

Si)

2

NH + 4 Me

3

SiCl + 2 NEt

3

ÿÿÿÿÿÿÿÿ! NoN

x

+ 2 NaCl + 4 (Me

3

Si)

2

O +

DME (reflux)

2 NEt

3

HCl + (1 ± 0.5x) N

2

(1)

In order gain further insight into this reaction, we

investi-gated the system by adding reagents sequentially, thus

al-lowing reaction intermediates to be isolated. The

observa-tions are summarized in Scheme 1. Only Me

3

SiCl showed

an initial significant rate of reaction towards Na

2

MoO

4

in

DME. MoO

2

Cl

2

(DME), 2, was isolated in high yield.

[17]

Reacting 2 with Me

3

SiCl, (Me

3

Si)

2

NH, and NEt

3

in

reflux-ing DME generated a black powder 1c, characterized to be

MoN

x

by XPS (Fig. 2). Treatment of 2 with (Me

3

Si)

2

NH

yielded a pale yellow liquid, 3, by distillation, which is yet

to be fully characterized. Compound 3 gradually darkened,

indicating further reaction at room temperature. A known

dimeric nitrido complex [N º Mo(OSiMe

3

)

3

×NH

3

]

2

, 4,

crys-tallized from the mixture as a minor product.

[18]

_Formation

of (Me

3

Si)

2

O, (Me

3

Si)

2

NH and a brown precipitate yet to

be characterized were also observed. Addition of NEt

3

to 3

enhanced the apparent rate of formation of 4. Addition of

pyridine to 3 resulted in the isolation of another nitrido

complex N º Mo(OSiMe

3

)

3

×py, 5.

[17]

Reacting 3 with Me

3

-SiCl, (Me

3

Si)

2

NH, and NEt

3

in refluxing DME generated a

black powder, 1d, shown to be MoN

x

by XPS (Fig. 2).

In an aprotic environment, (Me

3

Si)

2

NLi was allowed to

react with 2 in hexane to form a pale yellow liquid,

formu-lated as a nitrido complex N º Mo(OSiMe

3

)

2

(N(SiMe

3

)

2

),

6, in high yield. Contrary to the instability of 3 at room

temperature, 6 showed little sign of decomposition. A

pyri-dine adduct of 6, N º Mo(OSiMe

3

)

2

(N(SiMe

3

)

2

)×py, 7, was

crystallized in good yield from hexane. Compound 7 is a

mononuclear five-coordinate complex with a distorted

square pyramidal geometry about the metal center (Fig. 3).

Scheme 1.

Fig. 3. ORTEP drawing of 7, showing the numbering scheme for the non-hy-drogen atoms. Selected bond distances (Š) and angles (): Mo±N(1) = 1.640(3), Mo±N(2) = 1.973(3), Mo±N(3) = 2.334(3), Mo±O(1) = 1.921(3), Mo±O(2) = 1.924(2), N(1)±Mo±N(2) = 102.5(1), N(1)±Mo±N(3) = 92.0(1), N(2)±Mo±N(3) = 165.4(1), N(1)±Mo±O(1) = 106.6(1), N(1)±Mo±O(2) = 107.6(1), O(1)±Mo±O(2) = 139.1(1), O(1)±Mo±N(2) = 96.3(1), O(1)±Mo± N(3) = 77.7(1), O(2)±Mo±N(2) = 97.7(1), O(2)±Mo±N(3) = 79.2(1).

The Mo atom lies slightly above the basal plane while

the nitrido ligand occupies the apical position. The

N(1)±Mo distance is short, 1.640(3) Š. The overall

geom-etry of 7 is closely related to those of 4, 5, and

[N º MoCl

3

(N(SiMe

3

)

2

)]

±

.

[19]

When 6 was reacted with

Me

3

SiCl, (Me

3

Si)

2

NH, and NEt

3

in refluxing DME a black

powder 1e resulted that was identified as MoN

x

by XPS

(Fig. 2).

Comparing this observation with the experiment

employ-ing 3 to form molybdenum nitride, little difference exists

except that 6 was first prepared in an aprotic medium and

then exposed to a protic medium. Therefore, we propose

protonation of 6 to be an essential step. This step probably

converts the nitrido ligand to an imido ligand,

[20]

_then,

through condensation reaction steps, into MoN

x

. For 3, in

the presence of a base such as pyridine or NEt

3

,

deprotona-tion occurs and the nitrido complexes 4 and 5 are

gener-ated. In Scheme 2, a generalized Mo±N±Mo

polymeriza-tion route is proposed to account for the formapolymeriza-tion of the

molybdenum nitride. Intermolecular condensation

reac-tion, removal of a molecule HY from Mo=NH and Y±Mo

results in a Mo±N±Mo linkage. This step is comparable to

the M±O±M formation pathway proposed for sol-gel routes

to metal oxide materials.

[21]

_{The proposed Mo º N and}

Mo=NH species resemble the M±O

±

_{and M±OH species in}

a regular sol-gel process. Analogous reactions are known

for Mo±O±Mo and W±N±W connectivity formations.

[20,22]

Repeating the condensation step polymerizes the

mono-meric MoN-containing units, such as imido and nitrido

in-termediates, into clusters of oligomers. Ladder structures

found for [(

t

_BuCH

2

)

2

TaN]

5

and W

4

N

4

(NPh

2

)

6

(OBu)

2

may

be viewed as models to represent the initial stages of the

oligomerization.

[23,24]

_{Further polymerization causes these}

oligomers to coagulate into nanosized molybdenum nitride

powders.

In order to extend the chemistry to include other metals,

we attempted to prepare tungsten nitride powder by

react-ing Na

2

WO

4

with Me

3

SiCl, (Me

3

Si)

2

NH, and NEt

3

in

re-fluxing DME. The reaction did not proceed probably due

to the high W±O bond strength. However, reacting a

mix-ture of WCl

6

, (Me

3

Si)

2

NH, and NEt

3

in refluxing DME

produced a black precipitate. Preliminary characterization

of this black powder by XPS indicated the presence of

tungsten nitride with some residual chloride.

[25]

In summary, we have demonstrated a low-temperature

solution route to molybdenum nitride by reacting a mixture

of Na

2

MoO

4

, Me

3

SiCl, (Me

3

Si)

2

NH, and NEt

3

in refluxing

DME. From the molecular complexes observed in this

study, the process shows parallels to the sol-gel processing

of metal oxide materials in many ways. This study extends

the already versatile chemistry of interconversions among

various metal±nitrogen containing complexes by showing

an excellent correlation between molybdenum nitrido

com-plexes and molybdenum nitride. Further investigations to

extend our understanding of the reaction are in progress.

Experimental

All chemicals and solvents were manipulated under dry and oxygen-free N2atmosphere. Reactions carried out under Ar atmosphere showed same

result.

Compound 1: To Na2MoO4 (2.0 g, 9.7 mmol) suspended in DME

(100 mL), NEt3(5.5 mL, 40 mmol), (Me3Si)2NH (4.1 mL, 19 mmol), and

Me3SiCl (11.1 mL, 87.5 mmol) were added sequentially. The mixture

gradu-ally darkened within 1 h, after which it was refluxed for 12 h. During this time, a white solid deposited on the inner surface the condenser. An air-stable black precipitate was collected from the reaction mixture.

Compound 1a: After 1 was washed with H2O, the insoluble black

precipi-tate was collected and dried under vacuum.

Compound 1b: 1 was heated at 673 K under vacuum for 1 h. The black solid was collected.

Binding energies [eV]. 1a: Mo(3d5/2), 228.4; Mo(3d3/2), 231.5; Mo(3p3/2),

394.4; Mo(3p1/2), 411.9; N(1s), 397.0. 1b: Mo(3d5/2), 228.0; Mo(3d3/2), 231.3;

Mo(3p3/2), 394.4; Mo(3p1/2), 411.9; N(1s), 397.0. Based on the intensities of

C(1s) and O(1s) signals, both elements are judged to be insignificant compo-nents. 1c: Mo(3d5/2), 228.3; Mo(3d3/2), 231.5; Mo(3p3/2), 394.4; Mo(3p1/2),

411.6; N(1s), 397.1. 1d: Mo(3d5/2), 228.2; Mo(3d3/2), 231.4; Mo(3p3/2), 394.3;

Mo(3p1/2), 411.6; N(1s), 397.2. 1e: Mo(3d5/2), 228.4; Mo(3d3/2), 231.6;

Mo(3p3/2), 394.4; Mo(3p1/2), 411.8; N(1s), 397.3.

Compound 3: To 2 (5.0 g, 17 mmol) in hexane (100 mL), (Me3Si)2NH

(14.6 mL, 69.2 mmol) was added. After work up, a yellow liquid was iso-lated (3.1 g, 40 % yield based on Mo). Initially, the liquid was formuiso-lated to be (Me3SiN=)2Mo(OSiMe3)2based on NMR spectroscopy evidence [26]. 1_{H NMR (300 MHz, CDCl}

3, 25 C): d 0.11 (s, OSiMe3), 0.20 (s, NSiMe3).

Al-though the two signals appeared to correspond to an equal number of hy-drogen atoms, accurate integrations were difficult to obtain due to the broadening of the signal at 0.11 ppm. Contrary to the literature description of (Me3SiN=)2Mo(OSiMe3)2, 3 showed sign of decomposition and gradually

darkened at room temperature. We speculate that a minor proton-contain-ing molecule coexisted because regardless how carefully we prepared 3, stretchings of O±H (3779, 3730, 3640, 3570 cm±1_{) and N±H (3404, 3357,}

3241, 3161 cm±1_{) were observed in the IR spectra. The mass spectrometry}

(MS) data suggest that ions of three different complexes coexisted when 3 was evaporated into the spectrometer. MS (EI = 12 eV, direct inlet,98_Mo):

m/z = 379 (MoNO3Si3C9H27+), 450 (MoN2O2Si4C12H36+), 526 (MoN2O

3-Si5C15H46+± Me + 1). The assignments are m/z = 379, N º Mo(OSiMe3)3; m/

z = 450, (Me3SiN=)2Mo(OSiMe3)2; m/z = 526, a fragment of (Me3SiN=)

Mo(NHSiMe3)(OSiMe3)3. Thus, we speculate that when freshly prepared, 3

was a mixture of (Me3SiN=)2Mo(OSiMe3)2, a major product, and Me3SiOH,

a minor by-product. Upon standing at room temperature, Me3SiOH reacted

with (Me3SiN=)2Mo(OSiMe3)2through several addition±elimination and

si-lyl group migration steps. 3 thus became a mixture of molybdenum com-plexes, including a nitrido complex later isolated as 4 and 5.

Compound 6: To 2 (1.51 g, 5.20 mmol) stirred vigorously in hexane (50 mL), LiN(SiMe3)2 (1.74 g, 10.4 mmol) was added. The mixture was

stirred for 2 h. After work-up, 6 was isolated as a yellow liquid (2.0 g, 85 % yield based on Mo).1_{H NMR (300 MHz, toluene-d}

8, ±20 C): d 0.25 (s, 9H,

NSiaMe3), 0.33 (s, 18H, OSiMe3), 0.54 (s, 9H, NSibMe3);13C NMR (75 MHz,

toluene-d8, ±20 C): d 1.8 (OSiMe3), 2.7 (NSiaMe3), 5.0 (NSibMe3). MS (EI, 98_{Mo): m/z = 450 (M}+_).

Compound 7: To 6 (2.0 g, 4.4 mmol) in hexane (50 mL), pyridine (py) (1.5 mL, 19 mmol) was added. After stirring for 18 h, the solvent was re-moved, producing a yellow solid. Recrystallization from hexane yielded yel-low crystals (1.5 g, 64 % based on Mo).1_{H NMR (300 MHz, toluene-d}

8, ±10 C): d 0.14 (s, 18H, OSiMe3), 0.43 (s, 9H, NSiaMe3), 0.67 (s, 9H, NSibMe3), 6.42 (t, 2H, ±NCHCHCH), 6.79 (t, 1H, ±NCHCHCH), 8.44 (d, 2H, ±NCHCHCH);13_{C NMR (75 MHz, toluene-d} 8, ±10 C): d 2.0 (OSiMe3), 2.9 (NSiaMe3), 5.4 (NSibMe3), 124.8 (±NCHCHCH), 137.7 (±NCHCHCH),

150.7 (±NCHCHCH). MS (EI,98_{Mo): m/z = 379 (M}+_{± py). Crystal}

parame-ters of 7 at 298 K: space group P1, a = 10.499(1), b = 10.974(3), c = 13.001(3) Š, a = 83.03(2), b = 81.49(2), g = 80.31(2), V = 1453.1(5) Š3_{, Z =}

2, Dc= 1.206 g/cm3, Rf= 0.036, Rw= 0.034.

Received: April 20, 1998 Final version: June 30, 1998 ±

[1] L. E. Toth, Transition Metal Carbides and Nitrides, Academic Press, New York 1971.

[2] S. T. Oyama, The Chemistry of Transition Metal Carbides and Nitrides, Blackie Academic Professional, Glasgow 1996.

[3] H. O. Pierson, Handbook of Refractory Carbides and Nitrides, Noyes Publications, Westwood 1996.

[4] D. M. Hoffman, Polyhedron 1994, 13, 1169. [5] J. B. Wiley, R. B. Kaner, Science 1992, 255, 1093. [6] E. G. Gillan, R. B. Kaner, Inorg. Chem. 1994, 33, 5693.

[7] A. Hector, I. P. Parkin, J. Chem. Soc., Chem. Commun. 1993, 1095. [8] A. L. Hector, I. P. Parkin, Polyhedron 1995, 14, 913.

[9] A. L. Hector, I. P. Parkin, Chem. Mater. 1995, 7, 1728.

[10] Generally, molybdenum nitride powder is prepared by passing NH3

over MoO3at T > 873 K.

[11] L. Volpe, M. Boudart, J. Solid State Chem. 1985, 59, 332. [12] M. A. Sriram, P. N. Kumata, E. I. Ko, Chem. Mater. 1995, 7, 859. [13] S. Ramanathan, S. T. Oyama, J. Phys. Chem. 1995, 99, 16 365. [14] R. Fix, R. G. Gordon, D. M. Hoffman, Chem. Mater. 1991, 3, 1138. [15] R. Fix, R. G. Gordon, D. M. Hoffman, Chem. Mater. 1993, 5, 614. [16] A. R. West, Solid State Chemistry and Its Applications, Wiley, New

York 1984, p. 51.

[17] H.-T. Chiu, Y.-P. Chen, S.-H. Chuang, J.-S. Jen, G.-H. Lee, S.-M. Peng, Chem. Commun. 1996, 139.

[18] G. S. Kim, C. W. Dekock, J. Chem. Soc., Chem. Commun. 1989, 1166. [19] D. Fenske, A. Frankenau, K. Dehnicke, Z. Anorg. Allg. Chem. 1989,

574, 14.

[20] W. A. Herrmann, S. Bogdanovic, R. Poli, T. Priermeier, J. Am. Chem. Soc. 1994, 116, 4989.

[21] M. Prassas, L. L. Hench, in Ultrastructure Processing of Ceramics, Glasses, and Composites (Eds: L. L. Hench, D. R. Ulrich), Wiley, New York 1984, Ch. 9.

[22] T. E. Glassman, M. G. Vale, R. R. Schrock, Organometallics 1991, 10, 4046.

[23] M. M. B. Holl, P. T. Wolczanski, G. D. Van Duyne, J. Am. Chem. Soc. 1990, 112, 7989.

[24] Z. Gebeyehu, F. Weller, B. Neumüller, K. Dehnicke, Z. Anorg. Allg. Chem. 1991, 593, 99.

[25] Selected binding energies of tungsten nitride powder [eV]: W(4 f7/2),

33.2; W(4 f5/2), 34.5; Cl(2p), 199.2; N(1s), 398.3 (major), 400.6 (minor).

After the powder was heated at 573 K in vacuum for 2 h, the Cl(2p) and the minor N(1s) signals decreased to negligible. The N/W ratio is 0.5.

[26] H. W. Lam, G. Wilkinson, B. Hussain-Bates, M. B. Hursthouse, J. Chem. Soc., Dalton Trans. 1993, 1477.

A Novel Pathway to PbSe Nanowires at Room

Temperature**

By Wenzhong Wang,* Yan Geng, Yitai Qian, Mingrong Ji,

and Xianming Liu

Recently, one-dimensional (1D) structures with

meter diameter, such as nanowires (nanorods) and

nano-tubes, have attracted considerable attention due to their

special properties.

[1±13]

_{Compared with}

micrometer-diam-eter whiskers, they are expected to have remarkable

me-chanical properties, including electrical, optical, and

mag-netic properties that are in principle tunable by varying the

diameter and chirality.

[14,15]

_{These new nanoscale materials}

have potential applications in both mesoscopic research

and development of nanodevices. Previous work in this

field focused on carbon nanowires and nanotubes, which

were the by-product of fullerene research.

[16]

Convention-ally, carbon nanowires or nanotubes can be grown in an arc

discharge at a temperature of 3000 K,

[17,18]

_{by thermal}

de-position of hydrocarbons,

[19]

_{or vapor±liquid±solid (VLS)}

growth.

[3,4,12]

_{Comparatively little research has been carried}

out on other 1D materials and nearly all the previous

meth-ods of preparing nanowires or nanotubes require extreme

conditions. Therefore, one of the important goals of

materi-als scientists is to prepare nanoscale materimateri-als under milder

conditions. Here we report a novel route to PbSe

nano-wires: PbCl

2

, Se, and KBH

4

were kept in a sealed flask at

room temperature for 4 h using ethylenediamine as the

sol-vent. The study on PbSe is meaningful because it could be

widely used for IR sensors,

[20]

_{solar cells, infrared}

detec-tors,

[21]

_{chemical sensors,}

[22]

_{and so on. To our knowledge,}

the method here is the mildest route so far to produce

nanowires and it is reasonable to assume that other

nano-scale materials can be obtained by a similar process except

that so far only the reactant PbCl

2

has been substituted by

other MCl

n

compounds.

In a standard experimental procedure, an appropriate

amount of Se powder, PbCl

2

, and KBH

4

were placed in a

flask, which was filled with ethylenediamine up to 90 % of

its volume. The flask was then sealed and maintained at

about 10 C for 4 h. The precipitate was filtered and

washed with distilled water, the black product was

col-lected, and, finally, dried in vacuum at about 10 C for 12 h.

X-ray powder diffraction (XRD) was used to

character-ize the product. It was collected on a Rigaku D/max gA

ro-_______________________

±

[*] Dr. W. Wang, Dr. Y. Qian, Dr. M. Ji, Dr. X. Liu Structure Research Laboratory

University of Science and Technology of China Hefei, Anhui 230026 (People's Republic of China) Dr. W. Wang, Dr. Y. Geng, Dr. Y. Qian Department of Chemistry

University of Science and Technology of China Hefei, Anhui 230026 (People's Republic of China)

[**] This work is supported by the Chinese National Natural Science Fund and National Nanometer Materials Climbing Program.