C O MMU N I C A T I O N S
Acknowledgment. We thank Hyungrok Kim and Susan Kau-
zlarich for assistance with experiments performed at UCD, Ron
Baird and Eric McCullen for help with XPS, and Charles H. Winter
for critical reading of this manuscript. Acknowledgments are made
to the donors of the Petroleum Research Fund, administered by
the ACS, and the National Science Foundation (CAREER award
DMR-0094273 and IGERT-970952) for financial support of this
research.
Supporting Information Available: XPS spectra of Fe-P samples
reduced on mica (PDF). This material is available free of charge via
the Internet at http://pubs.acs.org.
References
(
(
(
1) Leslie-Pelecky, D. L.; Rieke, R. D. Chem. Mater. 1996, 8, 1770-1783.
2) Alivisatos, A. P. J. Phys. Chem. 1996, 100, 13226-13239.
3) Guzelian, A. A.; Katari, J. E. B.; Kadavanich, A. V.; Banin, U.; Hamad,
K.; Juban, E.; Alivisatos, A. P.; Wolters, R. H.; Arnold, C. C.; Heath, J.
R. J. Phys. Chem. 1996, 100, 7212-7219.
Figure 3. AFM topographic study of the particles after annealing. Surface-
confined precursors were calcined at 700 °C for 1 h, in 7% hydrogen in
nitrogen. (A) Precursor concentration: 2.3 mg/mL. (B) Cursor profile from
(
4) (a) Mi c´ i c´ , O. I.; Curtis, C. J.; Jones, K. M.; Sprague, J. R.; Nozik, A. J.
J. Phys. Chem. 1994, 98, 4966-4969. (b) Coffer, J. L.; Johnson, M. A.;
Zhang, L.; Wells, R. L.; Janik, J. F. Chem. Mater. 1997, 9, 2671-2673.
(c) Kher, S. S.; Wells, R. L. Nanostruct. Mater. 1996, 7, 591-603. (d)
Peng, X.; Wickham, J.; Alivisatos, A. P. J. Am. Chem. Soc. 1998, 120,
(
(
A). (C) Precursor concentration: 0.19 mg/mL. (D) Cursor profile from
B).
5343-5344.
reduction and confining the particles on surfaces, respectively.14
Accordingly, particles were reductively annealed directly on the
mica substrate.
(5) Hulliger, F. Struct. Bonding 1968, 4, 83-229.
(
6) (a) Previous reports of the synthesis of transition metal pnictide nano-
6b
particles have been limited to organometallic pyrolysis routes and
6
c,d
solvothermal syntheses. (b) Lukehart, C. M.; Milne, S. B.; Stock, S.
R. Chem. Mater. 1998, 10, 903-908. (c) Xie, Y.; Lu, J.; Yan, P.; Jiang,
X.; Qian, Y. J. Solid State Chem. 2000, 155, 42-45. (d) Xie, Y.; Lu, J.;
Yan, P.; Jiang, X.; Qian, Y. Chem. Lett. 2000, 114-115.
The results of directly annealing surface-confined phosphate
nanoparticle precursors at 700 °C are illustrated in Figure 3.15 At
high initial coverage, the resulting iron phosphide particles are 3D
aggregates ranging from 0.8 to 37 nm (average 13.4 ( 8.7 nm) as
shown in Figure 3A. Quantitative measurements of two relatively
large particles are shown in Figure 3B. The scale of these 3D
aggregates represents a significant growth in particle size relative
to the phosphate precursor employed for this experiment (2.18 (
(
7) Gopalakrishnan, J.; Pandey, S.; Rangan, K. K. Chem. Mater. 1997, 9,
2
113-2116.
(8) Fe P is ferromagnetic, T
25 K; other phases in the Fe-P phase diagram: FeP
2
c
) 266 K, and FeP is antiferromagnetic, T
N
)
1
Fe
2
(semicond.) and
5
3 c
P (ferromagnetic, T ) 716 K).
(9) (a) FeCl was treated with crystalline H PO and tri-n-octylamine in tris-
3
3
4
2
-ethylhexylphosphonate (TEHP) at 160 °C overnight, followed by
precipitation with methanol. The resultant nanoparticles have an Fe:P ratio
of 0.92 ( 5% (ICP-MS) and are soluble in pyridine, toluene, or hexane.
(
b) Riwotzki, K.; Meyssamy, H.; Kornowski, A.; Haase, M. J. Phys. Chem.
1.3 nm). However, at low coverage, phosphide nanoparticles exhibit
B 2000, 104, 2824-2828.
a near-spherical morphology with an average size of 1.41 ( 0.5
nm, as shown in Figure 3C and D. Assuming spherical geometry,
we found that annealing results in 35% reduction in particle
diameter (∼75% reduction in volume), consistent with oxygen loss
during the transformation of phosphate to phosphide (predicted
volume reduction from FePO : 72% for FeP, 79% for Fe P). On
4 2
the basis of data obtained for sintered samples (Figure 2), the
expected phase at 700 °C is FeP.16
The reduction of phosphorus is further confirmed in the XPS
spectra (Supporting Information). The P 2p XPS spectrum reveals
two peaks at 132.7 and 128.4 eV, respectively, with an intensity
ratio of 70:30. The lower energy peak is consistent with the binding
energy for FeP (129.5 eV),17 demonstrating that some reduction
has occurred. The presence of oxidized phosphorus (132.7 eV) is
not unexpected as the samples are exposed to air during introduction
to the XPS chamber. These data, including the observation of
oxidation, are consistent with those observed for other phosphide
nanoparticles (e.g., InP).3
In conclusion, the preparation of iron phosphide nanoparticles
presented here represents the first report of the use of preformed
nanoparticle precursors for the synthesis of nanoparticulate pnic-
tides. However, this general methodology is expected to be
applicable to a wide range of transition metals and pnicogens. To
date, we have successfully extended this approach to Ni-As
nanoparticulate phases and are also investigating other strategies
for limiting aggregation, including solution reduction routes, to
further expand the versatility of the method.
(
10) (a) Samples for AFM were prepared by depositing a 10 µL drop of
phosphate nanoparticle solution onto freshly cleaved mica (0001) surfaces,
followed by air-drying. The AFM scanner is home-constructed and
10b,c
employs the optical beam-deflection configuration.
controllers and software are from RHK Technology, Inc. Commercially
available Si cantilevers with force constants of 0.1 N/m (microsharp-
The electronic
3
N
4
ened cantilever, Thermo Microscopes) were used for imaging. The AFM
scanner was calibrated using Au(111) periodicity and step height. Contact-
resonance imaging (CRI) was used to minimize adhesion between the tip
10d
and surface, thus circumventing stick-slip motion during scanning. (b)
Kolbe, W. F.; Ogletree, D. F.; Salmeron, M. B. Ultramicroscopy 1992,
42, 1113-1117. (c) Liu, G.-Y.; Fenter, P.; Chidsey, C. E. D.; Ogletree,
D. F.; Eisenberger, P.; Salmeron, M. J. Chem. Phys. 1994, 101, 4301-
4306. (d) Wadu-Mesthrige, K.; Amro, N.; Garno, J. C.; Cruchon-Dupeyrat,
S.; Liu, G.-Y. Appl. Surf. Sci. 2001, 175-176, 391-398.
11) Lateral AFM sizes are a convolution of tip geometry. Without aggressive
deconvolution, it is more reliable to use measurements of nanoparticle
heights. Over 150 cursor profiles were analyzed for average particle size
determinations.
(
(12) Safety Note: Because of the use of hydrogen mixtures (7-15% in Ar or
2 3
N ), as well as the possibility of producing toxic byproducts (e.g., PH ),
all reduction reactions were either conducted in, or vented to, a fume
hood.
(
(
(
13) Powder diffraction file (PDF) data were obtained from the ICDD-PDF
database, release 2000.
14) Chen, Z. X.; Smith, G. C.; Putman, C. A. J.; ter Voert, E. J. M. Catal.
Lett. 1998, 50, 49-57.
15) AFM imaging of plain mica substrates indicates that the surface remains
atomically flat under the annealing conditions employed here.
(16) Although both iron and phosphorus are detected, the stoichiometry of the
2
reduced nanoparticle phase on mica (FeP, Fe P) has not been determined
due to sensitivity issues with conventional methods (e.g., XPS) on these
8
highly dilute samples. Planned magnetic studies will probe whether the
reduction temperature dependence of the phase transformation is identical
to that observed in the bulk, or if there are differences due to size effects,
and this will be presented in a full paper.
(
17) Binding energy data are available online at http://srdata.nist.gov/xps/.
JA028180V
J. AM. CHEM. SOC.
9
VOL. 125, NO. 14, 2003 4039