NANO LETTERS
Synthesis and Characterization of Dilute Magnetic Semiconductor Manganese-Doped Indium Arsenide Nanocrystals
2003 Vol. 3, No. 10 1441-1447
Cynthia A. Stowell, Robert J. Wiacek, Aaron E. Saunders, and Brian A. Korgel* Department of Chemical Engineering, Texas Materials Institute and Center for Nano- and Molecular Science and Technology, The UniVersity of Texas, Austin, Texas 78712-1062 Received June 18, 2003; Revised Manuscript Received August 19, 2003
ABSTRACT Organic monolayer-passivated Group III−V dilute magnetic semiconductor (DMS) MnxIn1-xAs nanocrystals, ranging from 2.2 to 10 nm in diameter, were synthesized by high-temperature colloidal arrested precipitation using the dehalosilylation of InCl3 and MnBr2 with tristrimethylsilylarsine in the coordinating solvent, trioctylphosphine. Doping levels up to x ) 0.05 (∼1020 cm-3) were obtainedsthree to 4 orders of magnitude higher than the thermodynamic Mn solubility in the InAs host lattice. Relatively size-monodisperse nanocrystals, with standard deviations ranging from ±11% to 17% as measured by TEM and small-angle X-ray scattering (SAXS) were achieved by size selective precipitation. The MnxIn1-xAs nanocrystals exhibit near-infrared photoluminescence (PL), with PL (i.e., exciton) energies red-shifted relative to InAs nanocrystals of equivalent size. Elemental analysis, magnetization, and electron spin resonance (ESR) measurements before and after a chemical surface exchange confirmed that the majority of dopant was located in the nanocrystal core. The doped magnetic nanocrystals behave as paramagnets down to the lowest measured temperature of 5 K, with total magnetic moments ranging from 2.0 to 3.7 µB/Mn atom that interact through antiferromagnetic superexchange, with exhange integrals (Jnn/kB) ranging from −2 to −4 K.
Introduction. The addition of impurity elements to a host lattice is a common approach for tuning electronic, optical, mechanical, and magnetic properties of materials. In the case of nanostructured materials, their small size and large surface area-to-volume ratios present a significant challenge to impurity doping. Extremely high doping concentrations must be achieved to incorporate a meaningful number of atoms into the structure. For example, 12 Mn impurity atoms in a 4 nm diameter InAs semiconductor nanocrystal corresponds to a doping concentration greater than 1020 cm-3 (compare this to typical n- and p-dopant levels in silicon of 1016 to 1018 cm-3!). Doping levels of this magnitude often exceed the solubility limits of the impurity element in the host lattice. Furthermore, high dopant solubility does not guarantee effective nanostructure doping. The very high surface areato-volume ratios in nanocrystals have been found to enhance impurity segregation to the particle surface through a “selfannealing” process in the core that intensifies with increasing synthesis temperature.1 Since reasonably high synthesis temperatures are typically required to achieve core crystallinity and high quality optical properties,2 this presents a challenge to nanocrystal doping.1,3 Nonetheless, other studies * Corresponding author: Phone (512) 471-5633; FAX (512) 471-7060; e-mail
[email protected] 10.1021/nl034419+ CCC: $25.00 Published on Web 09/11/2003
© 2003 American Chemical Society
have shown that surface segregation does not necessarily occur in all materials3 and can be eliminated in some cases by manipulating the reaction chemistry.1,4-7 Group III-V dilute magnetic semiconductor (DMS) nanocrystals, such as the Mn-doped InAs nanocrystals studied here, provide an interesting testbed for a study of impurity doping of nanostructures. The Mn solubility is lows in the range of 1016 to 1017 cm-3 (ref 8)syet Mn doping levels (∼1020 cm-3) far exceeding the solubility limit in GaAs and InAs hosts have been achieved by kinetically limited molecular beam epitaxy (MBE).9 The ability to achieve high dopant levels of magnetic impurity elements in these materials subsequently led to the discovery of unique properties, such as ferromagnetism and electric field dependent magnetic susceptibilities.10 These properties result from coupling between charge carriers and the Mn delectrons that rarely manifests itself in this manner in the more thoroughly studied Group II-VI DMS materials with very high Mn solubility due to the scarcity of charge carriers within those systems.11,12 Nanocrystals of Group III-V DMS materials could exhibit magnetic and optical properties modified by nanoscale dimensions. For example, enhanced electron-hole exchange interactions in quantum dots have been speculated to lead
to unique magnetic properties.13 In this context, ferromagnetic DMS thin films have been fabricated and studied; however, there has not been an example of DMS nanocrystals exhibiting ferromagnetism at any temperature.1,3,4,14-20 Due to the high Mn solubility in II-VI hosts, the synthetic nanocrystal research in the literature to date has focused almost entirely on Group II-VI DMS materials, including transition metal doped CdSe,1,4 ZnS,17 ZnSe3, CdS,19 and ZnO.6 Although the magnetic impurity in these materials produces properties such as magnetic circular dichroism and large Zeeman splitting, it is unlikely that this class of materials will give rise to ferromagnetism. Thus far, the only example of Group III-V DMS quantum dots has been synthesized by MBE, with a broad size and shape distribution.21 Colloidal nanocrystal synthesis could potentially yield nanocrystals with controlled size and shape, and the process could be scaled up to produce large quantities of material; however, it must be demonstrated that kinetic trapping of a magnetic impurity element in a crystalline semiconductor host is possible by arrested precipitation. In this paper, we present the colloidal synthesis of MnxIn1-xAs nanocrystals ranging from 2.2 to 10 nm in diameter, along with compositional, optical, and magnetic characterization. The MnxIn1-xAs nanocrystals were synthesized with Mn compositions as high as 2.5 atomic % (xMn ) 0.05). The nanocrystals exhibit size-dependent photoluminescence emission (PL) red-shifted with respect to InAs particles of equivalent size, and antiferromagnetic superexchange between Mn impurity atoms. In contrast to their Group II-VI counterparts, such as Mn-doped ZnSe and CdSe, these dopant levels are 2 orders of magnitude above the bulk solubility limit. Doping levels of this magnitude were accessed through the kinetic trapping during nanocrystal synthesis that occurs in the dehalosilylation synthesis reaction. MnxIn1-xAs nanocrystals were synthesized using a modification of the high-temperature dehalosilylation colloidal synthesis for InAs developed by Guzelian et al.22 Depending on the amount of nanocrystals desired for characterization, the synthesis was carried out in either a 1 mL stainless steel batch reactor or using standard airless procedures on a Schlenk line. The stainless steel reactors were loaded and sealed inside a nitrogen glovebox with 0.9 mL of a stock solution of 0.25 g InCl3 (Strem), 0.165 g tris(trimethylsilyl)arsine [(CH3)3Si]3As, prepared according to the procedure described by Wells et al.23) and 6.8 mL trioctylphosphine, (C8H17)3P (TOP, Aldrich). The reactor was placed in a preheated brass heating block at the synthesis temperature, which ranged between 250 °C and 280 °C (below the TOP boiling temperature of 290 °C). After reacting for 1 to 3 h, the particles were extracted from the cell with chloroform. Higher reaction temperature improved the crystallinity, while longer reaction times led to larger nanocrystals. Reaction byproducts were separated from the nanocrystals through alternating rinses with ethanol and chloroform. The nanocrystals were size selectively precipitated using chloroform and ethanol as the solvent-antisolvent pair. The procedure on the Schlenk line was slightly modified from the small 1442
Table 1. In, Mn, As Composition of TOP-Capped InMnAs Nanocrystals Determined by ICP-MS for Different Reaction Precursors of Identical Concentration and Reaction Conditions nanoparticle composition (%) precursor
Mn
In
As
MnBr2 MnCl2 Mn(II) phthalocyanine MnMe2
0.74 0.47 0.01
48.61 53.31 61.33 no reaction
50.65 46.22 38.66
batch reactions. In a three-necked flask under nitrogen, 20 mL TOP and 0.73 g InCl3 were stirred and heated to the reaction temperature (between 250 °C and 280 °C). Upon reaching the desired temperature, 0.71 g [(CH3)3Si]3As was injected by syringe. Under reflux, the reaction proceeded for 1 to 3 h. Mn doping was explored using a variety of precursors, including MnCl2, MnBr2, dimethylmanganese [Mn(CH3)2], and manganese(II) phthalocyanine (C32H16MnN8). Each precursor was added to the reaction mixture at a 0.16 M Mn concentration. The nanocrystals were characterized using high-resolution transmission and scanning electron microscopy (HRTEM and HRSEM) and small-angle X-ray scattering (SAXS). HRTEM was performed using a JEOL 2010F TEM at an operating voltage of 200 kV, and HRSEM images were acquired using a LEO 1530 SEM from samples drop cast on a glassy carbon substrate. Figures 1a and 1c show TEM images of InAs and Mn0.01In0.99As nanocrystals, respectively, after size selective precipitation. Prior to size selective precipitation, the InAs and MnxIn1-xAs nanocrystal size distributions were broad, with particles ranging from 2 to 10 nm in diameter. Size selective precipitation narrowed the size distribution to standard deviations about the average particle diameter (σ) less than (20% (as determined by both TEM and SAXS; see Figures 1b and 1d,24-30 with σ as low as (11% for the best samples). The Mn precursor chemistry significantly affects the doping, with MnBr2 giving the highest dopant concentrations. Table 1 shows the resulting dopant levels achieved from each Mn precursor.31 The addition of MnBr2 to the synthesis as an Mn source did not significantly affect the average particle diameter relative to the pure InAs synthesis. Dopant levels in 4 nm diameter TOP-capped MnxIn1-xAs nanocrystals reached as high as xMn ) 0.05. HRSEM, such as in Figure 1e, providing another confirmation of the overall quality of the MnxIn1-xAs nanocrystal synthesis. There was no evidence of a pure MnAs phase in any of the MnxIn1-xAs samples. To confirm whether the dopant was bound to the nanocrystal surface or located within the core of the nanocrystals, we performed a chemical surface exchange with pyridine on many MnxIn1-xAs samples. Mikulec et al.1 demonstrated that repeated exposure to pyridine removed surface-bound Mn from CdSe nanocrystals, as pyridine forms a stronger bond with the nanocrystal surface than TOP and strips the surface molecules as it replaces them. Following this approach, MnInAs nanocrystals were dissolved in approximately 1 mL of toluene and then stirred for 24 h in 15 mL of pyridine. The MnInAs nanocrystals were then Nano Lett., Vol. 3, No. 10, 2003
Table 2. Composition of InMnAs Particle Cores Determined by ICP-MS of Nonligand Exchanged, Size Selected Nanoparticles from the Same Batcha Mn
In
As
3.3 3.1 2.7 2.5 2.2
0.21 0.25 0.27 0.32 0.38
45.88 43.25 42.74 41.16 46.71
53.91 56.50 56.98 58.51 52.90
a
Figure 1. TEM images of (a) 4.5 nm TOP capped InAs nanocrystals and (c) 4.5 nm TOP capped Mn0.01In0.99As nanocrystals. (b) SAXS data of InAs nanocrystals dispersed in cyclohexane: (0) dp ) 4.4 nm, σ ) (0.18; (×) dp ) 4.2 nm, σ ) (0.14; (O) dp ) 3.6 nm, σ ) (0.20. (d) SAXS data for Mn0.003In0.997As nanocrystals in cyclohexane: (0) dp ) 3.8 nm, σ ) (0.15; (×) dp ) 3.7 nm, σ ) (0.17; (O) dp ) 3.76 nm, σ ) (0.11; (+) dp ) 3.64 nm, σ)(0.13; (4) dp ) 3.46 nm, σ ) (0.14. Size histograms determined by TEM agree with those found from SAXS to within (13%. (e) HRSEM image of 4 nm diameter Mn0.014In0.986As nanoparticles on a glassy carbon substrate.
precipitated using hexane as an antisolvent. A significant amount of Mn was found to be associated with the nanocrystal surface. For example, prior to ligand exchange, nanocrystals with xMn ) 0.048 decreased in Mn content to xMn ) 0.024. Approximately half the Mn dopant was located at the particle surface, as was typical for all of the samples synthesized by these procedures. Data supporting the effectiveness of a pyridine ligand exchange within this system is given in the Supporting Information. The Mn composition systematically increased with slower heat-up rates of the reaction vessel. For example, heat up times on the order of a few minutes gave xMn ) 0.015, whereas heat up times of 15 to 45 min led to significantly higher dopant levels of xMn ranging from 0.02 to 0.05. The Mn concentration also appeared to vary with particle Nano Lett., Vol. 3, No. 10, 2003
nanoparticle composition (%)
nanoparticle diameter, nm
The smaller nanocrystals contain proportionally more Mn.
Figure 2. Electron spin resonance (ESR) spectra measured at 115 K and 9.42 GHz: (a) dp ) 5 nm, TOP capped InAs nanocrystals in cyclohexane; (b) dp ) 5 nm, MnxIn1-xAs (x ) 0.01) TOP capped nanocrystals in cyclohexane; (c) dp ) 5 nm, pyridine capped MnxIn1-xAs (x ) 0.024) nanocrystals in powder.
diameter: smaller nanocrystals contained a higher Mn concentration, as shown in Table 2. During ligand exchange, a proportionally larger percentage of Mn was removed from the smaller particles due to their higher surface area-tovolume ratios and higher amount of surface-adsorbed Mn. Electron spin resonance (ESR, IBM Instruments, Inc. ER 300) spectroscopy provides another measure of Mn doping and the nature of the dopant. ESR spectra measured at 115 K are shown in Figure 2. The TOP-capped InAs and MnInAs nanoparticles were measured as concentrated dispersions in cyclohexane, and the pyridine-treated MnInAs nanocrystals were measured as a powder. The ESR spectra of MnInAs nanocrystals before and after chemical surface exchange revealed the characteristic sextet centered at H ) 3.3 kG (g ) 2.02) of the d5 configuration of Mn dopant.32,33 This configuration has been attributed to the localized trapping of an electron by a substitutional Mn2+ dopant.32 A reference sample of undoped InAs nanocrystals shows a flat background. The signal shape depends sensitively on the amount of Mn in the nanocrystals and the location of the ions in the particle. The hyperfine splitting found from curve b in Figure 1443
Figure 3. Room-temperature photoluminescence emission (PL) (λexc) 790 nm) and absorbance spectra for InAs and Mn0.02In0.98As nanocrystals dispersed in cyclohexane.
2 is 74 × 10-4 cm - 1, which is consistent with expectations for Mn tetrahedrally coordinated in an ionic lattice.33 As the Mn concentration increases from xMn ) 0.01 to xMn ) 0.024, the sextet signal weakens in the ESR spectra, as seen in curve c in Figure 2. Szczytko et al.34 have indicated that the absence of the hyperfine splitting results from increased Mn-Mn interactions at high dopant concentrations. The Mn dopant level of the sample with xMn ) 0.012 is consistent with dopant levels that have eliminated the hyperfine splitting in the ESR spectra obtained from thin MnInAs films.35 As a point of note, the ESR signal in curve c in Figure 2 could be observed at room temperature, while the sextet in the ESR spectra of InAs nanocrystals with lower Mn concentrations did not appear at temperatures exceeding 150 K. Figure 3 compares the absorbance (Cary 500 Scan, Varian) and PL (Fluorolog F-3 equipped with a Jorbin-Yvon/ SPEX InGaAs detector) spectra for size-selected InAs and MnInAs nanocrystals. All the PL spectra in Figure 3 were obtained at room temperature with an excitation wavelength of 790 nm. The exciton peak energies in the absorbance spectra and PL maxima are significantly blue-shifted from the bulk InAs band gap energy of 0.36 eV (3433 nm) due to quantum confinement. For InAs and MnInAs nanocrystals of equal 1444
Figure 4. Magnetization measurements of MnxIn1-xAs nanocrystals. (a) Temperature-dependent magnetic molar susceptibility measured under a magnetic field of 5000 Oe (the solid lines are the predicted values as calculated from eq 1 in the text) and (b) magnetization versus applied magnetic field measured at 5 K where (×) dp ) 4 nm, x ) 0.024 (after pyridine ligand exchange); (0) dp ) 4 nm, x ) 0.048; (+) dp ) 4.6 nm, x ) 0.014 (after pyridine ligand exchange); (O) dp ) 4 nm, x ) 0.02; (4) dp ) 4 nm InAs measured at 15 K.
size, the exciton peak energies appear to be similar, independent of Mn dopant. In contrast, Mn doping decreases the PL peak energy, relative to undoped InAs of equal size, and noticeably broadens the peak. The Mn-induced red shift in the PL peak energy is consistent with an Mn-related acceptor state, as has been found both experimentally and theoretically to lie approximately 0.1 eV above the valence band in the bulk.35,36 The increased PL peak breadth for the nanocrystals doped with Mn cannot be attributed to a difference in size distribution, as both TEM and SAXS confirmed similar size distributions for both samples; Nano Lett., Vol. 3, No. 10, 2003
however, one cannot completely rule out the effect of a compositional distribution in the sample. Nonetheless, the tail on the PL peak for the MnInAs nanocrystals is consistent with the additional impurity levels due to the Mn dopant.35,36 Temperature-dependent magnetization measurements of the MnInAs nanocrystals were conducted using a superconducting quantum interference device magnetometer (SQUID, Quantum Design). Thin films of nanocrystals were deposited onto a glass substrate and then scraped into a Lilly #4 Gelatin sample holder for the measurements. The temperaturedependent magnetization data shown in Figure 4a were obtained at a constant field at 5 kOe with temperatures ranging from 100 K to 5 K. The magnetization was also measured as a function of applied field at 5 K, as shown in Figure 4b. Pure InAs nanocrystals were measured as a baseline and found to be diamagnetic as expected, with a molar magnetic susceptibility χmolar ) - 49.2 × 10-6 cm3mol-1, very close to the literature value of χmolar ) 55.3 × 10-6 cm3mol-1.37,38 The MnInAs nanocrystals exhibit positive susceptibilities before and after chemical surface exchange, indicating conclusively that Mn incorporates into the nanocrystal core. Additionally, constant field temperature scans of InMnAs samples starting at 5 K produced no local maxima susceptibility values, indicating that there is not a blocking temperature associated with superparamagnetic materials. This also is strong evidence that there is no phase separation into MnAs, as it is ferromagnetic in the bulk and would be superparamagnetic on these length scales. Elemental analysis, ESR spectroscopy, and magnetization measurements confirm the most important result of this study, that kinetic trapping of Mn dopant in an InAs nanocrystal host lattice can be induced by choosing the appropriate reaction pathway for the synthesis. A closer analysis of the magnetization data provides information about the nature of the magnetic interactions in the MnInAs nanocrystals for comparison with the wellstudied films fabricated by kinetically controlled MBE. The molar susceptibilities are relatively low, ranging from χmolar ≈ 5 × 10-4 cm3mol-1 at room temperature up to χmolar ≈ 3 × 10-3 cm3 mol-1 at 5 K, indicating that the magnetic coupling is weak. An analysis of the magnetic interactions can be obtained by comparing the magnetic data with a hightemperature expansion of χmolar(T) to second order in inverse powers of temperature:39
χmolar(T) ) NA
(gµB)2 θ S(S + 1) 1 + , 3kBT T
[
]
(1)
where
θ)
S(S + 1) yMn 3kB
∑n znJn )
S(S + 1) yMn × 12Jnn. 3kB
(2)
In eqs 1 and 2, NA is Avogadro’s number, µB is the Bohr magneton, kB is Boltzmann’s constant, T is temperature, g is the Lande´ g-factor, S is the spin, yMn is the Mn mole Nano Lett., Vol. 3, No. 10, 2003
Table 3. Effective Bohr Magneton Number p, and the nearest Neighbor Exchange Integral Jnn Found by Fitting Eq 1 to Figure 5a diameter (nm)
xMn
ligand
p (µB/Mn atom)
Jnn/kB (K)
4 4 4.6 4
0.020 0.024 0.014 0.048
TOP Pyridine Pyridine TOP
2.2 3.66 2.73 3.03
-3.67 -4.29 -2.01 -2.12
fraction in the sample, zn is the number of nearest Mn neighbors to each Mn atom, Jn and Jnn are the distancedependent exchange integral and the nearest-neighbor exchange integral, respectively. Equation 1 was fit to the measured values of χmolar(T) shown in Figure 4a. From the high-temperature limit of eq 1, we can determine the effective number of Bohr magnetons p, contributed by each magnetic impurity atom to the magnetization: p ) g(JLS)xS(S + 1)µB where
[
]
3 1 S(S + 1) - L(L + 1) g(JLS) ) + 2 2 J(J + 1)
The value of g measured experimentally by ESR compares well to the value calculated by taking J ) S (total electronic angular momentum) and L ) 0 (orbital angular momentum), as is appropriate for Mn2+ within an InAs lattice: g(JLS) ) 2.39 The values of p and Jnn determined from these curve fits are shown in Table 3. Values of p range from 2.2 to 3.7 µB/Mn atom, which is significantly lower than p ) 5 (µB)/ (Mn atom) expected for Mn dopant with the d5 electron configuration. This lower value however is consistent with both theoretical and experimental findings. For example, Jain et al.36 computed a local magnetic moment of ∼4.1 µB/Mn atom as the Mn impurity in the lattice serves as an acceptors an idea that appears to be consistent with PL spectra. Experimentally, however, local magnetic moments as low as 2.4 µB/Mn atom have been measured for MnxGa1-xAs thin films.40,41 Although the low magnetic moments of the Mn impurity in the III-V semiconductor hosts still remain a subject of debate, it is believed that AsGa antisite defects could contribute to this behavior.40,41 As seen in Table 2, the MnInAs nanocrystals are As rich in the core, which could indicate a significant number of AsGa antisite defects. A high concentration of Mn atoms would energetically favor the formation of AsGa defects through acceptor-donor interactions, which would be consistent with the Mn d5 electron configuration measured by ESR spectroscopy in Figure 2. According to Bergqvist et al.,40 AsGa antisite defects induce antiferromagnetic coupling between Mn in the lattice. From Figure 4a, the nearest-neighbor exchange integrals, Jnn/kB, ranged from -2.0 to -4.3 K, indicating antiferromagnetic superexchange interactions between the Mn impurity atoms in the InAs host lattice. These values are higher than the measured bulk value of Jnn/kB ) - 1.6 K found for n-type 1445
Mn-doped InAs,42 yet is lower than Jnn/kB typical in Mndoped II-VI materials, such as CdMnTe (Jnn/kB ) - 8 K).11 In conclusion, we have demonstrated that kinetic trapping of dopants in a nanocrystal host lattice is possible, even in high-temperature arrested precipitation reactions. MnxIn1-xAs nanocrystals ranging from 2 to 10 nm in diameter were synthesized with up to xMn ) 0.05. Surface exchange and magnetic measurements confirmed that much of the dopant resides in the nanocrystal core and modifies the magnetic properties of the host material through antiferromagnetic superexchange interactions. On the basis of studies of bulk p-doped Group III-V DMS materials fabricated by MBE, the doping levels achieved here are sufficiently high to expect ferromagnetism.43 However, at temperatures as low as 5 K, only paramagnetic behavior was observed for the nanocrystals with the appearance of antiferromagnetic interactions between the Mn dopant atoms at lower temperatures. Ferromagnetism has generally been absent in bulk n-doped MnInAs films.42 In the nanocrystal, each Mn atom contributes significantly less than 5 µB/Mn atom (i.e., 2.2 to 3.7 µB/Mn atom) to the total magnetization of the nanocrystals, which may result from the donor-acceptor interactions between Mn2+ or dangling bonds and AsGa antisite defects which would be consistent with n-doping. The MnInAs quantum dot PL was redshifted and broader than the pure InAs PL due to the presence of the magnetic impurity. Potentially, kinetic impurity trapping in nanocrystals could enable the study of a wide range of doped nanostructures, including other DMS materials, such as Mn-doped GaP which has exhibited room-temperature ferromagnetism in the bulk.44,45 Acknowledgment. We thank Michael Sigman and Jairo Sinova for stimulating discussions, John Lansdown for ICPMS measurements, and Ronald Dass for his assistance with the magnetic measurements, and Alan H. Cowley for use of his facilities in the synthesis of [Si(CH3)]3As. We also thank the NSF, the Welch Foundation and the Texas Higher Education Coordinating Board for financial support. Supporting Information Available: Data (figures S1, S2, and S3) supporting the effectiveness of a pyridine ligand exchange within this system. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) Mikulec, F. V.; Kuno, M.; Bennati, M.; Hall, D. A.; Griffin, R. G.; Bawendi, M. G. J. Am. Chem. Soc. 2000, 122, 2532-2540. (2) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993, 115, 8706-8715. (3) Norris, D. J.; Yao, N.; Charnock, F. T.; Kennedy, T. A. Nano Lett. 2001, 1, 3-7. (4) Hanif, K. M.; Meulenberg, R. W.; Strouse, G. F. J. Am. Chem. Soc. 2002, 124, 11495-11502. (5) Raola, O. E.; Strouse, G. F. Nano Lett. 2002, 2, 1443-1447. (6) Radovanovic, P. V.; Norberg, N. S.; McNally, K. E.; Gamelin, D. R. J. Am. Chem. Soc. 2002, 124, 15192-15193. (7) Jun, Y.; Jung, Y.; Cheon, J. J. Am. Chem. Soc. 2002, 124, 615617. (8) Illegems, M.; Dingle, R.; Rupp, J. L. W. J. Appl. Phys. 1975, 46, 3059-3065. (9) Munekata, H.; Ohno, H.; Vonmolnar, S.; Harwit, A.; Segmuller, A.; Chang, L. L. J. Vac. Sci. Technol. B 1990, 8, 176-180. 1446
(10) Ohno, H.; Chiba, D.; Matsukura, F.; Omiya, T.; Abe, E.; Dietl, T.; Ohno, Y.; Ohtani, K. Nature 2000, 408, 944-946. (11) Furdyna, J. K. J. Appl. Phys. 1988, 64, R29-R64. (12) Ohno, H. Science 1998, 281, 951-956. (13) Hoffman, D. M.; Meyer, B. K.; Ekimov, A. I.; Merkulov, I. A.; Efros, A. L.; Rosen, M.; Couino, G.; Gacoin, T.; Boilot, J. P. Solid State Commun. 2000, 114, 547-550. (14) Shim, M.; Wang, C.; Norris, D. J.; Guyot-Sionnest, P. MRS Bull. 2001, 1005. (15) Kennedy, T. A.; Glaser, E. R.; Klein, P. B.; Bhargave, R. N. Phys. ReV. B 1995, 101, 452. (16) Khosravi, A. A.; Kundu, M.; Kuruvilla, B. A.; Shekhawat, G. S.; Gupta, R. P.; Sharma, A. K.; Vyas, P. D.; Kulkarni, S. K. Appl. Phys. Lett. 2995, 67, 2506. (17) Borse, P. H.; Srinivas, D.; Shinde, R. F.; Date, S. K.; Vogel, W.; Kulkarni, S. K. Phys. ReV. B 1999, 60, 8659-8664. (18) Qi, J.; Guo, X.; Sakurai, K.; Masumoto, Y. Scr. Mater. 2001, 44, 2315-2319. (19) Feltin, N.; Levy, L.; Ingert, D.; Pileni, M. P. J. Phys. Chem. B 1999, 103, 4-10. (20) Jeon, H. C.; Jeong, Y. S.; Kang, T. W.; Kim, T. W.; Chung, K. J.; Jhe, W.; Song, S. A. AdV. Mater. 2002, 14, 1725-1727. (21) Guo, S. P.; Ohno, H.; Shen, A.; Matsukura, F.; Ohno, Y. Appl. Surf. Sci. 1998, 132, 797-802. (22) Guzelian, A. A.; Banin, U.; Kadavanich, A. V.; Peng, X.; Alivisatos, A. P. Appl. Phys. Lett. 1996, 69, 1432-1434. (23) Wells, R. L.; Self, M. F.; Johansen, J. D.; Laske, J. A.; Aubuchon, S. R.; Jones, L. J. Inorg. Synth. 1997, 31, 150-158. (24) SAXS measurements were performed using a rotating copper-anode generator (Bruker Nonius, Molecular Metrology, Inc.) operated at 3.0 kW on InAs and InMnAs nanocrystals dispersed in cyclohexane. Scattered photons were collected on a multiwire gas-filled detector calibrated using silver behenate (CH3(CH2)20COOAg). All experimental data were corrected for background scattering and sample absorption. The term q is the magnitude of the scattering vector, q ) 4π/λ sin(θ), where λ is the wavelength (0.154 nm) and 2θ is the scattering angle. The angle-dependent scattered intensity I(q), from a dilute dispersion of noninteracting nanocrystals of radius R depends on the normalized size distribution N(R), and the shape factor P(qR),24 I(q) ∝ ∫N(R)P(qR)R6 dR.25-28 The average diameter Ravg and size distribution were determined using the shape function for spherical particles,24,29 P(qR) ) [3 (sin(qR) - qR cos(qR))/((qR)3)]2, and a Gaussian size distribution, N(R) ) 1/(σx2π) exp[-(R - Ravg)2/ 2σ2], where σ is the standard deviation. (25) Small-Angle X-ray Scattering; Glatter, O., Kratky, O., Eds.; Academic Press Inc.: New York, 1982. (26) X-ray Characterization of Materials; Lifshin, E., Ed.; Wiley-VCH: New York, 1999. (27) Korgel, B. A.; Fitzmaurice, D. Phys. ReV. B 1999, 59, 14191-14201. (28) Mattoussi, H.; Cumming, A. W.; Murray, C. B.; Bawendi, M. G.; Ober, R. Phys. ReV. B 1998, 58, 7850-7863. (29) Korgel, B. A.; Fullam, S.; Connolly, S.; Fitzmaurice, D. J. Phys. Chem. B 1998, 102, 8379-8388. (30) Guinier, A.; Fournet, G. Small-Angle Scattering of X-rays; Wiley: New York, 1955. (31) Elemental analysis was obtained using a micromass platform inductively coupled plasma mass spectrometer (ICP-MS). Samples were prepared by drop casting the nanocrystals into a plastic centrifuge tube. Concentrated nitric acid was added to dissolve the particles and then diluted with deionized water to provide manganese, arsenic, and indium concentrations within the required ranges based upon calibration curves created from elemental standards. (32) Szczytko, J.; Twardowski, A.; Palczewska, M.; Jablonski, R.; Furdyna, J.; Munekata, H. Phys. ReV. B 2001, 63, 085315. (33) Almelch, N.; Goldstein, B. Phys. ReV. 1962, 128, 1568. (34) Szcytko, J.; Twardowski, A.; Palczewska, M.; Jablonski, R.; Furdyna, J.; Munekata, H. Phys. Re. B 2001, 63, 085315. (35) Linnarsson, M.; Janzen, E.; Monemar, B.; Kleverman, M.; Thilderkvist, A. Phys. ReV. B 1997, 55, 6938-6944. (36) Jain, M.; Kronik, L.; Chelikowsky, J. R.; Godlevsky, V. V. Phys. ReV. B. 2001, 64, 245205. (37) To obtain χmolar(T) in cgs units of emu Oe-1 g-1 mol-1 (equivalent to units of cm3 mol-1), the measured magnetization (units of emu) was divided by the sample weight (units of g) and the applied magnetic field (units of Oe) to obtain the mass magnetic susceptibility, χmass(T). Since the sample mass also includes the weight of the Nano Lett., Vol. 3, No. 10, 2003
(38) (39) (40) (41)
ligands, the measured susceptibility was corrected by dividing through by the core inorganic volume fraction in the sample, φV: χmass,core(T) ) χmass(T)/φV. The term φV was typically 0.3, and χmolar(T) was then obtained multiplying χmass,core(T) by the core molecular weight. Sahu, T. Phys. ReV. B 1991, 43, 2415-2418. Ashcroft, N. W.; Mermin, N. D. Solid State Physics; Harcourt College Publishers: New York, 1976. Bergqvist, L.; Korzhavyi, P. A.; Sanyal, B.; Mirbt, S.; Abrikosov, I. A.; Nordstrom, L.; Smirnova, E. A.; Mohn, P.; Svendlindh, P.; Eriksson, O. Phys. ReV. B 2003, 67, 205201. Korzhavyi, P. A.; Abrikosov, I. A.; Smirnova, E. A.; Bergqvist, L.; Mohn, P.; Mathieu, R.; Svendlindh, P.; Sadowski, J.; Isaev, E. I.; Vekilov, Y. K.; Eriksson, O. Phys. ReV. Lett. 2002, 88, 187202.
Nano Lett., Vol. 3, No. 10, 2003
(42) Von Molnar, S.; Munekata, H.; Ohno, H.; Chang, L. L. J. Magn. Magn. Mater. 1991, 93, 356-364. (43) Ohno, H.; Munekata, H.; Penney, T.; Von molnar, S.; Chang, L. L. Phys. ReV. Lett. 1992, 68, 2664-2667. (44) Overberg, M. E.; Gila, B. P.; Thaler, G. T.; Abernathy, C. R.; Pearton, S. J.; Theodoropoulou, N. A.; McCarthy, K. T.; Arnason, S. B.; Hebard, A. F.; Chu, S. N. G.; Wilson, R. G.; Zavada, J. M.; Park, Y. D. J. Vac. Sci. Technol. B 2002, 20, 969-979. (45) Pearton, S. J.; Overberg, M. E.; Thaler, G. T.; Abernathy, C. R.; Kim, J.; Ren, F.; Theodoropoulou, N.; Hebard, A. F.; Park, Y. D. Phys. Status Solidi A 2003, 195, 222-227.
NL034419+
1447
