Recently, transition metal silicide nanowires have been widely studied [1-9] for their utilization in semiconductor device technologies. Low-resistivity silicides, such as TiSi2, CoSi2, and NiSi, have been applied for interconnection in CMOS devices . The group of refractory semiconducting silicides, composed of silicon and metals, have different physical properties that are useful and importantly meaningful. Among them, semiconducting silicides, such as CrSi2 and ß-FeSi2, with a narrow energy gap (0.1 to 0.9 eV) have been extensively investigated for their potential use in silicon-integrated optoelectronic devices  such as LEDs [12,13] and IR detectors . In particular, CrSi2 is a narrow bandgap (0.35 eV) semiconductor [15-17], offering applications in the Schottky barrier solar cell technology . Hexagonal CrSi2 with a C40-type structure has a high melting point and excellent resistance to oxidation, deformation, and stretching, being considered to be a potential structural material for aerospace and energy generation industries . Additionally, it is a thermoelectric conversion component that could be applied to generate electric power at high temperatures ; the figure of merit (ZT) of CrSi2 has been measured to be 0.25 at 900 K . CrSi2 also has good field emission with relatively low work function (3.9 eV)  as compared with generally studied field emission materials such as CNTs (5 eV)  and ZnO (5.3 eV) . With excellent intrinsic properties of CrSi2, one-dimensional CrSi2 nanowires are expected to improve field emission performances by bulk and thin film CrSi2. Though there have been some previous studies on CrSi2 nanowires [25-28], two special aspects can be found in this research. Firstly, we conducted a more systematic study on the influences of each processing parameter on growth. Secondly, we provided a low-cost and simple method to synthesize high-quality CrSi2 nanowires with very good physical properties.
In our experiments, we synthesized chromium disilicide nanowires with chemical vapor deposition (CVD) processes. Single-crystal Si (001) wafers, the native oxide of which was etched by BOE solution, were substrates. The metal source was from hydrous chromium chloride (CrCl3 · 6H2O) powders, and the flow gas is Ar gas (99.99%). The CrCl3 · 6H2O powders were put in the upstream zone of the furnace, where the temperature ranged from 700°C to 800°C, while the silicon (001) substrates were put in the downstream zone with the same temperature range. During the growth process, with oxygen environment, CrSi2 nanowires may transform to be CrSi2(core)/SiO2(shell) nanowires due to oxidation. To understand what factors influence the growth of chromium disilicide nanowires, we varied reaction time and temperatures of substrates and the metal source. Scanning electron microscopy (SEM), X-ray diffraction (XRD), and transmission electron microscopy (TEM) studies were conducted for morphology observation and structure identification of the nanowires. Additionally, physical properties, including magnetism (SQUID), photoluminescence (PL), and field emission (Keithley-237), were measured.
Results and discussion
In this work, we controlled different parameters to realize how they influence the nanowires’ growth, morphology, and physical properties. With source and substrate at 700°C and the flow gas of 120 sccm, we obtained dense CrSi2 nanowires with a length of approximately 20 μm as shown in Figure 1a by chemical vapor deposition. Interestingly, in Figure 1b, the nanowires grew from the particle with almost coherent growth direction and the morphology was rare. XRD analysis in Figure 1c shows (111), (003), and (112) major plane peaks, indicating that the nanowires have a C40 hexagonal structure. The TEM image of Figure 2a shows that the nanowires are 10 to 50 nm in diameter. In Figure 2b, the high-resolution transmission electron microscopy (HRTEM) image and the corresponding fast Fourier transform (FFT) pattern in the inset identifies the materials to be single-crystal CrSi2 nanowires of a hexagonal structure with lattice constants, a = 0.4428 nm and c = 0.6369 nm (JCPDS card no. 35–0781); the growth direction is , and the interplanar spacing of plane (003) is 0.2098 nm. Additionally, we tried 750°C with hydrogen as reducing atmosphere and obtained Cr5Si3 nanowires of approximately 10 μm in length and of a different morphology as shown in Figure 1d. In Figure 1e, we found that the nanowires grew from nanoparticles again. XRD analysis in Figure 1f shows two phases, CrSi2 and Cr5Si3; for further investigation on the atomic structures of the nanowires, we conducted TEM analysis as shown in Figure 2. From the TEM image of Figure 2c, the nanowire was of approximately 80 nm in diameter. The HRTEM image and the corresponding FFT pattern in the inset of Figure 2d confirm that the single-crystal Cr5Si3 nanowire has a BCT D8m structure with lattice constants, a = 0.9165 nm and c = 0.4638 nm (JCPDS card no. 51–1357); also, the nanowire is with  growth direction, and the interplanar spacing of plane (200) is 0.4571 nm.
The growth mechanism of the chromium silicide nanowires in this study is interesting. Figure 3 is the schematic illustration of the growth mechanism, showing the proposed growth steps of the CrSi2 nanowires. When the system was heated below 700°C, CrCl3 · 6H2O transformed to CrCl3 and H2O:
The CrCl3 gas molecules then agglomerated on the silicon substrate. As the system temperature reached the reaction temperature, 700°C, CrCl3 gas reacted with the silicon substrate to form CrSi2 nanoparticles and SiCl4 based on step (1) of Figure 3:
The SiCl4 product then reacted with CrCl3(g) to form CrSi2, following step (2) of Figure 3:
Notably, the CrSi2 nanowires precipitated from polygonal particles, and the growth direction seems consistent as shown in Figure 1b. The nanowires and polygonal particles may have the same stacking plane, (003), based on our TEM analysis, and nanowires grew from voids and defects on the surface of any polygonal particles with <001 > growth direction, following step (3) of Figure 3 as shown in a SEM image of Additional file 1: Figure S1. We conducted experiments with the heating times of 1.5, 4, and 12 h at 700°C, obtaining the corresponding results shown in Figure 4a, b, c, respectively. We found nanowires and particles at 1.5 h, more nanowires growing from particles at 4 h, and dense nanowires appearing with buried particles at 12 h, respectively. With a longer duration, more nanowires can overcome the activation energy, successfully nucleate, and grow to be nanowires, contributing to CrSi2 nanowires of a high density. According to the observations, we proposed that the mechanism of the nanowire growth is a self-catalytic process.
As the substrate temperature was at 750°C, CrCl3 gas reacted with H2 gas and the silicon substrate to form Cr5Si3 nanoparticles, HCl, and SiCl4, following step (i) of Figure 3:
The SiCl4 also reacted with CrCl3 to form CrSi2, which is the reason why the XRD analysis shows both CrSi2 and Cr5Si3 phases.
Also, we investigated the influence of the carrier gas flow rate when synthesizing chromium silicide nanowires. We conducted experiments at the gas flow rate of 60, 120, and 240 sccm at 700°C, obtaining the corresponding results shown in Figure 5a, b, c, respectively. It can be found that chromium disilicide nanowires appeared without particles at 60 sccm and with few particles at 120 sccm and that the morphology gradually transformed from nanowires to films at 240 sccm.
The CVD synthesis system can be divided into three sub-systems, which are momentum control system, mass transfer control system, and surface reaction control system. At a lower gas flow rate, mass transfer control system would be the main reaction mechanism, with which gas adsorption and desorption occurred on the Si wafer and fabrication of chromium silicide nanowires was preferred. On the other hand, at a higher gas flow rate, surface reaction control system would be the main reaction mechanism, with which CrCl3 reacted on the Si wafer surface by chemical vapor deposition; thus, chromium silicide films appeared.
In addition to understanding the growth behaviors of the chromium silicide nanowires, we explored their physical properties. Figure 6 is the field emission measurements for CrSi2 NWs, showing the plot of the current density (J) as a function of the applied field (E) with the inset of the ln(J/E2)-1/E plot. The sample was measured in a vacuum chamber pump to approximately 10−6 Torr. According to the Fowler-Nordheim (F-N) plot and the Fowler-Nordheim equation:
where J is the current density, E is the applied electric field, φ is the work function, and A, B are constants, respectively. We put +1,000 V on the sample with a 100-μm spacing between the anode and cathode, and we defined the turn-on field could obtain a current density of 10 μA/cm2 and the turn-on field we measured for CrSi2 nanowires was 7.5 V/μm. The field enhancement factor ß has been calculated to be 1,366 from the slope of ln(J/E 2) = ln(Aß 2/φ) − Bφ 3/2/ßE (for CrSi2, φ = 3.9 eV ), demonstrating that CrSi2 NWs are promising emitters. The outstanding field emission properties of CrSi2 NWs are attributed to their metallic property and special one-dimensional geometry with a high aspect ratio as compared with those of many other materials.
On magnetization analysis for chromium disilicide nanowires coated with a silicon oxide layer of a few nanometers in thickness, we prepared samples of 2.5 mm × 2.5 mm with the applied magnetic field of ±3,000 Oe perpendicular to the substrates. Notably, Figure 7 shows that the CrSi2/SiOx nanowires grown here were found to be ferromagnetic with the saturation magnetization of 8 × 10−7 emu, M R, remanence, of 2 × 10−7 emu, and H C, coercive force, of about 179 Oe, respectively, which is different from the antimagnetic behavior in CrSi2 and SiOX. The ferromagnetic characteristic results from the bonding formation between the Si sp hybrid orbitals and the Cr 3d orbitals at the SiOx/CrSi2 interface, where the oxygen atoms play an important role, bonding with silicon atoms and making chromium atoms with unpaired electrons, which contributes to ferromagnetism at nanoscale .
On photoluminescence analysis, Bhamu et al. studied the density of state (DOS) of CrSi2 bulk, including 1.33 eV, 0.56 eV above Fermi state, and 2.23 eV under Fermi state . Figure 8b shows our PL spectrum in the visible region for the CrSi2 nanowires, where the wide peak was present (red line) and through Gaussian fitting; the other two peaks, 396 nm (green line) and 465 nm (blue line), were calculated. Theoretically, the electron-hole pair recombinations of 1.33 eV, 0.56 eV conduct state to −2.23 eV valance state were 348 and 430 nm for CrSi2 bulk. In reality, the difference results from dimension, bulk, and nanowires; as the particle size reduces, wider bandgap light absorption band will move to shorter wavelengths, which is so-called blueshift ; however, there may be redshift as well; as the particle size decreases, the internal stress will increase, causing changes in the band structure  and the electron wave function overlap to increase the energy gap narrowing ; if the redshift factor is larger than the blueshift, then we will see redshift phenomenon, which is the case here.
In this study, using a CVD method, we have successfully synthesized chromium silicide nanowires of two phases with unique morphologies. Effects of some processing parameters, including the temperature, gas flow rate, and heating time, were investigated; for example, the growth of chromium disilicide nanowires were influenced by CrSi2 vapor supersaturation, CrSi2 vapor formation rate, and CVD control system. Also, the growth mechanism has been proposed. Field emission and photoluminescence measurements demonstrate that the CrSi2 nanowires are potential field-emitting and photovoltaic materials with a low turn-on field. Additionally, the magnetic property measurements for the CrSi2/SiOx nanowires, showing a ferromagnetic characteristic, demonstrate promising applications for magnetic storage and biological cell separation.
Competing interests The authors declare that they have no competing interests.
Authors’ contributions HFH and KCL conceived the study and designed the research. HFH conducted the experiments. HFH, PCT, and KCL wrote the manuscript. All authors read and approved the final manuscript.
KCL acknowledges the support from the National Science Council through grants 100-2628-E-006-025-MY2 and 102-2221-E-006-077-MY3.
Wu, WW; Lu, KC; Chen, KN; Yeh, PH; Wang, CE; Lin, YC: Controlled large strain of Ni silicide/Si/Ni silicide nanowire heterostructures and their electron transport properties, Appl Phys Lett, 2011, 97, S. 203110Crossref
Wu, WW; Lu, KC; Wang, CW; Hsieh, HY; Chen, SY; Chou, YC: Growth of multiple metal/semiconductor nanoheterostructures through point and line contact reactions, Nano Lett, 2010, 10, S. 3984–9CrossrefPubMed
Chou, YC; Lu, KC; Tu, KN: Nucleation and growth of epitaxial silicide in silicon nanowires, Mat Sci Eng R, 2010, 70, S. 112–25Crossref
Lu, KC; Wu, WW; Wu, HW; Tanner, CM; Chang, JP; Chen, LJ: In situ control of atomic-scale Si layer with huge strain in the nanoheterostructure NiSi/Si/NiSi through point contact reaction, Nano Lett, 2007, 8, S. 2389–94CrossrefPubMed
Lu, KC; Tu, KN; Wu, WW; Chen, LJ; Yoo, BY; Myung, NV: Point contact reactions between Ni and Si nanowires and reactive epitaxial growth of axial nano-NiSi/Si, Appl Phys Lett, 2007, 90, S. 253111Crossref
Liang, YH; Yu, SY; Hsin, CL; Huang, CW; Wu, WW: Growth of single-crystalline cobalt silicide nanowires with excellent physical properties, J Appl Phys, 2011, 110, S. 074302Crossref
Chen, LJ: An integral part of microelectronics, JOM, 2005, 57, S. 24–30Crossref
Derrien, J; Chevrier, J; Lethanh, V; Mahan, JE: Semiconducting silicide-silicon heterostructures: growth, properties and applications, Appl Surf Sci, 1992, 382, S. 56–8
Leong, D; Harry, M; Reeson, KJ; Homewood, KP: A silicon/iron-disilicide light-emitting diode operating at a wavelength of 1.5 mum, Nature, 1997, 387, S. 686–8Crossref
Bost, MC; Mahan, JE: An investigation of the optical constants and band gap of chromium disilicide, J Appl Phys, 1988, 63, S. 839–44Crossref
Shinoda, D; Asanabe, S; Sasaki, Y: Semiconducting properties of chromium disilicide, J Phys Soc Jpn, 1964, 19, S. 269–72Crossref
Bellani, V; Guizzetti, G; Marabelli, F; Piaggi, A; Borghesi, A; Nava, F: Theory and experiment on the optical properties of CrSi2 , Phys Rev B, 1992, 46, S. 9380–9Crossref
Mattheiss, LF: Electronic structure of CrSi2 and related refractory disilicides, Phys Rev B, 1991, 43, S. 12549–55Crossref
Anderson, WA; Delahoy, AE; Milano, RA: 8 percent efficient layered Schottky-barrier solar cell, J Appl Phys, 1974, 45, S. 3913–5Crossref
Bewlay, BP; Lipsitt, HA; Jackson, MR; Chang, KM: Processing microstructures and properties of Cr-Cr sub 3 Si, Nb-Nb sub 3 Si, and V-V sub 3 Si eutectics, Mater Manuf Processes, 1994, 9, S. 89–109Crossref
Nishida, I: The crystal growth and thermoelectric properties of chromium disilicide, J Mater Sci, 1972, 7, S. 1119–24Crossref
Rowe, DM: CRC handbook of thermoelectrics, 1995, Boca Raton, FL, CRC Press, S. 701
Chung, IJ; Hariz, A: Surface application of metal silicides for improved electrical properties of field-emitter arrays, Smart Mater Struct, 1997, 6, S. 633–9Crossref
Bonard, JM; Salvetat, JP; Stockli, T; Forro, L; Chatelain, A: Field emission from carbon nanotubes: perspectives for applications and clues to the emission mechanism, Appl Phys A, 1999, 69, S. 245–54Crossref
Minami, T; Miyata, T; Yamamoto, T: Transparent conducting zinc-co-doped ITO films prepared by magnetron sputtering, Surf Coat Technol, 1998, 108, S. 583–7Crossref
Hou, TC; Han, YH; Lo, SC; Lee, CT; Ouyang, H; Chen, LJ: Room-temperature ferromagnetism in CrSi2(core)/SiO2(shell) semiconducting nanocables, Appl Phys Lett, 2011, 98, S. 193104Crossref
Zhang, Y; Wu, Q; Qian, W; Liu, N; Qin, X; Yu, L: Morphology-controlled growth of chromium silicide nanostructures and their field emission properties, CrystEngComm, 2012, 14, S. 1659–64Crossref
Seo, K; Varadwaj, KSK; Cha, D; In, J; Kim, J; Park, J: Synthesis and electrical properties of single crystalline CrSi2 nanowires, J Phys Chem C, 2007, 111, S. 9072–6Crossref
Lee, CT; Li, TY; Chiou, SH; Lo, SC; Han, YH; Ouyang, H: First-principles analyses of unusual ferromagnetism observed in CrSi2(core)/SiO2(shell) nanocables, J Appl Phys, 2013, 113, S. 17E140
Bhamu, KC; Sahariya, J; Ahuja, BL: Electronic structure of ceramic CrSi2 and WSi2: Compton spectroscopy and ab-initio calculations, J Phys Chem Solids, 2013, 74, S. 765–71Crossref
Wang, Y; Herron, N: Nanometer-sized semiconductor clusters: materials synthesis, quantum size effects, and photophysical properties, J Phys Chem, 1991, 95, S. 525–32Crossref
Fu, H; Zunger, A: Electronic structure, surface effects, and the redshifted emission InP quantum dots, Phys Rev B, 1997, 56, S. 1496–508Crossref
Smith, CA; Lee, HWH; Leppert, VJ; Risbud, SH: Ultraviolet-blue emission and electron-hole states in ZnSe quantum dots, Appl Phys Lett, 1999, 75, S. 1688–90Crossref