Yang et al. 63

High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method

NanoWorld Journal

Research Article Open Access

http://dx.doi.org/10.17756/nwj.2016-032

Wen-zhi Yang1*, Wei-ming Huang1, Qiang Zheng2, Wei Huang1,3, Zi-ming Chen1, Fu-jun Shang1 and Bao-yu Zhang1

1Ningbo Branch of China Academy of Ordnance Science, Ningbo 315103, P. R. China2School of Materials Science and Engineering, Ningbo University of Technology, Ningbo 315016, P. R. China3College of Mechanical Engineering and Mechanics, Ningbo University, Ningbo 315103, P. R. China

*Correspondence to:Wen-zhi Yang, PhD Ningbo Branch of China Academy of Ordnance Science, Ningbo 315103, P. R. China Tel/Fax: +86 574 87902102E-mail: yangwenzhith@163.com

Received: October 21, 2016Accepted: December 10, 2016Published: December 12, 2016

Citation: Yang WZ, Huang WM, Zheng Q, Huang W, Chen ZM, et al. 2016. High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method. NanoWorld J 2(3): 63-68.

Copyright: © 2016 Yang et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 International License (CC-BY) (http://creativecommons.org/licenses/by/4.0/) which permits commercial use, including reproduction, adaptation, and distribution of the article provided the original author and source are credited.

Published by United Scientific Group

Abstract Free-standing silicon nano-crystals were synthesized from silicon multi-

crystals with high efficiency by induction plasma method. The particle morphology and distribution of nano powder were significantly influenced by powder feed rate, induction plasma power and the ratio of sheath gas. The average particle size monotonously increased with the increase of powder feed rate. The nano powder distribution became more and more concentrated as induction plasma power increased. The average size of nanopowder decreased obviously with the increase of H2 proportion. After the optimization of plasma parameters, mono-dispersed silicon nano-crystals were obtained with an average diameter of 20-85 nm and the mass yield reached a high level as 327 g/h. Meanwhile, the precursor utilization rate exceeded 81.8%. Under the TEM observation, all free-standing silicon nano-crystals had a uniform composition with a single crystal surrounded by a 1-2 nm thick amorphous silicon oxide shell. Although amorphous-like component was detected by Raman spectroscopy, the ensemble of silicon nano-crystals still showed a good crystallinity. The photoluminescence spectrum showed emission peaks in green region around 558 nm, which can be attributed to the oxide-related surface state of silicon nano-crystals.

KeywordsSilicon, Nano particles, Induction plasma, High efficiency synthesis

IntroductionSilicon nano-crystals play a very important role in nanotechnology and

this can be attributed to their outstanding properties and potential applications in several aspects [1]. As its most popular application, photovoltaic industry uses large amount of silicon-based nanostructures [2]. Due to their efficient photoluminescence (PL) properties in the visible range, silicon nano-crystals are also broadly applied in light emission field [3]. Besides these, silicon nano-crystals have been used to produce bulk nanostructured silicon samples for thermoelectric usage [4], and have been integrated into the anodes of lithium-ion batteries [5]. Furthermore, Free-standing silicon nano-crystals have great prospective in single particle transistor and a variety of other devices [6].

Lots of preparation methods have been proposed for the synthesis of free-standing silicon nano-crystals, such as thermal decomposition of silane [7], and the reduction of silicon halides [8]. More recently, nonthermal plasmas have been proposed for the synthesis of free-standing silicon nano-crystals

NanoWorld Journal | Volume 2 Issue 3, 2016 64

High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method Yang et al.

homogenous nucleation will lead to the formation of very fine particles. After the synthesis of silicon nano-crystals powder, an Ar-2vol% O2 gas mixture was flown for 24 hours to passivate the surface of nano powders. The cyclone acted to separate silicon nano-crystals from the excessive large particles that may contaminate the desired product. The primary collector contained 5 straight candles of porous metals filter (mesh of the filter is 10 mm), which further removed remained excessive large particles. Finally, silicon nano powders were collected in the glove box under Ar atmosphere.

Extensive material characterization was performed on the raw material powder and nano powder. An X-ray diffraction (XRD) using Cu Kα radiation (D/max-2500/PC, Rigaku Corporation, Japan) was conducted to analyze the phase composition. Particle morphology was observed by transmission electron microscope (Tecnai G2 20S-TWIN, FEI, Czech) with an accelerating voltage of 200 kV and scanning electron microscope (S4800, Hitachi, Japan). The average particle size of raw materials silicon multi-crystals and silicon nano-crystals were measured by BET test (ASAP 2020M, Micromeritics, USA). The size distribution of raw powder was evaluated by Microtrac particle size analyzer (Microtrac, S3500, USA). The surface information of silicon nano-crystals was obtained by Fourier-transform infrared spectroscopy (FTIR) (Nicolet 6700, Thermo, USA). Raman scattering (inVia-reflex, Renishaw, UK) was carried out to investigate the structure of nano powders. The photoluminescence spectrum was measured using high sensitive integrated fluorescence spectrometer (FluoroMax-4, Horiba, France) with 340 nm excitation.

Results and DiscussionFigure 2A shows the SEM micrographs of silicon multi-

crystals. The morphology of raw materials silicon multi-crystals is irregular. The distribution of particle size ranges between 1-78 mm, and the average particle size is 10.04 mm (Figure 2B). Since it is a complex physical and chemical process to synthesize the nano powder by induction plasma method using the silicon multi-crystals as a precursor, the plasma parameters, such as powder feed rate, induction plasma power and sheath gas proportion exert significant influence on

using silicon tetrachloride and silane as a precursor [9, 10]. One of the drawbacks of this technique is the fairly broad size distribution of nano-crystals. In a word, due to the lack of a simple, continuous, efficient, and inexpensive synthesis approach, the widespread utilization of free-standing silicon nano-crystals has been hindered. It is meaningful to explore an environment-friendly process, which can produce silicon particles with the desired structure and size. At the same time, high production rate and outstanding precursor utilization rate are also key factors need to be considered when preparing silicon nano-crystals.

For this reason, induction plasma method, which acquires the characteristics of high temperature, high energy density, and very short processing time, has attracted considerable attention, since its great potential for production of nano-sized powders. A wide range of nano powders have recently been prepared through induction thermal plasma process [11, 12]. However, there are few reports about the production of silicon nano-crystals by induction plasma using silicon multi-crystals as a precursor. In addition, synthesis of free-standing silicon nano-crystals from silicon multi-crystals is highly desirable for the reason that silicon multi-crystals are much cheaper and safer than other precursors, such as silicon tetrachloride and silane.

In the present article, the influence of plasma parameters on nano particle size distribution is revealed. And free-standing silicon nano-crystals are prepared with high efficiency via optimizing parameters using induction plasma method. The structure, morphology, crystal composition and photoluminescence properties of silicon nano-crystals are also investigated and discussed.

Material and MethodsSilicon nano-crystals were produced using radio frequency

(RF) helical resonator induction plasma (TIU-60, Tekna plasma systems Inc., Quebec, Canada). The typical 30-50 kW plasma was generated using a 5 MHz power supply. Because the in-flight vaporization of the powder and the absence of any electrode materials in contact with the plasma gases, induction plasma can offer a contamination-free approach particularly suitable for the preparation of silicon nano-crystals with high purity. Some pre-processing of silicon multi-crystals was necessary in order to obtain the uniform multi-crystals. Firstly, silicon multi-crystals were dried at 80 °C in vacuum for 24 h to remove the moisture. And then, the dried silicon multi-crystals were crushed by the Netzsch airflow mill to eliminate the large particle and aggregation. At last, silicon multi-crystals were graded by the air classifier to obtain the uniform multi-crystals. The schematic of plasma reactor was shown in Figure 1. Silicon multi-crystals as raw materials (> 99.99%, Lingyun silicon Inc., Xuzhou, China) were axially injected into the center of the Ar/H2 plasma with Ar carrier gas. At first, the silicon multi-crystals were evaporated in the plasma area, which is characterized by its ceramic plasma confinement tube construction surrounded by high velocity cooling water. The silicon vapor was subsequently subjected to a very rapid quenching by Ar/H2 quench gas in the vapor reactor, where

Figure 1: Schematic diagram of the induction plasma systems.

NanoWorld Journal | Volume 2 Issue 3, 2016 65

High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method Yang et al.

the particle morphology and properties of nano powder.

Powder feed rateThe powder feed rate plays an important role in the nano-

crystallization of silicon powder during the plasma process. Three powder feed rate are adopted, namely, 100, 400, and 800 g/h. Figure 3 shows the SEM micrographs of silicon nano-crystals at different powder feed rate. Apparently, the particle size distribution of nano powder is relative narrow at the powder feed rate of 100 g/h (Figure 3A). Most part of particles’ size lies between 150 to 200 nm. The particle size distribution becomes wider and wider as the powder feed rate increases. The distribution of powder is severe uneven when the powder feed rate is 800 g/h. Lots of small nano-particles adhered to the surface of the large particles, and some large particles are up to micrometer scale. Results of BET measurement are in accordance with the SEM micrographs of silicon nano-crystals. Table 1 exhibits the results of BET tests at different powder feed rate. The average particle size of silicon nano-crystals is 171 nm at the powder feed rate of 100 g/h, while the average particle size increases to 304 nm when the powder feed rate is 800 g/h. The average particle size monotonously goes up as powder feed rate increases. As the powder feed rate increases from 100 to 800 g/h, the quantity of raw powders passing through the plasma region increases sharply, which raise a greater demand of energy for nanocrystallization. However, the plasma power is fixed, which is too limited to let raw powders fully vapor at powder feed rate of 800 g/h. Owing to the poor heat absorption of vapor at high powder feed rate, the nanocrystallization efficiency of silicon multi-crystals drops significantly with the increase of powder feed rate.

Induction plasma powerThe energy of plasma region is closely connected with the

induction plasma power, which is a critical parameter to the formation of nano particles. The BET test results of silicon nano-crystals at different induction plasma power are shown in Table 2. With the increase of induction plasma power, the average particle size also increases to certain degrees. The average particle size of silicon nano-crystals is about 101 nm at 30 kW, and 171 nm at 50 kW. It is usually agreed that both the average particle size and the width and shape of the powder distributions are decisive indices of the product. Figure 4 exhibits the SEM micrographs of silicon nano-crystals at different induction plasma power. It is apparent that the nano powder distribution became more and more concentrated as the induction plasma powder increased. The distribution of nanopowder at 50 kW is much even than that of particles at lower power. Furthermore, the shape of nano powder at 50 kW is also better than that of powder at lower power. At 50 kW, the silicon nanoparticle shows the best sphericity and narrowest diameter distribution among the nanopowder.

Sheath gas proportionThe BET test results of silicon nano-crystals at different

sheath gas proportion are shown in Table 3. The results of BET test demonstrate that the average particle size of silicon nano-crystals decreases with the increase of H2 proportion in the sheath gas. With the increase of H2 from 5 to 20 standard liter per minute (slpm) in the sheath gas, the average particle size decreases from 132 to 67 nm. The conclusion is also proved by the SEM. The SEM micrographs of silicon nano-crystals at different sheath gas proportion are shown in Figure 5. It is clear that the size of nanopowder decreases as the proportion of H2 increases. Meanwhile, the nano powder distribution becomes more and more concentrated with the proportion of H2 increasing. The large particles (> 300 nm) are gradually eliminated as the proportion of H2 increases. The proportion of H2 plays an important role in plasma. As the proportion of H2 increases, the enthalpy and thermal conductivity of plasma increases obviously. And according to reality, the introduction of H2 effectively reduces the volume of plasma region, the reduction creates an enhanced energy density of plasma region. The enhancement of plasma energy density is beneficial to the formation of nano-crystals. Therefore, the nano powder

Table 1: The results of BET test at different powder feed rate.

Powder feed rate (g/h) The value of BET (m2/g)

The average particle size d (nm)

100 15.0827(151) 171(1)

400 11.7833(1257) 218(2)

800 8.4517(231) 304(1)

Table 2: The results of BET test at different induction plasma powder.

Induction plasma power (kW)

The value of BET (m2/g)

The average particle size d (nm)

30 25.4970(1577) 101(1)

40 19.5787(4294) 132(2)

50 15.0827(151) 171(1)

Figure 2: The SEM micrographs (A) and distribution of particle size (B) of silicon multi-crystals.

Figure 3: The SEM micrographs of silicon nano-crystals at different powder feed rate: (A) 100 g/h, (B) 400 g/h, and (C) 800 g/h.

Figure 4: The SEM micrographs of silicon nano-crystals at different induction plasma power: (A) 30 kW, (B) 40 kW, and (C) 50 kW.

NanoWorld Journal | Volume 2 Issue 3, 2016 66

High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method Yang et al.

distribution became more and more concentrated with the proportion of H2 increasing.

Optimal parameters of silicon nano-crystalsAnalyzing the impacts of each single plasma parameter

has on nano-crystals, we can conclude an optimized parameter setting on the base of a comprehensive discussion on multiplied factors. The key plasma parameters of silicon nano-crystals by induction plasma are shown in Table 4. It is known to us that the application of silicon nano-crystals is restricted by two critical factors. One is the production rate, and the other is convert ratio of precursor. The BET test results of silicon nano-crystals after optimizing parameters are shown in Table 5. The average particle size of silicon nano-crystals is 55 nm, and the mass production rate of nano powder has attained hectogram level per hour. The high production rate of 327 g/h reveals the high efficiency of silicon nano-crystals production, and the achieved 81.8% covert ratio lies a solid foundation for the massive application of silicon nano-crystals.

X-ray diffraction patterns of silicon multi-crystals (a) and silicon nano-crystals (b) are shown in Figure 6. The (111), (220), (311), (400), (331), and (422) diffraction peaks of diamond cubic silicon are clearly visible for both multi-crystals and nano-crystals. The ensemble crystalline fraction of the samples is very high [13]. However, compared with the raw materials, the silicon nano-crystals are spherically shaped with narrow size distribution. The average diameter of silicon nano-crystals is 55 nm by BET test (Table 5), which agrees with TEM results.

Transmission electron microscopy (TEM) is performed to verify the size and crystallinity of the material produced and to observe the shape and size distribution. TEM micrographs of silicon nano-crystals are shown in Figure 7. From bright field images (Figures 7A and 7B), it can be seen that the silicon nano-crystals exhibit spherical shape with diameter ranging from 20 to 85 nm. All silicon nano-crystals observed by TEM appear to be single crystals with no clear twin boundaries present. Figures 7C and 7D show high-resolution TEM (HRTEM) images of a typical example of the mono-dispersed silicon nano-crystals. Lattice fringes from the (111) planes confirm that the particles are crystalline. It is important to note that it is straightforward to identify lattice fringes throughout the sample by HRTEM, and this validly confirms the high degree of ensemble crystallinity of particles [13]. The HRTEM observations also reveal that each particle is composed of a single-crystalline silicon core covered by a 1-2 nm thick amorphous shell, which is ascribed to the oxidation of the silicon crystals (Figure 7D). Under the Ar-2vol% O2 atmosphere for 24 hours before collection, silicon nano-crystals are passivated and an amorphous silicon oxide thus formed, and this is proven by many former reports during the passivation process [13, 14]. The oxidation process can be described by the Elovich equation [10],

( )1 /ox o m md t r t ln t t= + ...............................................(1)

Table 3: The results of BET test at different sheath gas proportion.

sheath gas proportion (Ar/H2, slpm)

The value of BET (m2/g)

The average particle size d (nm)

30/5 19.5090(3188) 132(2)

30/10 25.4970(1577) 101(1)

30/20 38.2132(8644) 67(2)

Figure 5: The SEM micrographs of silicon nano-crystals at different sheath gas Ar/H2 proportion: (A) 30/5 slpm, (B) 30/10 slpm, and (C) 30/20 slpm.

Table 4: The optimal parameters of silicon nano-crystals by induction plasma.

Parameters Value

Central gas (Ar, slpm) 15

Sheath gas proportion (Ar/H2, slpm) 30/20

Quench gas (slpm) 100

Powder feed rate (g/h) 400

Induction plasma power (kW) 50

Reactor pressure (kPa) 80

Table 5: The results of BET test after optimizing parameters.

Item Value

BET (m2/g) 46.5086 (10914)

The average particle size d (nm) 55 (1)

Reaction time (h) 1

Raw materials in (g) 400

Nanocrystals out (g) 327

Convert ratio (%) 81.8

Figure 6: X-ray diffraction patterns of (A) silicon multi-crystals and (B) silicon nano-crystals.

NanoWorld Journal | Volume 2 Issue 3, 2016 67

High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method Yang et al.

typically observed in amorphous silicon [15]. Combing this test with XRD and TEM analysis, in spite of the presence of a 485 cm-1 shoulder, we can find that ensemble of the sample is dominated by crystalline fractions.

The photoluminescence spectrum of the silicon nano-crystals is presented in Figure 9. For this sample, the emission peaks in green region are around 558 nm. Lots of experiments have shown the evolution of the photoluminescence spectrum during size reduction in HF/HNO3 mixture. The green emission around 550 nm results from the size dependent band gap of silicon which originates from the quantum confinement effect [1, 16]. In this case, however, where silicon nano-crystals’ size range from 20 to 85 nm, the quantum confinement effect is not realistic. The green emission may be attributed to the oxide-related surface state of silicon nano-crystals [17].

ConclusionThe particle morphology and distribution of nano powder

were significantly influenced by powder feed rate, induction plasma power and the ratio of sheath gas. The average particle size monotonously increased with the increase of powder feed rate. The nano powder distribution became more and more concentrated as induction plasma power amount increased. The silicon nanoparticles synthesized at 50 kW show the best sphericity and narrowest diameter distribution among the series of nano powders. The average size of nanopowder decreases obviously as the proportion of H2 increases. And the nano powder distribution becomes more and more concentrated as the proportion of H2 goes up.

where the oxide thickness, dox as a function of time is determined by the monolayer growth rate, ro and characteristic growth time tm.

Further structural analyses of silicon multi-crystals and nano-crystals are also conducted. Fourier Transform infrared spectroscopy (FTIR) spectrum of silicon multi-crystals and nano-crystals are shown in Figure 8A. Both silicon multi-crystals and nano-crystals are exposed to air during the test procedure. All of the spectra are taken at room temperature, and normalized to the height of the strongest peak. Several bonds are evident in the spectrum which are most notably related to silicon bonds with oxygen, hydrogen, and hydroxy. Among them, the 1000-1250 cm-1 region is a characteristic of Si-O related modes [10, 13, 14]. Compared with that of multi-crystals, the intensity of Si-O stretching modes for nano-crystals is much higher, the difference means that silicon nano-crystals are much easier to oxidation than multi-crystals. After all, in consistence with TEM analysis, each silicon nano-crystals has an outer silicon oxide shell. Both silicon multi-crystals and nano-crystals exhibit a broad feature at 3400 cm-1, which can be attributed to OH [10]. In addition, there is also a broad absorption peak at 850-900 cm-1 in silicon nanocrytals, which can be attributed to Si-Hx modes or O-Si-Hx modes [10, 13].

Raman spectroscopy of silicon multi-crystals and nano-crystals are shown in Figure 8B. Each of the spectrum exhibits a two-peak Gaussian fit, which centered at 480 cm-1 and 512 cm-1, respectively. Peak position for the amorphous-like component locates between 480 cm-1and 495 cm-1, and the peak position for the crystalline component exists between 512 cm-1 and 520 cm-1 [10, 13, 15]. Owing to a breakdown of momentum conservation rule and the lack of long-range order in the crystal structure, broad peaks around 480 cm-1 are

Figure 7: TEM micrographs of silicon nano-crystals. (A & B) Bright field micrographs of nano-crystals produced using silicon multi-crystals. (C & D) High magnification TEM image of the silicon nano-crystals.

Figure 8: The (A) FTIR and (B) Raman spectrum of silicon multi-crystals and nano-crystals.

Figure 9: Photoluminescence spectra of the silicon nano-crystals.

NanoWorld Journal | Volume 2 Issue 3, 2016 68

High Efficiency Preparation, Structure and Properties of Silicon Nano-Crystals by Induction Plasma Method Yang et al.

After the optimization of process parameters, mono-dispersed silicon nano-crystals were obtained with an average diameter of 20-85 nm, and average size is about 55 nm. The mass yield reached a high level as 327 g/h. Meanwhile, the precursor utilization rate exceeded 81.8%. Under the TEM observation, all free-standing silicon nano-crystals had a uniform composition: with a single crystal surrounded by a 1-2 nm thick amorphous silicon oxide shell. When amorphous-like component was detected by Raman spectroscopy, the ensemble of silicon nano-crystals still showed a good crystallinity. The photoluminescence spectrum shows emission peaks in green region around 558 nm, which can be attributed to the oxide-related surface state of silicon nano-crystals.

AcknowledgementsThis work was sponsored by the Chinese Postdoctoral

Science Foundation (No. 2014M561795), Postdoctoral Scientific Research Project of Zhejiang Province (No. BSH1401037), Natural Science Foundation of Ningbo (No. 2015A610004), Start-up Grant from Ningbo University of Technology and the Technology Foundation for Selected Overseas Chinese Scholar, Ministry of Human Resources and Social Security of the People’s Republic of China.

References1. Mangolini L. 2013. Synthesis, properties and applications of silicon

nanocrystals. J Vac Sci Technol B 31: 020801. doi: 10.1116/1.4794789

2. Lazarenkova OL, Balandin AA. 2002. Electron and phonon energy spectra in a three-dimensional regimented quantum dot superlattice. Phys Rev B 66: 245319. doi: 10.1103/PhysRevB.66.245319

3. Cheng KY, Anthony R, Kortshagen UR, Holmes RJ. 2010. Hybrid silicon nanocrystal-organic light-emitting devices for infrared electroluminescence. Nano Lett 10(4): 1154-1157. doi: 10.1021/nl903212y

4. Petermann N, Stein N, Schierning G, Theissmann R, Stoib B, et al. 2011. Plasma synthesis of nanostructure for improved thermoelectric properties. J Phys D Appl Phys 44(17): 174034. doi: 10.1088/0022-3727/44/17/174034

5. Lee JK, Yoon WY, Kim BK. 2012. Electrochemical behavior of Si nanoparticle anode coated with diamond-like carbon for

lithium-ion battery. J Electrochem Soc 159(11): A1844-A1848. doi: 10.1149/2.045211jes

6. Ding Y, Dong Y, Bapat A, Nowak JD, Carter CB, et al. 2006. Single nanoparticle semiconductor devices. IEEE Trans Electron Devices 53(10): 2525-2531. doi: 10.1109/TED.2006.882047

7. Ostraat ML, De Blauwe JW, Green ML, Bell LD, Brongersma ML, et al. 2001. Synthesis and characterization of aerosol silicon nanocrystal nonvolatile floating-gate memory devices. Appl Phys Lett 79(3): 433-435. doi: 10.1063/1.1385190

8. Baldwin RK, Pettigrew KA, Garno JC, Power PP, Liu GY, et al. 2002. Room temperature solution synthesis of alkyl-capped tetrahedral shaped silicon nanocrystals. J Am Chem Soc 124(7): 1150-1151. doi: 10.1021/ja017170b

9. Mangolini L, Thimsen E, Kortshagen U. 2005. High-yield plasma synthesis of luminescent silicon nanocrystals. Nano Lett 5(4): 655-659. doi: 10.1021/nl050066y

10. Gresback R, Nozaki T, Okazaki K. 2011. Synthesis and oxidation of luminescent silicon nanocrystals from silicon tetrachloride by very high frequency nonthermal plasma. Nanotechnology 22(30): 305605. doi: 10.1088/0957-4484/22/30/305605

11. Guo JY, Fan XB, Dolbec R, Xue SW, Jurewicz J, et al. 2010. Development of nanopowder synthesis using induction plasma. Plasma Sci Technol 12(2): 188-199. doi: 10.1088/1009-0630/12/2/12

12. Li JG, Ikeda M, Ye R, Moriyoshi Y, Ishigaki T. 2007. Control of particle size and phase formation of TiO2 nanoparticles synthesized in RF induction plasma. J Phys D Appl Phys 40(8): 2348-2353. doi: 10.1088/0022-3727/40/8/S14

13. Yasar-Inceoglu O, Lopez T, Farshihagro E, Mangolini L. 2012. Silicon nanocrystal production through non-thermal plasma synthesis: a comparative study between silicon tetrachloride and silane precursors. Nanotechnology 23(25): 255604. doi: 10.1088/0957-4484/23/25/255604

14. Jurbergs D, Rogojina E, Mangolini L, Kortshagen U. 2006. Silicon nanocrystals with ensemble quantum yields exceeding 60%. Appl Phys Lett 88: 233116. doi: 10.1063/1.2210788

15. Hirasawa M, Orii T, Seto T. 2006. Size-dependent crystallization of Si nanoparticles. Appl Phys Lett 88: 093119. doi: 10.1063/1.2182018

16. Gupta A, Swihart MT, Wiggers H. 2009. Luminescent colloidal dispersion of silicon quantum dots from microwave plasma synthesis: exploring the photoluminescence behavior across the visible spectrum. Adv Funct Mater 19(5): 696-703. doi: 10.1002/adfm.200801548

17. Pavesi L, Dal Negro L, Mazzoleni C, Franzó G, Priolo F. 2000. Optical gain in silicon nanocrystals. Nature 408: 440-444. doi: 10.1038/35044012