Electroluminescent (EL) color shift of Si-rich SiOx with its O/Si composition detuned by changing the RF plasma powers during N2O/SiH4 vapor deposition are investigated. The higher O/Si composition ratio of Si-rich SiOx films when enlarging the RF plasma power contributes to an increment of the insulating SiO2 resistivity and the shrinkage silicon quantum dots (Si-QDs), indicating the fewer injected carriers in Si-QDs. The increasing oxygen content in Si-rich SiOx shortens the diffusion length of Si atoms to constrain the buried Si-QD size. In contrast to the blue-shift of EL peak wavelength induced by enlarging the RF plasma powers, the lengthening deposition time causes a thicker Si-rich SiOx film with more excessive Si atoms, thus providing larger Si-QDs for longer wavelength EL. The EL spectra of metal-oxide-semiconductor light-emitting diodes are red-shifted with increasing the Si-rich SiOx thickness due to the varied Si-QD size and degraded electron conductivity. The uniformity of Si-QDs in Si-rich SiOx layer contributes to the obvious wavelength shift when applying the biased current. The EL peak has a slightly blue-shifted phenomenon when the biased current increases under the band filling effect.
©2013 Optical Society of America
Anomalous-recipe deposition of Si-rich SiOx with the extremely high density of the embedded Si quantum dots (Si-QDs) has been extensively investigated as a new class of light emitting material over decades [1–9]. In spite of numerous experiments and theoretical calculations present in literature on the luminescent properties of Si nanostructures, only a few papers have been reported on the various electroluminescence (EL) characteristics of Si-QDs in Si-rich SiOx, mainly due to the difficulties of carrier injection and recombination within Si-QDs buried in a Si-rich SiOx matrix. Nevertheless, the possibility to fabricate an efficient Si-QD based light emitter represents a crucial step towards a practical application of Si nanostructure in optoelectronics. Versatile solutions have been recently developed to enhance the carrier injection efficiency, such as changing the work function of contact, shrinking the Si-QD bandgap, decreasing the Si-rich SiOx/matrix barrier height, and reducing the resistivity of host material, etc. . In addition, Si quantum dots (Si-QDs) embedded in the other host matrices, such as SiNx, SiC, ZnO, etc, were treated as another solution to decrease the energy barrier at the interface between the host matrix and Si-QD [11–16]. Cheng and associates have employed the Si-rich SixC1-x films with buried Si-QDs to for Si-based LEDs, which effectively decreases the turn-on voltage down to 4.2 V and enhances the turn-on current up to 0.42 mA . In particular, the EL wavelength of Si-rich SixC1-x based LEDs can be centered blue-shifted to 570 nm. Kurokawa et al. have applied Si-QD/a-SiC superlattice to obtain the tunable photoluminescence (PL) wavelength from 1170 nm to 780 nm . Rodriguez and associates reported that the PL of the Si-rich SiNx film with buried Si-QDs can be detuned from 550 nm to 375 nm . Park et al. further derived a theoretical formula of E(eV) = 1.56 + 2.40/d2 to acquire a relationship between PL wavelength and Si-QD size for Si-rich SiNx film with buried Si-QDs . Recently, Kuo’s group also observe the PL of Si-QDs embedded in ZnO host matrix at central wavelength of 685 nm . To obtain an efficient room-temperature EL, both the metal/Si-rich SiOx/Si and the metal/n-Si/Si-rich SiOx/p-Si based light emitting diodes (LEDs) were demonstrated [17,18], in which Fowler-Nordheim (F-N) and direct p-n junction barrier tunneling mechanisms were known to play important roles on the light emission from Si-QDs. However, the EL responses of these devices are usually not efficient due to the requirement of extremely high electric field for carriers tunneling through the insulating oxide channel [19,20]. In our previous work, making the Si-QD based metal-oxide-semiconductor light-emitting diodes (MOSLEDs) on Si nanopillar to obtain the external quantum efficiency up to 0.1% has been realized . Growing the Si-rich SiOx film embedded size-tunable Si-QD upon a high-aspect-ratio Si nanopillar array not only enlarges the luminescent power but also induces anomalous wavelength shift with increasing Si-rich SiOx thickness. In this work, the MOSLEDs are fabricated by plasma-enhanced chemical vapor deposition (PECVD) with different RF plasma powers and deposition time to control the O/Si composition and thickness of Si-rich SiOx layer. The effect of thickness and composition variations of Si-rich SiOx film on the shift of Si-QDs based EL wavelength and color are discussed.
2. Experiment setup
The Si-rich SiOx films were prepared on a p-type (100)- oriented Si substrate by using the PECVD system at the pressure and substrate temperature of 67 Pa and 450°C, respectively. The N2O and SiH4 fluence were controlled at 150 and 33 sccm, respectively. The deposition time was varied to detune the Si-rich SiOx thickness of 150 nm and 350 nm, while the RF plasma power was controlled at 60 and 70 W. Afterwards, the Si-rich SiOx films were annealed in a quartz furnace at 1100°C under N2 atmosphere for 10 min to synthesize the Si-QDs . For I-V and EL measurements, Al and indium tin oxide (ITO) were chosen for bottom and top electrodes with thickness of 200 and 500 nm as deposited by e-gun and thermal evaporation, respectively. The EL ranging from 300 to 900 nm were resolved by a monochromator (CVI, model DK240) with a UV-VIS-NIR photomultiplier (Hamamatsu, model R928) and digital multimeter (HP, model 34401A). The current-voltage (I-V) response of MOSLEDs with buried Si-QDs was measured by programmable electrometer (Keithley, model 237). The MOSLEDs were inserted into the silicon integral sphere head (ILX, model OMH-6703B) and the total power was detected by power multimeter (ILX, model OMM-6810B) .
3. Results and discussion
The XPS analysis clearly shows the corresponding O/Si composition ratio of 1.45 and 1.62, respectively, in the Si-rich SiOx grown at the RF plasma powers of 60 W and 70 W, indicating that the Si-rich SiOx matrix gradually transfers to pure SiO2 structure with increasing RF plasma powers, as shown in Fig. 1 . In more detail, the N2O molecules obtain more kinetic energy to decompose in PECVD chamber when the RF plasma power increases in the PECVD system, because the dissociation energy of N2O (112 kcal/mol) is much higher than that of SiH4 (75.6 kcal/mol), yielding more O to facilitate the synthesis of Si-rich SiOx. Thus, a higher O/Si composition ratio can easier be obtained with higher RF plasma powers during PECVD growth. Moreover, the variation of O/Si composition ratio in the Si-rich SiOx has a slight effect on the electrical property of MOSLEDs. The Si-rich SiOx film gradually transforms to pure SiO2 layer when enlarging the RF plasma power, contributing to an increment of insulating SiO2 resistivity. Indeed, the carrier is more difficultly injected into Si-QDs. Concurrently, the O/Si composition ratio of Si-rich SiOx film also contributes to the shrinkage of Si-QD size .
The HRTEM bright-field view image for the Si-QDs embedded in the 350-nm-thick Si-rich SiO1.45 and SiO1.62 samples and their corresponding size distribution are also shown as Fig. 2 . The Si-QD size for all Si-rich SiOx films grown under the different RF plasma powers has a relatively broadened distribution. The average size and standard deviation of Si-QDs embedded in the 350 nm-thick Si-rich SiO1.45 and SiO1.62 layers are 2.1 ± 0.2, and 1.7 ± 0.1 nm, respectively, in Fig. 2. The higher density of excess Si atoms for the Si-rich SiOx grown at the lower RF plasma powers contributes to the lower O/Si composition ratio. The Si atoms can easily move and self-aggregate because of various Si and inadequate O atoms in Si-rich SiO1.45 film. Moreover, the larger Si-QDs embedded in Si-rich SiOx film is synthesized if the lower-O/Si-ratio SiOx film conserves excessive Si atoms. With increasing plasma power, the TEM-estimated volume density of Si-QDs in the 350-nm-thick Si-rich SiO1.45 and SiO1.62 films is enlarged from 3.4 × 1018 to 5.9 × 1018 cm−3, respectively, in Fig. 2. Because the moving range of the excessive Si atoms is restricted to self-aggregate under the specific annealing temperature and the annealing time, the high-O/Si-ratio Si-rich SiO1.62 film easily contributes to the smaller Si-QD and the higher TEM-estimated volume density .
The comparison of Si-QD size distribution between the 150-nm- and 350-nm-thick Si-rich SiO1.62 is observed by using the HRTEM bright-field view image, as shown in Fig. 3 . The Si average size and standard deviation of the 150 nm-thick Si-rich SiO1.62 film is 1.6 ± 0.1 nm, as shown in Fig. 3. The larger average Si-QD size in thicker Si-rich SiOx layer could be easier formed than that in the thinner Si-rich SiOx layer because a diffusion length of Si atoms is not seriously confined. Moreover, the TEM-estimated volume density of Si-QDs in 150-nm-thick Si-rich SiO1.62 film is reduced to 2.1 × 1018 cm−3 in Fig. 3. Similar trend is also observed for the Si-rich SiO1.45 sample.
The upper part of Fig. 4 shows the I-V curves of Si-QD based MOSLEDs with the 150-nm- and 350-nm-thick Si-rich SiO1.45 films. The corresponding turn-on voltage enlarges from 85 V to 200 V as the turn-on electric field remains as a constant of 5.6 × 106 V/cm. The lower part of Fig. 4 reveals the I-V curves of Si-QD based MOSLEDs with the 150-nm- and 350-nm-thick Si-rich SiO1.62 film. The turn-on voltage is 99 and 230 V, but increasing the turn-on electric field to 6.6 MV/cm. In addition, the MOSLED device with thicker Si-rich SiO1.62 layer could suffer from higher current. When the carriers inject into the MOS structure, there are some thermal energy released to the Si-rich SiO1.62 film. Hence, the more Si-QDs within thicker Si-rich SiO1.62 sample facilitate more carrier transportation paths and higher thermal dissipation capability as compared the thinner ones. In principle, the smaller Si-QDs decrease the effective dielectric constant and enhance the barrier height of F-N tunneling to degrade overall tunneling probability [9,24], which makes carriers injected in Si-QDs under the stronger voltage. Therefore, the Si-rich SiO1.62 based MOSLEDs have the larger turn-on voltage and lower turn-on current. Similar observation on the decreased tunneling current contributed by shrinking Si-QD size has also been reported by Chakraborty et al. .
The maximum EL power of 10.6 and 81.6 nW with corresponding power density of 0.21 and 1.62 mW/cm2 and power slopes of 3 and 11.4 mW/A, respectively, are determined for the 150-nm- and 350-nm-thick Si-rich SiO1.45 based MOSLEDs, as shown in the upper part of Fig. 5(a) . In contrast, the maximum output powers of 150-nm and 350-nm thick Si-rich SiO1.62 based MOSLEDs are 5-5.5 times larger than those of the 150-nm- and 350-nm-thick thick SiO1.45 based MOSLEDs (55 and 469 nW), while the power slope is 53.82 and 115.2 mW/A, as shown in the lower part of Fig. 5(a). The one-order-of-magnitude increasing power slope of Si-QDs based MOSLEDs with thicker Si-rich SiO1.62 and more Si-QDs leads to a significant enhancement on the external quantum efficiency of these devices. This mainly results from an improved carrier recombination probability inside buried Si-QDs under a better confinement of injected carriers within Si-QDs. It is evident that the EL intensity increased with the current density flowing through the devices because the number of injected carriers in the Si-rich SiO1.62 matrix increases. The larger Si-QD density contributes to a more recombination probability to enhance the EL power emission power. In principle, the external quantum efficiency of MOSLEDs are compared by defining it as the ratio of the output photon numbers and input electron numbers, as described by
The EL saturation at high current density was clearly observed, and several factors may be important in leading to the reduction in efficiency at high current density in LEDs [26–28] analogous to those reported in other material systems [26–28]. The weakly confined QDs are formed from the Si-QDs embedded in Si-rich SiOx matrix, thus the small band offset potentially leads to carrier leakages at high current density [26,29]. However, further studies are still required to confirm the dominant factors leading to EL saturation in the LEDs.
It could be explained that the Si-QDs embedded in Si-rich SiOx layer is uniform and the amount of Si-QDs is proportionally increased when the thickness of Si-rich SiOx film increases. In comparison, the external quantum efficiency of 350-nm-thick Si-rich SiO1.45 based MOSLED is 4.7 × 10−5, which is one order of magnitude smaller than that of the Si-rich SiO1.62 based MOSLED. Attributing to the higher density of Si-QDs embedded in the Si-rich SiO1.62 film, the EL intensity and the external quantum efficiency of the Si-rich SiO1.62 film is observed to be higher than those of the Si-rich SiO1.45 film. The EL peaks are located at 460 and 500 nm for the MOSLED samples with 150- and 350-nm-thick Si-rich SiO1.45, as shown in Fig. 6 . In contrast, Fig. 6 reveals that the EL peaks are blue-shifted to 420 and 450 nm when fabricating the MOSLED by growing the 150- and 350-nm-thick Si-rich SiO1.62 films.
Basically, there are two main mechanisms known to contribute carrier transport through the Si-rich SiOx layer, including the direct tunneling and the Fowler-Nordheim (F-N) tunneling. For an MOSLED, the I-V analysis depends on the oxide thickness. If the oxide film is thin enough, the carriers could transport between the metal and the semiconductor to make the MOSLED behave like a Schottky-barrier diode. When the applied voltage on an MOSLED is smaller than the barrier height of the metal/oxide interface, the electrons can tunnel through the whole thin oxide layer to contribute direct tunneling current in this case. However, the direct tunneling terminates due to the thicker oxide film and the F-N tunneling process will start to dominate the carrier transport. The F-N tunneling occurs at the extremely high biased field when the metal/oxide barrier has a triangular shape for carriers tunneling through only a part of the oxide layer. The oxide film does not inhibit the current flow after tunneling through the triangular barrier. The current density could be increased with the typical behavior of the F-N tunneling by increasing the applied voltage. For all samples, the direct tunneling at the low voltage depends on the excessive Si concentration in the Si-rich SiOx film. In other words, the carrier transport is mainly due to the F-N tunneling mechanisms as described byTable 1 .
The effective barrier height slightly enhances when increasing the Si-rich SiOx thickness at the different RF plasma power. The barrier height plays an important role on the external quantum efficiency of Si-QD based MOSLEDs. The carriers with energy can easily escape from the Si-QDs due to a lower barrier height because the carrier transport is mainly dependent on the matrix in which the Si-QDs are embedded. The wave function for the electrons and holes in the Si-QD will overlap to increase a probability of the non-phonon-assisted recombination. The tunneling probability of the carriers between Si-QDs for the square potential well is given by
On the other hand, the EL peak wavelength of Si-QDs based MOSLED device with Si-rich SiOx grown at same RF plasma power would slightly red-shift with increasing thickness. This can be confirmed with the evidence on the EL peak wavelength of Si-QD based MOSLEDs with different Si-rich SiOx thickness shown in Fig. 7 . It is found that the EL of MOSLEDs with thicker Si-rich SiOx layer slightly red-shifts to a longer wavelength, where in the Si-rich SiOx sample with lower O/Si composition ratio induces a larger wavelength shift with increasing the thickness of Si-rich SiOx layer. Such a red-shifted phenomenon on the EL of devices with increasing Si-rich SiOx thickness could be explained by means of the relationship between the varied Si-QD size and degraded electron conductivity. Since it is easy to form bigger Si-QDs in thicker Si-rich SiOx film as the diffusion length of Si atoms is not seriously confined, thus the average Si-QD size could be relatively larger in thicker SiOx film than that in thinner SiOx film. Although the Si-QDs could also have a broadened size distribution in thicker Si-rich SiOx layer congenitally from PECVD fabrication, the EL can be triggered at a relatively lower resistance region, where the Si-QD embedded in Si-rich SiOx are closely congested to provide a carrier transportation path and to facilitate the radiative recombination within the buried Si-QDs. In the Si-QD based MOSLEDs, EL can only be observed under forward biased condition. Based on the F–N tunneling model, the recombination of electrons and holes can only tunnel into Si-QD from Si substrate (holes) and ITO (electron) when the device is highly biased, such that most of the Si-QDs near the oxide/substrate interface begins to be populated by holes. The increasing of applied current results in a serious hole population in Si-QD toward the central region of Si-rich SiOx film.
In addition, the average size of Si-QDs in the central region of Si-rich SiOx film is larger than those in the edge region, providing the EL spectra red-shifts with increasing biased current. Alternatively, if the average size of Si-QD in the central region of Si-rich SiOx film is smaller than that in the edge region, the EL spectra would blue-shift with increasing biased current. In general, the EL wavelength of Si-QD embedded MOSLED device with thicker Si-rich SiOx film is longer, which is originated from the aggregated Si-QDs with bigger size. In our case with the thickness and bias dependent wavelength shifting phenomenon obtained from Fig. 7, there is a less distinct trend for the EL peak wavelength shift with the varying biased current of MOSLED as compared to that changes with Si-rich SiOx thickness. These results elucidate that the distribution of Si-QD is relatively uniform in the Si-rich SiOx film although there existed a broadened size distribution of Si-QDs in different regions. The same result is also found by contrasting the EL of Si-rich SiO1.62 MOSLED. In more detail, the Fig. 7 shows the position of EL peak as a function of applied current and the EL peak slightly blue-shifts when the biased current increases, which is mainly due to the band filling effect occurred within the Si-QDs. It is evident that the EL intensity increases with the current density flowing through the device because the number of injected carriers in the Si-rich SiOx matrix increases. As the carrier injection effect becomes larger, most carriers would occupy the higher energy states of Si-QD and subsequently contributes to light emission of shorter wavelength. Obviously, the band filling effect is enhanced with thicker Si-rich SiOx film by comparing the upper two graphs and lower two graphs. Because the emitting optical power is more intense from thicker Si-rich SiOx film, there are more carriers participating in the radiative electron-hole combination. Since the carrier filling at energy states is a probability event dominated by the law of large-number carriers, hence, the band filling effect is statistically apparent as more electron-hole recombination centers (Si-QDs) embedded inside in thicker Si-rich SiOx film.
In conclusion, the O/Si composition variation and Si-rich SiOx thickness induced electroluminescent wavelength shift of Si-rich SiOx based blue and green MOSLEDs is observed and elucidated. The XPS analysis for SiOx samples reveals that the increasing O/Si composition ratio from 1.45 to 1.62 when enlarging the RF plasma power from 60 to 70 W. The higher RF plasma power facilitates the growth of Si-rich SiOx film with a larger O/Si composition ratio, thus providing a small distribution range for smaller Si-QDs. Moreover, for the thicker Si-rich SiOx layer, the Si-QD size slightly increases because of the more excess Si atoms in the Si-rich SiOx layer. By decreasing Si-QD size from 2.1 nm to 1.6 nm, the EL wavelength is blue-shifted from 500 nm to 420 nm. The turn-on voltage of MOSLEDs grown with increasing RF plasma powers enlarges from 85 V to 99 V because of the decreasing tunneling probability for smaller Si-QDs. It also contributes to the reducing turn-on current. Moreover, the MOSLEDs with the thicker Si-rich SiOx layer has a larger turn-on voltage due to an enlargement of insulating Si-rich SiOx films. The maximum EL emission power increases up to 469 nW at the 350-nm-thick SiO1.62 grown MOSLED due to a more recombination probability with the larger Si-QD volume density. The smaller Si-QD provides a deeper quantum well to leads to a better carrier confinement for carriers, which contributes to the higher external quantum efficiency. The EL wavelength of thicker Si-rich SiO1.45 MOSLED has an obviously red-shifted phenomenon due to a relationship between the varied Si-QD size and degraded current conductivity. The bigger Si-QDs easily provide carrier transportation paths to the radiative recombination because of its weak quantum confinement. In addition, the uniformity of Si-QDs in Si-rich SiOx layer contributes to an obvious wavelength shift when applying the biased current. In our case, the wavelength is only slightly shifted to the shorter wavelength when increasing the biased current due to the band filling effect. The band filling effect enhances with thicker Si-rich SiOx film because of the more carrier participation in the electron-hole recombination. The band filling effect is statistically apparent as more electron-hole recombination centers (Si-QDs) embedded inside in thicker Si-rich SiOx film although the carrier filling at energy states is a probability event dominated by the law of large-number carriers. Therefore, the Si-rich SiOx MOSLEDs could obtain the higher external quantum efficiency and EL emission power when further enlarging the RF plasma power and Si-rich SiOx thickness, which also contribute to the larger turn-on voltage and lower turn-on current.
This work was partially supported by the National Science Council, Taiwan, R.O.C. and the Excellent Research Projects of National Taiwan University, Taiwan, R.O.C., under grants NSC 100-2221-E-002-156-MY3, NSC 101-2622-E-002-009-CC2, NSC 101-ET-E-002-004-ET and 99R80301.
References and links
1. L. T. Canham, “Silicon quantum wire array fabrication by electrochemical and chemical dissolution of wafers,” Appl. Phys. Lett. 57(10), 1046–1048 (1990). [CrossRef]
3. G. G. Qin, A. P. Li, B. R. Zhang, and B.-C. Li, “Visible electroluminescence from semitransparent Au film/extra thin Si-rich silicon oxide film/p-Si structure,” J. Appl. Phys. 78(3), 2006–2009 (1995). [CrossRef]
4. H.-Z. Song, X.-M. Bao, N.-S. Li, and J.-Y. Zhang, “Relation between electroluminescence and photoluminescence of Si+-implanted SiO2,” J. Appl. Phys. 82(8), 4028–4032 (1997). [CrossRef]
5. C.-H. Lin, S.-C. Lee, and Y.-F. Chen, “Strong room-temperature photoluminescence of hydrogenated amorphous silicon oxide and its correlation to porous silicon,” Appl. Phys. Lett. 63(7), 902–904 (1993). [CrossRef]
7. G.-R. Lin, Y.-H. Pai, and C.-T. Lin, “Microwatt MOSLED using SiOx with buried Si nanocrystals on Si nano-pillar array,” J. Lightwave Technol. 26(11), 1486–1491 (2008). [CrossRef]
8. G.-R. Lin, C.-J. Lin, and C.-K. Lin, “Enhanced Fowler-Nordheim tunneling effect in nanocrystallite Si based LED with interfacial Si nano-pyramids,” Opt. Express 15(5), 2555–2563 (2007). [CrossRef] [PubMed]
9. B.-H. Lai, C.-H. Cheng, and G.-R. Lin, “Multicolor ITO/SiOx/p-Si/Al light emitting diodes with improved emission efficiency by small Si quantum dots,” IEEE J. Quantum Electron. 47(5), 698–704 (2011). [CrossRef]
10. G.-R. Lin, C.-J. Lin, and H.-C. Kuo, “Improving carrier transport and light emission in a silicon-nanocrystal based MOS light-emitting diode on silicon nanopillar array,” Appl. Phys. Lett. 91(9), 093122 (2007). [CrossRef]
11. C.-H. Cheng, C.-L. Wu, C.-C. Chen, L.-H. Tsai, Y.-H. Lin, and G.-R. Lin, “Si-rich SixC1-x light-emitting diodes with buried Si quantum dots,” IEEE Photonics J. 4(5), 1762–1775 (2012). [CrossRef]
12. A. Rodriguez, J. Arenas, and J. C. Alonso, “Photoluminescence mechanisms in silicon quantum dots embedded in nanometric chlorinated-silicon nitride films,” J. Lumin. 132(9), 2385–2389 (2012). [CrossRef]
13. N.-M. Park, C.-J. Choi, T.-Y. Seong, and S.-J. Park, “Quantum confinement in amorphous silicon quantum dots embedded in silicon nitride,” Phys. Rev. Lett. 86(7), 1355–1357 (2001). [CrossRef] [PubMed]
14. K.-Y. Kuo, S.-W. Hsu, P.-R. Huang, W.-L. Chuang, C.-C. Liu, and P.-T. Lee, “Optical properties and sub-bandgap formation of nano-crystalline Si quantum dots embedded ZnO thin film,” Opt. Express 20(10), 10470–10475 (2012). [CrossRef] [PubMed]
15. Y. Kurokawa, S. Tomita, S. Miyajima, A. Yamada, and M. Konagai, “Photoluminescence from silicon quantum dots in Si quantum dots/amorphous SiC superlattice,” J. Appl. Phys. 46(35), L833–L835 (2007). [CrossRef]
16. G.-R. Lin, Y.-H. Pai, C.-T. Lin, and C.-C. Chen, “Comparison on the electroluminescence of Si-rich SiNx and SiOx based light-emitting diodes,” Appl. Phys. Lett. 96(26), 263514 (2010). [CrossRef]
17. F. Iacona, G. Franzo, and C. Spinella, “Correlation between luminescence and structural properties of Si nanocrystals,” J. Appl. Phys. 87(3), 1295–1303 (2000). [CrossRef]
18. G. Franzò, A. Irrera, E. C. Moreira, M. Miritello, F. Iacona, D. Sanfilippo, G. Di Stefano, P. G. Fallica, and F. Priolo, “Electroluminescence of silicon nanocrystals in MOS structures,” Appl. Phys., A Mater. Sci. Process. 74(1), 1–5 (2002). [CrossRef]
19. C.-J. Lin and G.-R. Lin, “Defect-enhanced visible electroluminescence of multi-energy silicon-implanted silicon dioxide film,” IEEE J. Quantum Electron. 41(3), 441–447 (2005). [CrossRef]
20. G.-R. Lin, C.-J. Lin, C.-K. Lin, L.-J. Chou, and Y.-L. Chueh, “Oxygen defect and Si nanocrystal dependent white-light and near-infrared electroluminescence of Si-implanted and plasma-enhanced chemical-vapor deposition-grown Si-rich SiO2,” J. Appl. Phys. 97(9), 094306 (2005). [CrossRef]
21. B.-H. Lai, C.-H. Cheng, Y.-H. Pai, and G.-R. Lin, “Plasma power controlled deposition of SiOx with manipulated Si Quantum Dot size for photoluminescent wavelength tailoring,” Opt. Express 18(5), 4449–4456 (2010). [CrossRef] [PubMed]
22. G.-R. Lin, C.-J. Lin, and K.-C. Yu, “Time-resolved photoluminescence and capacitance-voltage analysis of the neutral vacancy defect in silicon implanted SiO2 on silicon substrate,” J. Appl. Phys. 96(5), 3025–3027 (2004). [CrossRef]
23. G.-R. Lin and C.-J. Lin, “Improved blue-green electroluminescence of metal-oxide-semiconductor diode fabricated on multirecipe Si-implanted and annealed SiO2/Si substrate,” J. Appl. Phys. 95(12), 8484–8486 (2004). [CrossRef]
24. G.-R. Lin, C.-H. Chang, C.-H. Cheng, C.-I. Wu, and P. S. Wang, “Transient UV and visible luminescent dynamics of Si-rich SiOx metal–oxide–semiconductor light-emitting diodes,” IEEE Photonics J. 4(5), 1351–1364 (2012). [CrossRef]
25. G. Chakraborty, S. Chattopadhyay, C. K. Sarkar, and C. Pramanik, “Tunneling current at the interface of silicon and silicon dioxide partly embedded with silicon nanocrystals in metal oxide semiconductor structures,” J. Appl. Phys. 101(2), 024315 (2007). [CrossRef]
26. N. Tansu and L. J. Mawst, “Current injection efficiency of InGaAsN quantum-well lasers,” J. Appl. Phys. 97(5), 054502 (2005). [CrossRef]
27. H. Zhao, G. Liu, R. A. Arif, and N. Tansu, “Current injection efficiency induced efficiency-droop in InGaN quantum well light-emitting diodes,” Solid-State Electron. 54(10), 1119–1124 (2010). [CrossRef]
28. I. E. Titkov, D. A. Sannikov, Y.-M. Park, and J.-K. Son, “Blue light emitting diode internal and injection efficiency,” AIP Adv. 2(3), 032117 (2012). [CrossRef]
29. N. Tansu and L. J. Mawst, “The role of hole leakage in 1300-nm InGaAsN quantum-well lasers,” Appl. Phys. Lett. 82(10), 1500–1502 (2003). [CrossRef]