Properties of phosphorus-boron co-doped c-Si quantum dots/SiNx:H thin film prepared by PECVD in-situ deposition

doi:10.21203/rs.3.rs-4207464/v1

Download PDF

Article

Properties of phosphorus-boron co-doped c-Si quantum dots/SiNx:H thin film prepared by PECVD in-situ deposition

https://doi.org/10.21203/rs.3.rs-4207464/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 16 Sep, 2024

Read the published version in Scientific Reports →

You are reading this latest preprint version

Co-doping of phosphorus and boron elements into crystalline silicon quantum dot (c-Si QD) is an effective approach for enhancing the photoluminescence (PL) performance. In this paper, we report on the preparation of hydrogenated silicon nitride (SiN_x:H) thin films embedded with phosphorus-boron co-doped c-Si QDs via plasma enhanced chemical vapor deposition (PECVD) route. Mixed dilution including hydrogen (H₂) and argon (Ar) is applied in the in-situ deposition process for optimizing the deposition process. The P-B co-doped c-Si QD/SiN_x:H thin films exhibit a wide range of PL spectra. The emission is greatly improved especially for the short-wavelength light when compared to the SiO_x:H thin film containing P-B co-doped c-Si QDs. The effects of H₂/Ar flow ratio on the structural and optical characteristics of thin films are systematically investigated through a series of characterizations. Experimental results show that various properties, such as crystallinity, QD size, optical band gap and doping concentrations, are effectively controlled by tuning H₂/Ar flow ratio. Based on the red-shift of QCE-related PL peak, the successful P-B co-doping into Si QDs are verified. Finally, a comprehensive discussion has been made to analyze the influence of H₂-Ar mixed dilution on the film growth and impurity doping in detail in this paper.

Physical sciences/Materials science/Materials for optics

Physical sciences/Nanoscience and technology/Nanoscale materials/Quantum dots

Si quantum dot

silicon nitride

co-doping

PECVD

Crystalline silicon quantum dot (c-Si) has been proved to be a promising candidate for novel optoelectronics and photovoltaic devices by virtue of quantum confinement effect (QCE). Similar as other QD materials, the properties of c-Si QD are determined by QCE ^[1–3]. Accordingly, by simply controlling QD size, the emission/absorption spectra, radiative lifetimes, and the conductivity are able to be effectively modified for the intended application ^[4–7]. As benefit from the peculiarities, c-Si QD show great potential in manufacturing high-performance optoelectrical devices, in particular for photoluminescence (PL). With the shrink of size to nanoscale, c-Si QD should not be considered as an indirect bandgap semiconductor, quasi-direct radiative recombination takes place due to QCE and leading to an enhancement of emission ^[8]. Therefore, the PL intensity of c-Si QD is much stronger than that of bulk Si. When comparing with other QD materials (e.g. group II − VI QD, group III − V QD, perovskite QD), c-Si QD is competitive for commercial application owing to the nontoxicity and low cost for PL application.

As for c-Si QD, its bandgap energy is higher than that of bulk Si (~ 1.12 eV) because of QCE. It means a poor PL performance in near-infrared region. One approach for solving the problem is introducing impurities into Si QD, and PL energy can be effectively lowered. Nevertheless, single n-type or p-type doping leads to the PL quenching which results from the Auger interaction between photo-excited electron-hole pairs and carriers supplied by doping ^[9]. With the aim at eliminating Auger effect, researchers adopt co-doping method which enable P and B impurity atoms to be doped in c-Si QD simultaneously, and PL quenching can be inhibited through the charge compensation of n-type and p-type carriers ^[10–11]. Additionally, a further shift towards low-energy side for PL peak takes place as induced by the donor − acceptor (D − A) transitions deriving from P-B co-doping ^[12].

Nowadays the most commonly used method for producing P-B co-doped Si QD is co-sputtering. Through this way, P-B co-doped c-Si QDs can be obtained as embedded in silicon oxide (SiO_x) matrix. With a large bandgap energy over ~ 9.0 eV, SiO_x is considered as a suitable matrix material for enhancing QCE. However, the Si = O bonds at the interface between Si QDs and SiO_x results in the interface states which locate within band gap. The interface states weaken the emission of short-wavelength light, thus acting as an obstacle for the realization of Si-based full-color PL devices ^[13]. In order to extend the PL wavelength region, the influence of Si = O bonds should be eliminated. Researchers attempt to obtain P-B co-doped colloidal Si QDs by etching away SiO_x matrix ^[14] or directly synthesis through non-thermal plasma method ^[15]. Then the independent P-B co-doped colloidal Si QDs can be further assembled to thin film by spin-coating for device application ^[16].

Although the thin films based on P-B co-doped colloidal Si QDs show better substrate compatibility and emission ability in short wavelength range, current production techniques applied is relative complex and unsuitable for the high-efficiency film fabrication. In comparison, PECVD in-situ deposition is more efficient in preparing impurity-doped Si QD-based thin film under low temperature since there is just one step for fabricating the Si QD-based thin film with the structure of Si QDs/dielectric matrix ^[17–18]. Among various matrix materials, silicon nitride has prominent advantages in PL application by virtue of the relatively large band gap energy (~ 5.3 eV). Moreover, for Si QDs embedded in silicon nitride matrix, the PL emission in short wavelength range can be enhanced due to the absence of Si = O bonds. To our knowledge, there remains no reports concerning the investigations on the P-B co-doping of Si QD-based thin films with the structure of c-Si QD/SiN_x.

With the aim at realizing a wide wavelength region of PL spectrum, we have investigated the properties of P-B co-doped c-Si QD/SiN_x:H thin films. PECVD in-situ deposition technic has been carried out for preparing samples in this work. The mixed dilution including H₂ and Ar is adopted for promoting crystallization and doping since the effect of H₂-Ar mixed dilution has been verified previously ^[19]. Various characterizations have been implemented for studying the properties of P-B co-doped c-Si QD/SiN_x:H thin films, and the effect of dilution on crystallization and doping process has been illustrated in detail.

In this research, P-B co-doped Si QD/SiN_x:H thin films were deposited though PECVD route. A conventional capacitively coupled RF (13.56 MHZ) PECVD system has been employed with the electrode area of 144 cm² and the parallel electrode distance of 2 cm. Before experiments, p-type silicon (100) wafer and glass were clean by acetone, ethanol and deionized water in subsequence for the utilization as substrate. When the background vacuum was evacuated to 5×10^− 7 torr, the dilution gases H₂ and Ar, as well as the precursors silane (SiH₄) and ammonia (NH₃) were introduced into reaction chamber, then followed by the doping source gases phosphine (PH₃) and diborane (B₂H₆). The flow of SiH₄, NH₃, PH₃ and B₂H₆ were fixed at 5, 5, 0.25 and 0.125 sccm respectively, while the total flow of dilution was fixed at 500 sccm. During deposition, the reaction pressure was maintained at 0.5 torr for all the samples, while the substrate temperature and RF power were set to 150°C and 120 W respectively. In addition, the thin films for TEM observation were directly deposited on carbon-coated copper grids with the thickness of 30 ~ 60 nm.

For obtaining the crystallinity of P-B co-doped Si QD/SiN_x:H thin films, room-temperature Raman measurement was performed using an in Via Reflex Raman spectrometer with a laser at 514 nm. High-resolution transmission electron micrographs (HRTEM) images were obtained by means of a JEOL-JEM 2100 transmission electron microscope operating at 200 kV. A structural characterization of samples was carried out utilizing a fourier transform infrared spectrometer (Nicolet Is10) for studying FTIR spectra. An x-ray photoelectron spectroscopy (XPS) ESCALAB 250 was applied for investigating the chemical bonding configurations of samples as well as atomic concentrations of critical elements. A Horiba Jobin-Yvon T64000 system with a 325 nm He-Cd laser as the excitation source was applied for studying the photoluminescence (PL) properties of samples. The optical band gap of sample was estimated from the transmittance spectra obtained by a UV-1601 UV-visible spectrophotometer, with the thickness measured by a Dektak 6M step profiler.

The degree of crystallinity of samples are measured by the Raman spectrometer. The Raman spectra in the range of 200–1000 cm^− 1 of the P-B co-doped SiN_x:H thin films prepared at different H₂ flow are presented Fig. 1a. It can be seen that for all the samples, a significant broad band locating at ~ 480–550 cm^− 1 is identified. A sharp peak centered approximately at ~ 520 cm^− 1 is also observed. It is the Raman spectral signature of crystalline Si component, and can be found for the samples deposited at H₂ flow = 300, 400 and 500 sccm. At H₂ flow = 200 sccm, it can be concluded that there is no c-Si component in the thin film due to the absence of sharp peak. When the H₂ flow increases from 300 sccm to 500 sccm, the sharp peak emerges and gradually grows, indicating that the crystallization of P-B co-doped SiN_x:H thin film is promoted as more H₂ is introduced in dilution. This is in good agreement with previous reports ^[20].

Quantitative analysis of crystallinity degree relies on the crystalline volume fraction (F_c), which help us understand the structural evolvement of the deposited thin films. For calculating F_c, all the spectra of P-B co-doped SiN_x:H thin films in the wavenumber range of 384–552 cm^− 1 are fitted through Gaussian deconvolution method, and the typical best-fitting spectrum is shown in the Fig. 1b for the sample prepared at H₂ flow = 300 sccm. There are four Gaussian-shaped sub-peaks can be divided from the spectrum: (1) a broad peak assigned to the asymmetric Si-N bond stretching mode at ~ 465 cm^{− 1 [21]}; (2) a broad peak of the amorphous silicon (a-Si) phase at ~ 480 cm^− 1; (3) a narrow peak denoting the intermediate phase containing grain boundaries and Si ultra-nanocrystalline at ~ 510 cm^{− 1 [22–23]}; (4) a narrow Lorentz-shaped peak corresponding to the component of crystalline Si at ~ 520 cm^− 1. Considering the peaks of intermediate phase and c-Si co-exist in spectra, it can be deduced that the c-Si content in the thin films consists of Si QDs, and the amorphous content surrounding the Si QDs is composed of amorphous silicon nitride and amorphous silicon since Si-N bond and a-Si co-exist in the thin films. Considering the intermediate phase as a crystalline content, F_c can be estimated by the equation ^[24]:

F _c= (I_a + I_i)/(βI_a + I_i + I_c)

where I_a, I_i and I_c denote the integrated intensities of the amorphous, intermediate and crystalline component respectively. I_a is defined as the sum of integrated intensities of peaks corresponding to amorphous silicon nitride and a-Si. The cross-section ratio β depends on the size of the Si QDs and the excitation wavelength, and can be expressed as ^[24]:

β(D) = 0.1 + exp (-D/250)

where D is the QD size in nm, and can be assumed as a unity in the case of the tiny nanocrystallites of a few nanometers. The F_c value varying with H₂ is summarized in Table.1. At H₂ flow = 200 sccm, the thin film is completely amorphous. When H₂ flow achieves at 300 sccm, c-Si content forms with the F_c of 31.3%. Then as H₂ flow increases, the F_c gradually grows. The maximum F_c value of 47.1% is obtained at H₂ flow = 500 sccm.

In order to directly view the detailed internal microstructure of P-B co-doped SiN_x:H thin films, TEM micrographs are shown in Fig. 2. It can be seen in Fig. 2a that the thin film deposited at H₂ flow = 200 sccm has a cluster-like structure, where no Si QDs are founded. This phenomenon coincides well with the Raman spectra as the thin film is completely amorphous. Figure 2b shows the typical morphology of Si QDs which are formed in the P-B co-doped SiN_x:H thin film deposited at H₂ flow = 400 sccm, there can be observed numerous tiny spherical-shaped Si QDs with dense distribution, the QD size is estimated to be in the range of ~ 5–8 nm. The inset of Fig. 2b reveals the magnified details of microstructure by presenting the HRTEM image of Si QDs. The clearly visible crystallographic planes are observed for the Si QDs. The inter planer spacing of the well-defined crystallographic planes can be estimated to be ~ 3.0 Å, corresponding to the Si (111) lattice planes. Therefore, it is believable that the Si QDs are highly crystallized. The average QD size estimated from the TEM images has been presented in Table.1. As H₂ flow grows from 300 to 500 sccm, the average QD size increases from 4.76 to 8.58 nm. It is reasonable since the F_c value has the same varying tendency.

Table I. Morphological, compositional and optical properties of P-B co-doped SiN_x:H thin films

H₂ Flow (sccm)	F_c (%)	Average QD size (nm)	Optical band gap (eV)
200	N/A	N/A	2.78
300	31.3	4.76	2.24
400	34.4	6.17	1.85
500	47.1	8.58	1.68

FTIR spectroscopy is carried out to investigate the chemical bonding configurations of P-B co-doped SiN_x:H thin films. Figure 3 shows the FTIR absorbance spectra of the samples deposited with various H₂ flow. Different absorption bands corresponding to the typical vibration modes can be identified for all the samples, namely the Si–H wagging mode of a-Si at ~ 630 cm^− 1, the Si–N stretching mode at ~ 840–870 cm^− 1, the N–H rocking mode at ~ 1150 cm^− 1, the Si–H stretching mode at ~ 2100–2200 cm^− 1, and the N–H stretching mode at ~ 3300–3350 cm^{− 1 [25]}. As H₂ flow increases, the intensity of Si-N absorption band shrinks significantly, indicating the decreasing volume fraction of SiN_x:H. No P- or B-related absorption bands are found, it is because the P and B concentrations in the in-situ deposited thin films are rather low.

The elemental composition and bonding states of the deposited thin films are investigated by XPS measurement. Figure 4a shows the Si 2p core level XPS spectra for all the deposited thin films. While a typical best-fitted spectrum of the sample deposited at H₂ flow = 400 sccm is presented in Fig. 5a. As seen, five Gauss-shaped sub-peaks are extracted from the spectrum. According to previous works, the Si 2p peak in the range of ~ 99.8 eV to ~ 104 eV are determined by five different Si binding states Siⁿ⁺ (n = 0–4), and n represents the number of nitrogen atoms bonding with one silicon atoms ^[26]. Thereinto, the Si⁰ state at ~ 99.8 eV denotes the chemical structure of pure Si, meaning the presence of c-Si QDs as for our samples. When combining the Raman results with the Si 2p core level XPS spectra, it can be concluded that the SiN_x:H network is a mixture including four kinds of Si binding states. At H₂ flow = 200 sccm, a peak is observed centering near the position of Si⁴⁺ state. Since no Si⁰ peak can be found in the spectrum, the thin film is believed to be completely amorphous, which coincides well with the Raman results. As H₂ flow increases, the peak gradually shifts towards the low energy side, implying the increment in the proportion of c-Si component.

Figure 4b shows the XPS spectra of P 2p core level for the deposited thin films. At ~ 129.4 eV and ~ 133.3 eV, two peaks can be found for all the samples. The deconvolution of spectra is carried out, and the peak at ~ 129.4 eV can be separated to two sub-peaks. As shown in Fig. 5b, the spectrum of the sample deposited at H₂ flow = 400 sccm is presented. The two sub-peaks are respectively assigned to P-Si (~ 129.3 eV) and P-P (130.4 eV) chemical bonds, while the peak at ~ 133.3 eV denotes the P-N bonds in the matrix or at the QD interface ^[27]. With the increasing H₂ flow, the intensity of P-Si/P-P peak gradually enhanced while that of P-N peak shrinks, implying that the P-Si and P-P bonds tend to form in the case of large H₂ flow, whereas large Ar flow favors the formation of P-N bonds. According to the XPS spectra, it is possible that P-doping is promoted as more H₂ flow are introduced. It is notable that P-doping efficiency should not be directly related to P-Si peak, because successful P-doping depends on the P atoms penetrating into c-Si QDs and occupying Si sites, whereas the P-Si bonds formed at the QD interface or inside SiN_x:H matrix make no contribution to effective P-doping. The conclusion is the same for B-doping. Additionally, no peak corresponding to P-B bonds is visible in the spectra, meaning that there are seldom P-B compounds formed in the thin films.

Figure 4c shows the XPS spectra of B 1s core level for all the thin films. It can be observed that there is a broad peak at ~ 191 eV. This peak is attributed to the B-N chemical bonds, and it can be found for the thin film deposited at H₂ flow = 200 sccm. Since B-N bonds only form in the matrix or at the QD interface, the B-N peak correlates with the B atoms locating outside c-Si QDs. As H₂ flow increases, the B-N peak shrinks while the B-Si peak at ~ 187.eV emerges and grows. Similarly, the spectra of P 2p core level are the same in varying tendency, meaning that large H₂ flow is favorable for both P and B to be bonded to Si. Figure 5c shows the deconvolution of spectrum for the sample deposited at H₂ flow = 400 sccm. Besides B-N and B-Si peaks, there is another broad sub-peak at ~ 185 eV. According to previous studies, the peak is considered to be caused by the plasmon loss peak of Si ^[28]. Similar as P 2p spectra, no peak assigned to P-B bonds is found.

The atomic concentration ratios of compositional elements are shown in Fig. 6. All the atomic concentrations are calculated based on the XPS spectra. Figure 6a shows the atomic concentrations of Si and N varying with H₂ flow. As seen, the Si concentration increases significantly with the increasing H₂ flow whereas N concentration exhibits a opposite varying tendency. We calculate the N/Si concentration ratio, and find it decreases from 73.79–18.14%. Therefore, we conclude that N atoms incorporate into P-B co-doped SiN_x:H thin films more easily in the case of Ar-riched dilution. Correspondingly, a great amount of Si-N bonds perform as the barrier which interrupts the orderly growth of c-Si crystallographic planes, so the crystallization is suppressed. Figure 6b shows the atomic concentrations of P and B varying with H₂ flow. It is easily found that both the P and B concentrations gradually decrease with the increasing H₂ flow. As the H₂ flow increases from 200 sccm to 500 sccm, the calculated P/Si concentration ratio decreases from 2.56–0.49%, while the calculated B/Si concentration ratio decreases from 1.80–0.43%. Similar to N, it is more easily for P and B atoms to incorporate in thin films in the case of Ar-riched dilution. However, these P and B tend to bond with N other than Si according to the weak P-Si and B-Si peaks shown in Fig. 4. Correspondingly, there is no improvement in doping efficiency in spite of the larger impurity concentration.

The optical band gap energy (E_opt) is estimated by measuring the absorbance and reflectance using the Tauc equation which is defined as αhν = B(hν-E_opt)^{2 [29]}. α is the absorption coefficient, h is the Planck’s constant, ν is the frequency of the radiation, B is the edge width parameter. The E_opt value as a function of H₂ flow is shown in Table.1. With the H₂ flow increasing from 200 to 500 sccm, the E_opt value decreases from 2.78 eV to 1.68 eV. The E_opt value is determined by both the SiN_x:H matrix and Si QDs for our samples. So it is reasonable that E_opt grows wider under the combined effects of more SiN_x:H content and smaller c-Si QDs in the case of Ar-riched dilution.

Figure 7 represents the PL spectra of the P-B co-doped SiN_x:H thin films deposited with different H₂ flow. All the PL spectra are within a wide photo energy range of ~ 1.25–3.50 eV. Two peaks can be found respectively centering at ~ (P1) and (P2), it can be deduced from constant peak position that the two PL peaks are defect-related. According to previous studied, P1 at ~ 3.01 eV is attributed to the recombination from the conduction band to N⁴⁺ level, while P2 at ~ 2.41 eV originates from the radiative recombination at Si-dangling bond (K⁰) center ^[8]. Additionally, a P3 peak with varying center in the range of 1.45 eV ~ 2.11 eV is observed for the thin films deposited at H₂ flow = 300, 400 and 500 sccm. It is prominent that the PL mechanism of P3 is different from that of P1 and P2. Based on the varying tendency of QD size, P3 can be attributed to QCE since the red-shift takes place with the increasing QD size as the H₂ flow increases from 300 to 500 sccm. Note that the intensity of P3 gradually decreases with the increasing QD size. This probably results from the variation of P- and B-doping concentration in c-Si QDs and will be illustrated later.

The comparison of PL performance for the co-doped and un-doped SiN_x:H thin films deposited at H₂ flow = 400 sccm is presented in Fig. 8. The three peaks P1, P2 and P3 can be observed for both spectra in which all the peaks are larger in intensity for the un-doped sample. For the co-doped sample, a red-shift takes place for the QCE-related P3 peak, suggesting the impurity levels are involved in the optical transitions. When P and B are simultaneously doped in Si QDs, the transition between donor and acceptor levels which locates within band gap leads to lower PL energy. Therefore, the red-shift is a sign of the realization of P-B co-doping, so it is verified that the P-B co-doped c-Si QD/SiN_x:H thin films have been successfully prepared by PECVD in-situ deposition method.

According to the measurement results presented above, P-B co-doped c-Si QD/SiN_x:H thin films have been successfully prepared through PECVD in-situ deposition method. The production consists of only one step, and the process temperature is limited as low as 150 ℃. As for the co-sputtering method, the formation of c-Si QD depends on the high process temperature (generally > 1000 ºC) which is beyond the limit of most substrate materials. On the other hand, impurity atoms can be driven towards QD interface and finally move out of c-Si QD in the case of high temperature. This phenomenon is known as self-purification effect ^[30], and it plays a negative role in impurity doping. Unlike co-sputtering method, plasma plays the key role other than temperature for the in-situ deposition. In this process, the Si-based and N-based precursors are dissociated in plasma to be ionized Si-based and N-based radicals, these ionized radicals gradually agglomerate to fabricate amorphous network in which c-Si QDs are self-assembled at nucleation centers. Meanwhile, the ionized B-based and P-based radicals captured by the defects in the growing Si QDs are rapidly overlaid by the Si-based radicals which reach subsequently. As a result, these B and P atoms are trapped in Si QDs. Since the whole process lasts only a few milliseconds ^[15], there is no chance for impurity atoms to escape, instead they permanently locate inside QDs and form displacement doping. Evidently, self-purification effect become rather weak in this plasma-dominated route which is a kinetic growth process. Impurity doping becomes more efficient due to the little heat created during deposition.

The growth mechanism of Si QD-based thin film prepared by the way of PECVD in-situ deposition has been intensively studied, and the effect of dilution gas on film growth have been clearly clarified ^[31–33]. As for this research, H₂ and Ar are applied for creating plasma. It is well known that H plasma is able to induce crystallization for Si-based film, while the Ar plasma has the same function. What’s more, Ar are much lower in threshold energy comparing with H₂ since there are two kinds of metastable states Ar* ^[34]. It means that Ar plasma possesses higher energy when excited at the same power, so it is more efficient in creating nucleation centers by breaking weak Si-Si bonds. Furthermore, dissociation takes place in Ar plasma via direct interaction or the electron-accelerating route ^[35]. Consequently, dissociation rates of various kinds of molecules can be significantly improved when Ar flow is increased, enabling the incorporation of more atoms into network. As for our samples, the N/Si concentration ratio is only 18.14% at H₂ flow = 500 sccm. Considering the same flow of SiH₄ and NH₃, obviously the dissociation of NH₃ is insufficient in the case of H₂-riched dilution. As more Ar flow are introduced, the N/Si concentration ratio grow larger, indicating that the dissociation rate of NH₃ greatly improves, and it becomes close to that of SiH₄ in Ar-riched dilution. As verified by previous research, larger N/Si concentration ratio leads to smaller Si QDs and lower F_c ^[13]. This is because that more incorporated N leads to larger volume of SiN_x:H which plays a role as the barrier against the growth of c-Si QD by inhibiting the fabrication of crystallographic plane. Our results coincide well with the previous researches, and the N/Si concentration ratio can be easily tuned by varying the flow ratio of H₂/Ar.

Similar as the N/Si concentration ratio, both the P/Si and B/Si concentration ratios increase when larger Ar flow is applied, as revealed in Fig. 6b. It is reasonable as the PH₃ and B₂H₆ molecules are dissociated more efficiently in Ar-riched dilution. More P and B atoms are able to diffuse into the thin film and form n- or p-type doping. However, the impurity atoms outside c-Si QDs make no contribution to effective doping. Therefore, B/Si or P/Si concentration ratio should not be used for accurately measuring doping level. In consideration of this point, we have to identify the P-B co-doping effect indirectly through analyzing the PL performance. According to Fig. 8, the realization of P-B co-doping in Si QDs can be verified according to the red-shift of QCE-related P3 peak. Notably, the defect-related P1 and P2 peaks shrink for the P-B co-doped sample, it is possible that the impurity atoms passivate the defects, and correspondingly inhibit the defect-related radiative recombination. Moreover, P3 peak exhibits a significant shrinkage with the decreasing H₂ flow as shown in Fig. 7. It implies that when simultaneously doped by P and B, larger Si QDs possess lower PL intensity. Hori et al has demonstrated that the large co-doped Si QDs are p-type semiconductor since donors and acceptors are not well compensated. As a result, the Fermi level is very close to the HOMO level ^[35]. As for our samples, it can be deduced that the Auger recombination in the uncompensated larger Si QDs induces the PL quenching, and this effect will enhance with the increasing QD size.

Since our aim is synthesizing a type of film material which possesses excellent PL properties while is suitable for efficient production. The P-B co-doped c-Si QDs/SiN_x:H thin films exhibit a wide range of PL spectra from ~ 1.25 eV to ~ 3.50 eV. In contrast, the PL energy of P-B co-doped c-Si QDs embedded in SiO_x can hardly surpass 2.1 eV because of the Si = O bonds ^[36]. It suggests the P-B co-doped c-Si QDs/SiN_x:H thin films are more appropriate for the application on light-emitting devices by virtue of the wide range of PL spectra. What’s more, the production of P-B co-doped c-Si QDs/SiN_x:H thin films is simply as there is only one step, superior to the multi-step process such as the co-sputtering technic, as well as the methods for obtaining co-doped colloidal Si QDs. Based on all the illustration above, the P-B co-doped Si QDs/SiN_x:H thin films should be considered as a promising candidate for novel high-performance PL devices.

In conclusion, we have investigated the fabrication process as well as the properties of P-B co-doped Si QDs/SiN_x:H thin films prepared by PECVD in-situ deposition method. H₂-Ar mixed dilution has been introduced during deposition. Through various measurements, we have demonstrated the successful co-doping of P and B into the Si QDs embedded in SiN_x:H matrix by applying H₂-Ar mixed dilution. It is also found that by tuning H₂/Ar flow ratio, various properties, such as QD size, optical band gap and PL peak, can be effectively controlled. As for the QD-related PL peak, PL quenching has been observed together with a red-shift of peak position as QD size increases. This has been attributed to the imperfect compensation of P and B atoms in large Si QDs. In contrast with the undoped sample, the shrinkage in the intensity of QD-related PL peak has proved the successful co-doping. All the spectra are in a wide PL energy range from infrared region to ultraviolet region. In summary, P-B co-doped Si QD/SiN_x:H thin film can be fabricated by PECVD in-situ deposition method at relatively low substrate temperature. This work will be enlightening for the design of novel QD-based luminescence device.

Author Contribution

Zhifeng Gu wrote the main manuscript text. Feng Shan and Jia Liu reviewed the manuscript.

Data Availability

All data that support the findings of this study are included within the article

Park, N. M., Choi, C. J., Seong, T. Y., & Park, S. J. Phys. Rev. Lett. 86, 7 (2001).
Terada, S., Ueda, H., Ono, T., & Saitow, K. ACS Sustainable Chem. Eng. 10, 1765−1776 (2022).
D. Li, J. Chen, T. Sun, Y. Zhang, J. Xu, W. Li and K. Chen, Opt. Express. 30, 12308 (2022).
Q. Masaadeh, E. Kaplani and Y. Chao, Electronics 11, 2433 (2022).
X. Yu, Z. Ma, Z. Shen, W. Li, K. Chen, J. Xu and L. Xu, Nanomaterials 12, 2459 (2022).
R. Tsubata, K. Gotoh, M. Matsumi, M. Wilde, T. Inoue, Y. Kurokawa, K. Fukutani, N. Usami, ACS Appl. Nano Mater. 5, 1820−1827 (2022).
Q. Xu, I. T. Cheong, L. Meng, J. G. C. Veinot and X. Wang, ACS. Nano. 15, 18429−18436 (2021).
B. Sain and D. Das, Phys. Chem. Chem. Phys. 15, 3881 (2013).
B. S. Joo, S. Jang, M. Gu, N. Jung and M. Han, J. Alloy. Compd. 801, 568–572 (2019).
M. Fujii, K. Toshikiyo, Y. Takase, Y. Yamaguchi and S. Hayashi, J. Appl. Phys. 94, 1990 (2003).
M. Fujii, Y. Yamaguchi, Y. Takase, K. Ninomiya and S. Hayashi, Appl. Phys. Lett. 85, 1158 (2004).
R. Limpens, M. Fujii, N. R. Neale and T. Gregorkiewicz, J. Phys. Chem. C. 122, 6397–6404 (2018).
T. W. Kim, C. H. Cho, B. H. Kim and S. J. Park, Appl. Phys. Lett. 88, 123102 (2006).
M. Fukuda, M. Fujii, H. Sugimoto, K. Imakita and S. Hayashi, Opt. Lett. 36, 4026–4028 (2011).
R. Limpens, G. F. Pach and N. R. Neale, Chem. Mater. 31, 4426–4435 (2019).
H. Sugimoto, M. Fujii and K. Imakita, Nanoscale 6, 12354 (2014).
D. Das and S. Samanta, Physica E 128, 114615 (2021).
D. Das and S. Samanta, Mater Chem Phys 243, 122628 (2020).
X. Zhang, A. Wu, S. Shi and F. Qin, Surf. Coat. Technol. 228, S412–S415 (2013).
C. Z. Chen, S. H. Qiu, C. Q. Liu, Y. D. Wu, P. Li, C. Y. Yu and X. Y. Lin, J. Phys. D. Appl. Phys. 41,195413 (2008).
L. V. Mercaldo, E. M. Esposito, P. D. Veneri, G. Fameli, S. Mirabella and G. Nicotra, Appl. Phys. Lett. 97, 153112 (2010).
P. Zhang, S. Li, D. Li, L. Ren, Z. Qin, L. Jiang and J. Xu, Opt. Laser. Technol. 157, 108706 (2023).
Q. Cheng, S. Xu and K. Ostrikov, J. Mater. Chem. 19, 5134–5140 (2009).
D. Raha and D. Das, Sol. Energy Mater. Sol. Cells 95, 3181–3188 (2011).
B. H. Kim, C. H. Cho, T. W. Kim, N. M. Park and G. Y. Sung, Appl. Phys. Lett. 86, 091908 (2005).
B. Li, T. Fujimoto, N. Fukumoto, K. Honda and I. Kojima, Thin Solid Films 334, 140–144 (1998).
P. J. Wu, Y. C. Wang and I. C. Chen, J. Phys. D: Appl. Phys. 46, 125104 (2013).
C. Song, J. Xu, G. Chen, H.C. Sun, Y. Liu, W. Li, L. Xu, Z.Y. Ma and K.J. Chen, Appl. Surf. Sci. 257, 1337 (2010).
S. Park, E. Cho, D. Song, G. Conibeer and M. A. Green, Sol. Energy Mater. Sol. Cells. 93, 684–690 (2009).
G. M. Dalpian and J. R. Chelikowsky, Phys. Rev. Lett. 96, 226802 (2006).
F. Chaibi, R. Jemai, H. Aguas, H. Khemakhem and K. Khirouni, J. Mater. Sci. 53, 3672–3681 (2018).
D. Das, D. Raha and K. Bhattacharya, J. Nanosci. Nanotechnol. 9, 5614–5621 (2009).
S. Veprek, Z. Iqbal, J. Brunner and M. Schärli, Phil. Mag. B. 43, 527–547(1981).
A. Kole and P. Chaudhuri, Thin Solid Films 522, 45–49 (2012).
Y. Hori, S. Kano, H. Sugimoto, K. Imakita and M. Fujii, Nano Lett. 16, 2615–2620 (2016).
M. V. Wolkin, J. Jorne and P. M. Fauchet, Phys. Rev. Lett. 82, 197 (1999).

No competing interests reported.

Download PDF

Journal Publication

published 16 Sep, 2024

Read the published version in Scientific Reports →

Editor invited by journal
09 Apr, 2024
Submission checks completed at journal
09 Apr, 2024
First submitted to journal
02 Apr, 2024

You are reading this latest preprint version

Properties of phosphorus-boron co-doped c-Si quantum dots/SiNx:H thin film prepared by PECVD in-situ deposition

Status:

Journal Publication

Version 1

Abstract

Figures

I. Introduction

II. Experimental

III. Results

IV. Discussion

V. Conclusion

Declarations

Author Contribution

Data Availability

References

Additional Declarations

Status:

Journal Publication

Version 1