In Situ-Doped Silicon Thin Films for Passivating Contacts by Hot-Wire

Subscriber access provided by EKU Libraries

Surfaces, Interfaces, and Applications

In-situ Doped Silicon Thin Films for Passivating Contacts by Hot-Wire Chemical Vapor Deposition with a High Deposition Rate of 42 nm/min Shenghao Li, Manuel Pomaska, Jan Hoß, Jan Lossen, Mirko Ziegner, Ruijiang Hong, Friedhelm Finger, Uwe Rau, and Kaining Ding ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b10360 • Publication Date (Web): 30 Jul 2019 Downloaded from pubs.acs.org on August 3, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

In-situ Doped Silicon Thin Films for Passivating Contacts by Hot-Wire Chemical Vapor Deposition with a High Deposition Rate of 42 nm/min Shenghao Li,†,‡ Manuel Pomaska,‡ Jan Hoß,§ Jan Lossen,§ Mirko Ziegner,⊥ Ruijiang Hong,*,† Friedhelm Finger,‡ Uwe Rau,‡ Kaining Ding,‡ †Institute

for Solar Energy Systems, Guangdong Provincial Key Laboratory of Photovoltaic

Technology, School of Physics, Sun Yat-sen University, 510006 Guangzhou, China ‡IEK-5

Photovoltaik, Forschungszentrum Jülich, 52425 Jülich, Germany

§International

⊥IEK-2

Solar Energy Research Center – ISC Konstanz, 78467 Konstanz, Germany

Microstructure and Properties of Materials, Forschungszentrum Jülich, 52425 Jülich,

Germany KEYWORDS: microstructure factor, domain size, silicon thin film, crystallization, selective contact, silicon surface passivation, phosphorus doping profile, Auger recombination,

ABSTRACT: Hot-wire chemical vapor deposition (HWCVD) was used to deposit in-situ doped amorphous silicon layers for poly-Si/SiOx passivating contacts at a high deposition rate of 42 1 ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

nm/min. We investigated the influence of a varied phosphine gas (PH3) concentration during the deposition on (i) the silicon film properties and (ii) the passivating contact performances. The microstructural film properties were characterized before and after a high temperature crystallization step to transform a-Si films into poly-Si films. Before the crystallization, the silicon layers become less dense as the PH3 concentrations increase. After the crystallization, an increasing domain size is derived for higher PH3 concentrations. Sheet resistance is found to decrease as domain size increased and the correlation between mobility and domain size was discussed. The performances of the passivating contact was measured and a firing-stable open circuit voltage of 732 mV, contact resistivity of 8.1 mΩ∙cm2 and a sheet resistance of 142 Ω/□ could be achieved with the optimized PH3 concentration. In addition, phosphorous doping tails into the crystalline silicon were extracted to evaluate the Auger recombination of the passivating contact.

1. Introduction Selective passivating contacts for crystalline silicon (c-Si) solar cells have attracted many research interests for their potential to improve solar cell efficiencies.1 Different materials such as SiC,2 MoOx3 and also organic materials4 are investigated for the application on selective passivating contacts. Among these materials, passivating contacts with polycrystalline silicon (poly-Si) on ultra-thin silicon oxide have shown high potential as a full area passivation technology for silicon solar cells application. By using this technology, power conversion efficiencies of up to 26.1% has been achieved on c-Si solar cells.5 In the fabrication of polySi/SiOx passivating contact, amorphous silicon (a-Si) layers are deposited and transformed into poly-Si by a high temperature annealing.6 For the deposition of a-Si layers, different chemical vapor deposition (CVD)7 and physical vapor deposition (PVD) techniques are proposed such as 2 ACS Paragon Plus Environment

Page 3 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


low pressure chemical vapor deposition (LPCVD),8-10 plasma enhanced chemical vapor deposition (PECVD),11-13 hot-wire chemical vapor deposition (HWCVD),14 and sputtering.15 Among these technologies, HWCVD has the potential for the highest deposition rate.16 Despite the high deposition rate, this technology enables good interfaces by avoiding ion damage caused by plasma. Excellent performances have been reported when a-Si films deposited by HWCVD were applied on silicon thin film solar cells and heterojunction solar cells.17-18 Recently, we fabricated poly-Si/SiOx carrier selective passivating contacts with both good passivation quality and low contact resistivity by performing POCl3 diffusions on intrinsic HWCVD a-Si layers.14 However, although HWCVD is a single-side deposition technology, the POCl3 diffusion would introduce phosphorus diffusion on both sides of the silicon wafer. To fabricate the targeted solar cells shown in Figure 1, additional steps to protect the front side of the solar cells would be necessary if POCl3 diffusion is performed. In order to simplify the fabrication process, in-situ doped silicon layers deposited by HWCVD for passivating contacts were developed in this work to avoid the POCl3 diffusion. In the case of LPCVD deposition, transforming from intrinsic silicon layers to in-situ doped silicon layers leads to a much lower deposition rate.19 The typical deposition rate reduces from around 10 nm/min for intrinsic layers to around 1 nm/min for in-situ doped layers.20 In the present work, however, a deposition rate of 42 nm/min was achieved for the in-situ doped layers by using HWCVD. After the a-Si layer deposition, thermal annealing in N2 was used for the crystallization so this process is well adapted to the solar cell process chain and no additional protection for front side is needed. In order to investigate the influence of phosphorus doping on the silicon thin films, we characterized the microstructure of in-situ doped silicon films before and after crystallization by Fourier-transform infrared (FTIR) spectroscopy and X-ray diffraction

3 ACS Paragon Plus Environment


Page 4 of 23

(XRD), respectively. For the determination of passivation quality, implied open-circuit voltage (iVOC) was measured by quasi-steady-state photoconductance (QSSPC). Contact resistivity (ρc) and sheet resistance (Rsheet) were measured by transmission line measurement (TLM) and fourpoint probe measurement, respectively. Secondary ion mass spectrometry (SIMS) was measured to determine the phosphorus concentration in the poly-Si and electrochemical capacitance voltage (ECV) was measured to reveal the effective doping profiles of the passivating contacts.

Figure 1. Targeted solar cell with full-area rear passivating contact. 2. Experimental The HWCVD a-Si:H layers were deposited at a working pressure of 0.01 mbar and a substrate heater temperature of 350 °C. The H2 flow rate was 10 sccm and the (SiH4+PH3) flow rate was also 10 sccm. In this work the PH3 concentration is defined as PH3/(SiH4+PH3), where the variables represent the respective gas flow rates. Three curled tantalum filaments were used in the deposition and the filament temperature was 1750 °C. The filament-to-substrate distance was 100 mm. For the measurement of iVOC as a measure for passivation quality, symmetric samples were fabricated. Phosphorus doped n-type c-Si wafers (1 Ω∙cm, as-cut thickness 180 μm) were sawdamage etched in NaOH and cleaned. After cleaning, SiOx layers of about 1.5 nm were fabricated on both sides of the wafers using the nitric acid oxidation of Si (NAOS) method21 for


Page 5 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


chemical passivation.22 The wafers were then covered by 100 nm in-situ doped a-Si:H layers on both sides using HWCVD followed by annealing in N2 at 825 °C for 30 minutes. After the annealing, 80 nm SiNx layers were deposited on top of the silicon layers at 450 °C to provide hydrogen for the interface passivation.23 The devices were completed by firing at a peak temperature of 820 °C to test the temperature stability of the passivation during sintering of the metal contacts. For the Rsheet measurement, a-Si:H precursor layers were deposited on top of ptype c-Si wafers after the same etching, cleaning and SiOx fabrication as for the symmetric lifetime samples described above. The precursor silicon layers were then transformed into polySi layers together with the lifetime samples in the same annealing process. XRD spectra were measured on a D8 ADVANCE diffractometer (Bruker AXS GmbH) under grazing incidence with incidence angle α = 1° in parallel beam geometry (setup: Cu long fine focus tube operated at 40 kV and 40 mA, parallel beam mirror, parallel plate collimator, SOL-X detector). In order to obtain good signal-to-noise ratios from the thin silicon films, each data point was measured for 120 s to reduce the measurement fluctuation. After measurement, the program TOPAS V6 (Bruker AXS GmbH) was used to analyze the line profile. 3. Results Our previous study indicates that the passivation quality of the passivating contact is related to the microstructure of the precursor a-Si:H layer.14 Intrinsic precursor layers with low microstructure factor give rise to high iVOC after POCl3 diffusion. The microstructure factor R is defined as:24 R

A2100 A2100  A2000

(1)



Page 6 of 23

where A2100 and A2000 are the integrals under 2100 cm-1 and 2000 cm-1 absorption peak mode in the FTIR spectra, respectively.25 The 2000 cm-1 absorption peak is commonly associated with isolated Si-H bonding embedded in rather small cavities in the a-Si bulk, while the 2100 cm-1 peak is associated with clustered Si-Hx groups (x=1 or 2) in larger cavities.26 Besides, the H content can also be extracted from the FTIR spectra.27 In this work FTIR spectroscopy was performed on the in-situ doped a-Si:H films deposited by HWCVD and the microstructure factor and hydrogen (H) content were extracted from the measured spectra. The results are compared with intrinsic a-Si:H film (PH3 concentration=0%) properties as shown in Figure 2. The microstructure factor increases from 0.26 to 0.37 as the PH3 concentration increased from 0% to 3.5%, which indicates a less dense material structure. On the other hand, the H contents of all the layers were around 10±1% for all PH3 concentrations. For all the samples the deposition rates were 42±3 nm/min and there was no obvious difference in deposition rate between intrinsic and different in-situ doped layers. This result is in contrast to LPCVD deposition, in which introducing doping gas reduces the deposition rate significantly.1920

Figure 2. a) FTIR spectra and b) microstructure factors and hydrogen content of precursor aSi:H layers deposited with different PH3 concentrations. The dash lines are linear fits for guiding the eye. 6 ACS Paragon Plus Environment

Page 7 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


After the N2 annealing, the crystallized in-situ doped silicon layers was analyzed by XRD measurement. The line profiles of the silicon (111), (220), (311), (331) and (422) reflections are shown in Figure 3a. The domain size is determined as integral breath based mean column length LVol,IB and crystallites are considered to consist of columns of cells along the scattering direction.28-30

LVol ,IB 

  IB  cos 

(2)

Here, λ is the wavelength of the radiation, βIB is the integral breadth (defined as the width of a rectangle with the same height and area as the line profile) and θ is the Bragg angle. As shown in Figure 3b, the derived domain size increases as the PH3 concentration increases from 0.5% to 3.0%. However, an even higher PH3 concentration of 3.5% leads to a reduced domain size.

Figure 3. a) XRD spectra and b) domain sizes of crystallized poly-Si layers deposited with different PH3 concentrations. Scanning electron microscope (SEM) was performed on the crystallized in-situ doped silicon layers to characterize the surface morphology. The results are shown in Figure 4. Rough surface morphologies are shown in all the poly-Si layers. However, no obvious change is observed on samples with different PH3 concentration. 7 ACS Paragon Plus Environment


Page 8 of 23

Figure 4. SEM characterization of crystallized poly-Si layers deposited with different PH3 concentrations. a) 0.5%, b) 1.5%, c) 2.0%, d) 2.5%, e) 3.0%, f) 3.5%. The performances of the fabricated passivating contact using different PH3 concentrations are shown in Figure 5, i.e. iVOC, ρc, and Rsheet. The iVOC after crystallization increased when the PH3 concentration increased from 0.5% to 3.0% but decreased as the PH3 concentration was further increased to 3.5%. The iVOC improved significantly after SiNx deposition and subsequent firing. The best iVOC of 731 mV was achieved at a PH3 concentration of 2.0%. All of the samples showed a ρc lower than 25 mΩ∙cm2. When the PH3 concentration increased from 0.5% to 2.0%, ρc decreased from 22.5 mΩ∙cm2 to 8.1 mΩ∙cm2, while further increasing the PH3 concentration yielded a constant ρc.


Page 9 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. a) iVOC, b) ρc and c) Rsheet of poly-Si/SiOx passivating stacks with different PH3 concentrations. Figure 5c shows a decreasing Rsheet from 5466 Ω/ □ to 106 Ω/□ for PH3 concentrations increasing from 0.5% to 3.0%, while at 3.5% PH3 concentration the Rsheet increased to 355 Ω/□ for annealing temperature of 825 °C. At 2.0% PH3 concentration, where the best iVOC was measured, Rsheet of 422 Ω/□ was measured. In order to reduce the Rsheet of the poly-Si/SiOx passivating stacks at 2.0% PH3 concentration, an annealing research with different annealing times and annealing temperatures was performed, as shown in Figure 6. Here the a-Si:H layer thickness was 200 nm in order to obtain a lower Rsheet. The results show a slightly increased iVOC and a slightly decreased Rsheet as the annealing temperature increases from 825 °C to 875 °C. Although the iVOC after crystallization increases as the annealing time increases from 5 min to 30 9 ACS Paragon Plus Environment


Page 10 of 23

min, the influence of annealing time on iVOC after SiNx and firing as well as Rsheet is ignorable. An exceptionally low Rsheet is measured when annealed at 850 °C and 30 min but the reason is unclear. With annealing at 875 °C and 30 min, firing-stable iVOC of 732 mV and Rsheet of 142 Ω/ □ was achieved.

Figure 6. a) iVOC and b) Rsheet of poly-Si/SiOx passivating stacks with different annealing times and annealing temperatures. In order to further discuss the passivation qualities at different PH3 concentrations, the phosphorus concentrations as well as the effective doping profiles were measured after crystallization using SIMS and ECV, respectively. Spectroscopic ellipsometry measurement showed that the layer thickness was 75 nm after crystallization. As shown in Figure 7a, the phosphorus concentration increased as PH3 concentration increased from 0.5% to 3.5%. On the other hand, the effective doping concentration in poly-Si layer increased as the PH3 concentration increased from 0.5% to 2.0%. However, further increasing the PH3 concentration did not give rise to higher effective doping concentration. Instead deeper doping tails in c-Si were measured as the PH3 concentration increased from 2.0% to 3.5% (Figure 7b).


Page 11 of 23 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 7. a) SIMS phosphorus profiles and b) ECV effective doping profiles of the passivating stacks. 4. Discussions 4.1 Silicon film properties A positive correlation between domain size and PH3 concentration from 0.5% to 3.0% are shown in Figure 3b. This result is in agreement with previous studies from Wada et al. who reported the same trend for increasing doping concentrations.31 The diffusion-enhanced growth model is used to explain the trend. In this model the grain growth would be enhanced by an increase in diffusion coefficient, while it has been reported that Si self-diffusion coefficient (Si in Si) increases as the phosphorus doping concentration increases.32 On the other hand, the microstructure of a-Si:H films could also influence the crystallization of the silicon films during annealing. K. Sharma et al. reported that an increase in microstructure factor leads to the development of larger grain sizes after annealing,33 which is also observed in this work. When the PH3 concentration is increased, the microstructure factor increased and also the domain size increased, as shown in Figures 2b and 3b. This trend is explained by the different nucleation rates of the silicon films. In high microstructure factor films there are more disordered domains around the surface of voids,34 resulting in lower nucleation rates.35 Also, it is 11 ACS Paragon Plus Environment


Page 12 of 23

reported that large grain poly-Si films can be obtained by suppressing the nucleation rate during annealing.33 Thus, in high PH3 concentration samples the domain sizes are larger since the nucleation rate was slower than that of low PH3 concentration samples during annealing. The crystallization of a-Si:H at relatively low temperatures (

In Situ-Doped Silicon Thin Films for Passivating Contacts by Hot-Wire

Recommend Documents