Search

This paper reports on a systematic and quantitative assessment of light induced degradation (LID) and regeneration in full Al-BSF and passivated emitter rear contact cells (PERC) along with the fundamental understanding of the difference between the two. After LID, PERC cells showed a much greater loss in cell efficiency than full Al-BSF cells (∼0.9% vs ∼0.6%) because the degradation in bulk lifetime also erodes the benefit of superior BSRV in PERC cells. Three main regeneration conditions involving the combination of heat and light (75 °C/1 Sun/48 h, 130 °C/2 Suns/1.5 h and 200 °C/3 Suns/30 s) were implemented to eliminate LID loss due to BO defects. Low temperature/long time (75 °C/48 h) and high temperature/short time (200 °C/30s) regeneration process was unable to reach 100% stabilization. The intermediate temperature/time (130 °C/1.5 h) generation achieved nearly full recovery and stabilization (over 99%) for both full Al-BSF and PERC cells. We discussed the effect of temperature, time and suns in regeneration mechanism for two cells.

Most p-type Si solar cells involves phosphorus-doped emitter by POCl 3 diffusion or phosphorus ion-implantation. Although the formation of the phosphorus emitter is known to getter impurities like Fe, the difference in the impact of these two gettering techniques on cell performance is not well quantified. Therefore, this paper compares the gettering efficiency of POCl 3 diffusion and phosphorus ion-implantation on Czochralski(Cz) and cast quasi-mono Si wafers. Cz-Si wafers were used to measure bulk lifetime and iron concentration before and after POCl 3 diffusion and phosphorus implantation with different doses. Increase in phosphorus implantation dose improved the gettering efficiency by increasing bulk lifetime and decreasing iron concentration but remained inferior to POCl 3 diffusion partly due to single side gettering as opposed to double side gettering with higher phosphorus surface concentration during POCl 3 diffusion. Moreover, large-area solar cells were fabricated on cast quasi-mono and Cz-Si wafers to quantify the impact of these emitters on cell parameters. POCl 3 diffused cast quasi-mono cells showed ~0.4% higher efficiency due to higher bulk lifetime compared to phosphorus implanted emitter. However, phosphorus implanted Cz cells gave ~0.3% higher efficiency due to lower emitter saturation current, resulting from the benefit of in-situ oxide surface passivation during the implant anneal.

Solar cells fabricated on boron (B)-doped Czochralski (Cz) Si wafers in the photovoltaic industry are known to suffer from light-induced degradation (LID) in efficiency. This paper reports on promising LID-free large-area indium (In)-doped Cz Si solar cells. Two different commercial-grade B-doped Cz materials were included for comparison. To study the impact of LID on the cell structure, ion-implanted large-area (239 and 242.22 cm2) screen-printed full aluminum (Al) back-surface field (BSF) baseline cells, as well as higher performance passivated emitter rear cells (PERC) with oxide passivation and local Al BSF, were fabricated. In-doped PERC cells achieved 20.3% efficiency, while the B-doped cells gave efficiencies of 20.7% and 20.5% from low- (2 Ω⋅cm) and high-resistivity (6.2 Ω⋅cm) substrates, respectively. It was found that initial efficiency of In-doped PERC cells was ∼0.2% lower due to lower bulk lifetime and higher back-surface recombination velocity. However, In-doped PERC cells showed no LID and surpassed the B-doped PERC cell efficiency by 0.3–0.5% after 0.8-sun 48-h illumination at 37 °C.

We examine the potential to enhance cell performance using wide bandgap metal oxide films as full-area rear contacts to p-type crystalline Si (c-Si) solar cells. We aim to introduce a band offset through wide bandgap nickel oxide rather than introducing a band bending through transition metal oxides (e.g. MoO x , V 2 Ox, WO x ). Our numerical simulation shows it is possible to achieve one-sun efficiency of 21.6% with the open-circuit voltage (V oc ) of 652 mV, the short-circuit current density (J sc ) of 39.9 mA/cm2 and the fill factor (FF) of 82.5% when the back surface recombination velocities is 100 cm/s at p-Si/NiO x .

Field effect passivation by negative charge is important for back surface of p-type PERC and front surface of n-type PERT with boron emitter. Currently, Al 2 O 3 passivation is widely used for both surfaces due to its negative charge. However, Al 2 O 3 tool has high operation cost and safety related issues due to the use of TMA precursor. In this paper, a novel plasma charging tool is used to inject negative charge into the SiNx/SiO 2 passivation stack to passivate boron emitter and boron-doped silicon surface. Large area bifacial n-type PERT and p-type PERC cells were fabricated with controlled negative charge injection to demonstrate the effectiveness of this technique. Field effect passivation by injecting 1E13cm−2 negative charge by this plasma charging method gave ∼1.35% and ∼0.36% increases in absolute cell efficiencies for n-type PERT and p-type PERC cells, respectively. Detailed characterization and analysis show that these improvements are due to enhanced surface passivation resulting from the formation of accumulation layer on p-type surfaces which reduces the SRH recombination and shunting associated with depletion and inversion layers.

Al 2 O 3 film with SiNx capping layer is widely used for rear side passivation of p-type PERC cells and passivation of p+ emitter in n-PERT cells because of very effective field-induced passivation by high density of negative charge in Al 2 O 3 (5e12∼1e13cm−2). This paper reports on a promising field-effect passivation by charge injection in SiO2/SiNx stack using a novel low-cost plasma charging method which can replace plasma ALD Al2O3. In addition, this tool injects either positive or negative charge in a controlled manner. It is demonstrated that emitter saturation current density(Joe) of a SiO2/SiNx passivated boron emitter decreases from ∼80fA/cm2 to ∼50fA/cm2 after −7.9e12cm−2 negative charge injection, which is equivalent to the Al2O3/SiNx passivated boron emitter. In addition, a 0.4% increase in absolute efficiency was observed after the injection of 1e13cm−2 negative charge in the SiO 2 /SiNx passivated boron emitter. Sentaurus device modeling was performed to estimate the impact of field-effect passivation by extracting Joe values as a function of injected charge in SiO 2 /SiNx passivated boron and phosphorus emitters. It was found that charge injection is more effective for boron emitters. And the field-effect passivation quality saturated after ∼1e13cm−2 charge in both types of emitters. We expect negative and positive charging on both sides of the cell structure will further enhance field-effect passivation and achieve even higher cell efficiency.

This paper shows the results and the limitations of a 21% N-Cz 239cm2 screen printed cell with blanket p+ and n+. In addition, we show the properties and impact of tunnel oxide capped with doped n+ polysilicon and metal on the back side which can overcome those limitations. Since both the doped n+ layer and the metal contact are outside the bulk silicon wafer, the Jo is dramatically reduced resulting in much higher Voc. Process optimization resulted in high iVoc of 728mV on the symmetric structures. The un-metallized cell structure with Al2O3/SiN passivated lightly doped p+ emitter and a tunnel oxide / n+ poly back also gave high iVoc of 727mV. The finished screen-printed 132cm2 device gave a Voc of 683mV, Jsc of 39.4mA/cm2, FF of 77.6% and an efficiency of 20.9%. Cell analysis show that implementation of a selective emitter can give higher efficiency.

In this paper, novel and promising high efficiency light-induced degradation (LID) free indium-doped Cz Si cells are presented. Two different commercial grade and large area B-doped Cz materials were included for comparison. Ion-implanted large area (239 and 242.22 cm2) screen printed full Al-BSF cells as well as passivated emitter rear contact (PERC) cells with oxide passivation and local aluminum back surface field were fabricated. In-doped PERC cells achieved 20.3% efficiency while the B-doped cells gave the efficiencies of 20.7% and 20.5% from low (2 Ω-cm) and high resistivity (6.2 Ω-cm) substrates, respectively. Although the initial efficiency of In-doped PERC cells was slightly lower than B-doped cells, In-doped PERC cells surpassed the low and high resistivity B-doped PERC cells by 0.5% and 0.3%, respectively, in absolute efficiency after 0.8 sun 48-hour illuminations at 37°C.

In this work, we investigate a transparent conducting oxide (TCO)-based, passivated contact for the potential use as a passivated tunnel contact to p-type Si. As a surface passivation layer, the Al2O3 films with varying the thickness are deposited using plasma-enhanced atomic layer deposition (PEALD) at 200 °C, followed by post-deposition annealing. For a ∼15 nm thick Al2O3 layer, a high level of surface passivation is achieved, characterized by the effective surface recombination velocity (Seff,max) of

We examine thermally evaporated MoO x films as a full-area rear contact to crystalline p-type Si solar cells for efficient hole-selective contacts. Prior to front- and rear-metallization, the implied open-circuit voltage (iV oc) is evaluated to be 646 mV with implied fill factor (iFF) of 82.5% for the tunnel SiO x /MoO x rear contacted cell structure with the passivated emitter on the textured surface, showing it is possible to achieve an implied 1-sun efficiency of 20.8%. Numerical simulation reveals that the electron affinity (χ) of the MoO x material strongly influences the performance of the MoO x contacted p-Si cell. Simulated band diagrams show that the values in χ of the MoO x layer must be sufficiently high in order to lower junction recombination, indicating that the highest efficiency of 21.1% is achievable for a high χ of 5.6 eV of MoO x films and back surface recombination velocity of

Both ion implanted and POCl 3 diffused emitters are used for industrial production of p-type Si solar cells. Formation of phosphorus doped emitter is known to perform gettering of impurities, however, gettering efficiency of these two diffusion techniques is not well quantified for single crystal Cz Si and Cast multi or Quasi-mono Si wafers. In addition, ion implantation can provide higher quality emitter with in situ oxide passivation[1]. This paper compares the performance of Quasi-mono and Cz Si cells fabricated with POCl 3 and ion-implanted emitters. Quasi-mono wafers have more defects and are expected to benefit from gettering. Large area screen printed p-type industrial cells with full Al-BSF cell structure were fabricated on commercial grade single crystal Cz Si and two different Cast Quasi-mono Si wafers with 50% and 80% mono-crystalline regions. Bulk lifetime was measured to evaluate the gettering efficiency of each technology before and after each emitter formation[2]. POCl 3 diffusion gave greater improvement in bulk lifetime of Quasi-mono wafers compared to ion implanted wafers, resulting in 0.7 ∼ 1% higher absolute efficiency and over 5 ∼ 15mV higher Voc compared to the implanted cells. However, in the case of Cz cells, bulk lifetime remained high and comparable for the two emitters. Therefore, Cz cells with implanted emitter gave 0.4% higher efficiency and 7mV higher Voc due to a higher quality emitter with in situ front oxide passivation.

Front metal contact induced recombination and resistance are major efficiency limiting factors of large-area screen-printed n-type front junction Si solar cells with homogeneous emitter and tunnel oxide passivated back contact (TOPCON). This paper shows the development of a selective boron emitter (p+/p++) formed by a screen-printed resist masking and wet chemical etch-back process, which first grows a porous Si layer and subsequently removes it. Various wet-chemical solutions for forming porous Si layer are investigated. An industrial compatible process with sodium nitrite (NaNO2) catalyst is developed to uniformly etch-back the ∼47 Ω/◻ atmospheric pressure chemical vapor deposited heavily doped boron emitter to ∼135 Ω/◻ by growing a 320 nm porous Si layer within 3 min and subsequently removing it. After etching back, the boron emitter was subjected to a thermal oxidation to lower the surface concentration and the emitter saturation current density J0e. Various etched-back emitters were evaluated by meas...