Engineering topics

Harvey, Steve (ORCID:0000000161207062)

Publications and source records attributed to Harvey, Steve (ORCID:0000000161207062).

7 records

DOE OSTI2023

Understanding Hydrogen Passivation Mechanism in Poly-Si Passivating Contacts: Insights from Effusion Studies

Silicon PV is the dominant PV technology, and captures >95% world market share, supporting -100 GW/year. Diffused and passivated pn junctions along with other rear side passivation technologies are the mainstream technologies and will continue to remain so in the coming decades. Passivated contacts, using tunnel oxide passivation stacks at the rear side, will gain market share from about 10% in 2022 up to 58% within the next 10 years. Most mature approaches use passivating layers of hydrogenated Al2O3 and SiNx. LPCVD SiNx provides less hydrogen for passivation compared to PECVD SiNx but can retain the hydrogen up to higher temperatures which is beneficial for firing. Although SiNx provides large amount of hydrogen, this doesn't necessarily translate to good passivation. AlOx:H contributes to retaining H at higher temperatures. Water molecules are also detected during H-effusion and may play a role in passivation of the oxide/wafer interface.

effusion study↗

DOE OSTI2023

Self Assembled Monolayers for Passivated Contacts

Passivated contacts mitigate defects typically encountered due metallization of solar cells. We deposit amorphous silicon (a-Si:H) on an oxidized silicon wafer via PECVD and anneal at high temperature to crystallize into polysilicon passivated contact. One drawback is the absorption of the polysilicon between grid fingers, so removal of this material is desirable to maximize Jsc. Alternatively, interdigitated back contact cells rely on a gap between n- and p- fingers, which is commonly etched to ensure electronic isolation. We utilize a self assembled monolayer (SAM) using hexamethyldisilazane (HMDS) as a precursor to pattern and etch amorphous silicon (a-Si:H) and polysilicon without the need for photoresist. Ultraviolet light exposure oxidizes the HMDS by photocleaving the organic groups [1] of the SAM leaving a patterned SiO2. Directly soaking this in TMAH will eventually etch the SAM and the silicon, where the SiO2 serves as an etch mask. Inversely, a dilute HF dip selectively etches this SiO2 and the SAM remains. A subsequent soak in TMAH selectively etches the underlying silicon, where the SAM serves as an etch mask. Importantly, we find that the SAM can remain intact for metallization, where we measure 10 mO-cm2 specific contact resistivity on n-type polysilicon.

passivation↗

DOE OSTI2023

Mechanism of Hydrogen Passivation in Passivated Contact Si Solar Cells

In this abstract, we discuss the mechanism of hydrogen passivation on symmetrical n-Si/ultra-thin SiO2/polySi structures. The hydrogen was introduced from different hydrogen-containing dielectric layers (AlOx:H and SiNx:H and their stacks), as well as by forming gas anneal (FGA). The effusion of hydrogen both from the dielectric layers and in the underlying poly-Si was explored using a quadrupole mass spectrometer (QMS) and FTIR spectroscopy. We show that the strength of hydrogen bonds depend on the deposition technique as well as hydrogenation mechanism. While a PECVD SiNx:H loses hydrogen at a peak temperature of ~450 degrees C, LPCVD SiNx:H, although having less hydrogen compared to PECVD, effuses at a peak temperature of ~850 degrees C. This becomes important to maintain passivation of passivated contacts after firing of metal contacts at high temperatures. On the other hand, it has been observed that SiNx:H provides larger amount of hydrogen to poly-Si after FGA compared to FGA treated AlOx:H, which acts as a capping layer and helps in retaining the hydrogen up to higher temperatures.

hydrogen bonding configuration↗

DOE OSTI2022

Dopant Compensation within the Intrinsic Poly-Si Isolation Region in Poly-Si/SiOx Passivated IBC Si Solar Cells

We report on the effect of dopant compensation within intrinsic poly-Si regions between p- and n-type fingers of poly-Si/SiOx passivated interdigitated back contact (IBC) solar cells using intrinsic poly-Si as the isolation region between the doped poly-Si fingers. First, we show that dopants from the doped fingers contaminate the intrinsic gap, resulting in doping of the entire intrinsic gap and overlap of the dopant tails from each finger. Next, we show that despite this doping across the gap, shunting between the doped fingers does not occur. We show that this is a result of trap-assisted compensation creating a highly resistive intrinsic region, preventing shunt. We simulate shunt resistance across the gap based on local carrier concentration and deep trap density. We show that trap defects within the poly-Si enhance compensation between the dopant tails. We experimentally confirm these predictions by scanning spreading resistance microscopy of the gap showing ~20 um domain with resistivity ~10^7 ..omega.. cm. Additionally, Kelvin probe force microscopy are compared to finite element simulations which result in the same approximate shape for potential profile, indicating diode behavior across the isolation region. These results demonstrate the powerful effect that trap defects have within the poly-Si isolation region and suggest that precision patterning is not as essential as once thought.

dopant compensation↗

DOE OSTI2021

Trap-Assisted Dopant Compensation Prevents Shunting in poly-Si Passivating Interdigitated Back Contact Silicon Solar Cells

Interdigitated back contact (IBC) solar cells achieve the highest efficiencies of single-junction architectures, but complicated patterning of the rear fingers and spreading of dopants during processing inhibit their mainstream adoption due to concerns of shunting between the IBC fingers. One method of simplifying patterning at the rear is by using contact masks combined with plasma-enhanced chemical vapor deposition (PECVD) or ion implantation. However, the intrinsic isolation region becomes contaminated during high-temperature annealing by lateral diffusion of dopants and during masked PECVD by spreading of dopant radicals through region between the mask and the substrate. Despite this contamination, we show through scanning spreading resistance microscopy and Kelvin probe force microscopy that a ~20 µm wide compensating region exists with high enough resistivity to prevent shunting. We model this p-i-n poly-Si system using two simulation models: a simple resistor model considering only the capture of charge carriers by trap defects in poly-Si to reduce the conductivity, and a more refined 1-dimensional finite element model using Poisson’s equation, drift-diffusion equations, and recombination of carriers. Using this model, we show that high defect density significantly decreases the current across the region between the p- and n-type fingers, preventing shunting.

atom probe↗

DOE OSTI2020

Synthesis of CdSeCdSexTe1-x/CdTe for Graded Solar Cells

Here, we have developed CdSe x Te 1-x /CdTe bilayer deposition for graded solar cell technology. Smoothly graded Se profiles with > 16% efficiency with Jsc=28 m A/cm 2 and 20-ns bulk lifetime has been observed observed. The CdSeTe reveals longer carrier lifetimes than CdTe within the same device.

41 EE - Solar Energy Technologies Office (EE-4S)↗

DOE OSTI2020

Chapter 15: Summary and Outlook

In the various chapters of this book, numerous characterization techniques are presented that can be applied to thin-film solar cells to determine (micro)structural, compositional, electrical, and optoelectronic properties. What has yet to be more fully elucidated is to what extent these characterization techniques can be combined, in a correlative way, to enhance the information gathered on materials and devices. Indeed, it is valuable to consider combining techniques to verify the relevance of the measured materials properties or to obtain them on different length scales-to compare surface with bulk properties, or to correlate structure and composition of materials with electrical and optoelectronic properties.

41 EE - Solar Energy Technologies Office (EE-4S)↗