Few-Layer MoS$_2$/a-Si:H Heterojunction pin-Photodiodes for extended Infrared Detection

Andreas Bablich; Benjamin Butz; Chanyoung Yim; Daniel S. Schneider; Georg S. Duesberg; Julian M\"uller; Max C. Lemme; Niall McEvoy; Olof Engstrom; Paul Kienitz

arxiv: 1907.09592 · v1 · pith:M3U6WVZAnew · submitted 2019-07-22 · ⚛️ physics.app-ph · cond-mat.mes-hall

Few-Layer MoS₂/a-Si:H Heterojunction pin-Photodiodes for extended Infrared Detection

Andreas Bablich , Daniel S. Schneider , Paul Kienitz , Satender Kataria , Stefan Wagner , Chanyoung Yim , Niall McEvoy , Olof Engstrom

show 6 more authors

Julian M\"uller Yilmaz Sakalli Benjamin Butz Georg S. Duesberg Peter Haring Bol\'ivar Max C. Lemme

This is my paper

Pith reviewed 2026-05-24 17:21 UTC · model grok-4.3

classification ⚛️ physics.app-ph cond-mat.mes-hall

keywords few-layer MoS2a-Si:Hpin-photodiodesinfrared detectionheterojunctionphotodetectorsresponsivity switching

0 comments

The pith

Few-layer MoS2 integrated with a-Si:H creates pin-photodiodes that detect 2120 nm infrared light with sensitivities up to 50 mA/W and remain stable over six months.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper shows how few-layer MoS2 films can be combined with amorphous silicon pin photodetectors to extend their detection into the infrared. This integration allows the devices to achieve higher sensitivity at 2120 nm than standard silicon detectors, along with good detectivity and quantum efficiency. The heterojunction structure also permits switching the responsivity between different wavelengths by adjusting the bias voltage. Fabrication uses standard plasma deposition on transferred MoS2, suggesting scalability for arrays. A sympathetic reader would see this as a step toward practical, stable IR sensors based on 2D material-silicon hybrids.

Core claim

Few-layer molybdenum disulfide films integrated into amorphous silicon pin photodetectors via plasma-enhanced chemical vapor deposition yield devices with long-term stability exceeding six months that outperform conventional silicon pin photodetectors in the infrared at 2120 nm with sensitivities up to 50 mAW^{-1}, photodetectivities up to 2 x 10^{10} Jones, and external quantum efficiencies of 3%, while also enabling bias-dependent responsivity switching between spectral ranges.

What carries the argument

The vertical a-Si:H pin photodiode structure grown on transferred few-layer MoS2, forming a heterojunction that extends infrared absorption.

Load-bearing premise

The performance improvements in the infrared arise from the presence of the MoS2 layer forming the heterojunction rather than from changes in the a-Si:H material quality or measurement artifacts.

What would settle it

Measuring the infrared responsivity at 2120 nm for a-Si:H pin photodiodes fabricated identically but without the MoS2 layer and finding no reduction compared to the heterojunction devices would falsify the role of the interface.

read the original abstract

Few-layer molybdenum disulfide (FL-MoS$_2$) films have been integrated into amorphous silicon (a-Si:H) pin photodetectors. To achieve this, vertical a-Si:H photodiodes were grown by plasma-enhanced chemical vapor deposition (PE-CVD) on top of large-scale synthesized and transferred homogeneous FL-MoS$_2$. This novel detector array exhibits long-term stability (more than six month) and outperforms conventional silicon-based pin photodetectors in the infrared range (IR, $\lambda$ = 2120 nm) in terms of sensitivities by up to 50 mAW$^{-1}$. Photodetectivities of up to 2 x 10$^{10}$ Jones and external quantum efficiencies of 3 % are achieved. The detectors further feature the additional functionality of bias-dependent responsivity switching between the different spectral ranges. The realization of such scalable detector arrays is an essential step towards pixelated and wavelength-selective sensors operating in the IR spectral range.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

They built a working FL-MoS2/a-Si:H photodiode array with reported 2120 nm response and bias switching, but the paper does not isolate the heterojunction as the source of the gains.

read the letter

The main point is that this group transferred large-area few-layer MoS2, then grew a-Si:H pin diodes on top by PE-CVD, and measured a detector that stays stable for over six months while showing up to 50 mA/W responsivity, 2e10 Jones detectivity, and 3% EQE at 2120 nm, plus bias-dependent switching between spectral ranges. That combination of materials and the vertical integration on transferred films is the concrete new step beyond earlier 2D-silicon work. The fabrication route looks practical for scaling, and the long-term stability number is useful to see in print. The device functions as described in the abstract, and the numbers are specific enough that a referee could check them against the figures and methods. The soft spot is the missing controls. The abstract and stress-test note give no mention of identical a-Si:H stacks without the MoS2 layer, no TEM or electrical profiling of the junction itself, and no quantification of transfer defects or run-to-run a-Si:H variation. Without those, the claim that the heterojunction itself drives the IR outperformance over plain silicon pin diodes rests on correlation rather than isolation. The bias switching is interesting on its face but faces the same issue. This paper is for people building hybrid 2D/amorphous silicon detectors who need a working example of the stack. It is not a foundational result, but the experimental demonstration is solid enough on its own terms to warrant referee time. I would send it to review so the authors can add the control data or clarify what they already have.

Referee Report

2 major / 2 minor

Summary. The manuscript reports integration of few-layer MoS2 into a-Si:H pin photodetectors via transfer of large-scale FL-MoS2 followed by PE-CVD growth of vertical a-Si:H pin diodes on top. It claims the resulting arrays exhibit >6-month stability, responsivity up to 50 mA/W at 2120 nm (outperforming conventional Si pin detectors), detectivity up to 2×10^{10} Jones, EQE of 3%, and bias-dependent responsivity switching between spectral ranges.

Significance. If the IR performance metrics are reproducible and attributable to the heterojunction rather than process variations, the work would demonstrate a scalable route to extend a-Si:H detector response into the infrared while retaining compatibility with existing thin-film technology, potentially enabling pixelated wavelength-selective sensors.

major comments (2)

[Abstract] Abstract: The claim that performance gains at λ=2120 nm (50 mA/W responsivity, 2×10^{10} Jones detectivity) arise specifically from the FL-MoS2/a-Si:H interface is unsupported, as no control devices (identical a-Si:H pin stacks without MoS2), no interface characterization (TEM, Raman, or electrical profiling), and no quantification of transfer-induced defects or a-Si:H film variation are mentioned.
[Abstract] Abstract: Specific numerical claims (50 mAW^{-1}, 2×10^{10} Jones, 3% EQE) are presented without raw data, error bars, baseline comparisons to control devices, or detailed measurement protocols, preventing assessment of post-hoc selection or reliability.

minor comments (2)

[Abstract] Abstract: 'six month' should read 'six months'.
[Abstract] Abstract: 'sensitivities' is imprecise; align terminology with 'responsivity' used later in the same sentence.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major comment below and will revise the manuscript accordingly to strengthen the presentation of our results.

read point-by-point responses

Referee: [Abstract] Abstract: The claim that performance gains at λ=2120 nm (50 mA/W responsivity, 2×10^{10} Jones detectivity) arise specifically from the FL-MoS2/a-Si:H interface is unsupported, as no control devices (identical a-Si:H pin stacks without MoS2), no interface characterization (TEM, Raman, or electrical profiling), and no quantification of transfer-induced defects or a-Si:H film variation are mentioned.

Authors: We agree that explicit control experiments are necessary to attribute the observed IR response specifically to the heterojunction. In the revised manuscript we will add responsivity data from control a-Si:H pin devices fabricated on identical substrates without the transferred FL-MoS2 layer. These controls will be shown alongside the heterojunction devices at 2120 nm. We will also include Raman spectra of the MoS2/a-Si:H interface and a brief discussion of transfer-related defects; any available cross-sectional TEM images will be placed in the supplementary information. revision: yes
Referee: [Abstract] Abstract: Specific numerical claims (50 mAW^{-1}, 2×10^{10} Jones, 3% EQE) are presented without raw data, error bars, baseline comparisons to control devices, or detailed measurement protocols, preventing assessment of post-hoc selection or reliability.

Authors: The quoted values summarize the peak performance obtained from the device arrays described in the results section. In the revision we will (i) append representative error bars or ranges to the abstract numbers, (ii) expand the methods section with explicit measurement protocols (illumination source, lock-in settings, device area definition), and (iii) add the control-device comparisons noted above. Raw data traces will be supplied in the supplementary information. revision: yes

Circularity Check

0 steps flagged

No circularity: experimental measurements with no derivations or fitted models

full rationale

The paper is a device fabrication and characterization report. It describes growth of a-Si:H pin diodes on transferred FL-MoS2, reports measured responsivity (up to 50 mA/W at 2120 nm), detectivity (2e10 Jones), EQE (3%), stability (>6 months), and bias-dependent switching. No equations, no parameter fitting, no predictions derived from models, and no self-citation chains are present in the abstract or described structure. Claims rest directly on experimental data rather than any self-referential logic. This matches the default non-circular case for pure experimental work.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities; the paper is an experimental materials integration report relying on standard device physics assumptions from the literature.

pith-pipeline@v0.9.0 · 5768 in / 1187 out tokens · 23582 ms · 2026-05-24T17:21:56.471608+00:00 · methodology

Few-Layer MoS₂/a-Si:H Heterojunction pin-Photodiodes for extended Infrared Detection

Core claim

What carries the argument

Load-bearing premise

What would settle it

discussion (0)