One Polarized Scan Maps an Entire 2D Alloy Series — and Its Dual Emission
Most 2D semiconductors emit light one way. This one does it two ways at once — and which way depends on how thick the crystal is.
Rhenium disulfide (ReS2) and its selenium-substituted cousin ReSe2 belong to a small, unusual family of 2D transition-metal dichalcogenides: instead of the hexagonal symmetry that gives materials like MoS2 their familiar in-plane uniformity, ReX2 compounds crystallize in a distorted, triclinic structure built from chains of Re4 metal clusters. That distortion makes the material’s optical, electrical, and structural properties genuinely different depending on which in-plane direction you measure along — a property called in-plane anisotropy that’s of real interest for polarization-sensitive photodetectors, flexible optoelectronics, and next-generation transistors.
The Obstacle — Direct or Indirect? The Answer Kept Changing With Thickness
For a 2D semiconductor to be useful in light-emitting devices, its band gap generally needs to be direct — meaning an electron can drop straight to the valence band and emit a photon, rather than needing an extra momentum-changing step. Whether bulk ReS2 has a direct or indirect band gap has been a genuinely unsettled question: absorption measurements in bulk material pointed to an indirect gap, but photoluminescence (PL) measurements kept showing an emission peak consistent with a direct transition, even in thick crystals where indirect behavior would normally be expected to dominate. What was missing was a single measurement series, across the entire ReS2-to-ReSe2 compositional range and across a wide range of crystal thicknesses, that could show both behaviors — direct and indirect — showing up in the same material and explain when each one appears.
The Innovation — Polarized μPL Across an Entire Alloy Series
In a study published in Nanotechnology (Ho, Liu & Lin, 2017, Nanotechnology, 28, 235203, DOI: 10.1088/1361-6528/aa6f51), researchers at National Taiwan University of Science and Technology grew the complete ReS231Se3 series (x = 0 to 2, in steps of 0.2) by chemical vapor transport, then systematically measured polarized micro-photoluminescence (μPL) across the entire composition range — with light polarized both along and perpendicular to the crystal’s distorted b-axis — on samples ranging from thin multilayers (under 75 nm) to bulk crystals (over 1 μm thick).
Polarization is the key variable that had been missing from earlier PL studies of this material: because the crystal’s Re4 cluster chains only run in one in-plane direction, the two dominant excitonic transitions near the band edge — labeled E1ex and E2ex — should, in principle, be individually excitable by choosing the polarization direction. If that separation could be demonstrated cleanly and consistently across the whole alloy series, it would settle both the direct/indirect question and establish a practical route to polarization-selective ReX2 light-emitting devices. Crystal growth and stoichiometry across the series were independently verified by HRTEM and energy-dispersive X-ray analysis: lattice constants increase smoothly with selenium content, following Vegard’s law almost exactly — direct structural confirmation that the whole compositional series forms a genuine, continuously tunable solid solution rather than a phase-separated mixture.
Key Results — Two Polarizations, Two Excitons, Two Emission Pathways
The case builds across five results: separate the excitons by polarization, confirm that separation is a continuous physical response, show how it tunes with composition, show what changes with thickness, and confirm the thickness-dependent feature independently by a second optical method.
Two Perpendicular Polarizations Cleanly Separate the E1 and E2 Excitons
Polarized μPL spectra of ReS2, ReS1․1Se0․2, ReS1․)Se0․6, ReS0․1Se0․4Se0․4, ReS0․4Se1․), and ReSe2 show the E1ex exciton (~1.463 eV in ReS2) emerging clearly along the b-axis polarization, and the E2ex exciton (~1.502 eV) emerging perpendicular to it — a clean polarization-dependent split that matches, within about 0.02 eV, prior polarized absorption measurements of the same transitions.
Rotating the Polarizer 180° Traces Out a Clean Sinusoidal Intensity Swing
Sweeping the polarization angle continuously from 0° to 180° in a ReS1․1Se0․2 sample shows the E1ex and E2ex emission intensities following clean |cos θ| and |sin θ| dependences respectively — textbook dichroic behavior that confirms the polarization sensitivity isn’t an artifact of two fixed measurement angles, but a continuous, predictable physical response tied to the crystal’s b-axis orientation.
Exciton Energies Shift Predictably as Selenium Replaces Sulfur
Across the full x = 0 to 2 series, both E1ex and E2ex decrease smoothly in energy as selenium content increases, fitting a quadratic composition dependence with small bowing parameters. The energy separation between them, ΔE = E2ex − E1ex, shrinks linearly with selenium content — directly tracking the change in lattice constant as selenium substitutes for sulfur.
Beyond 70 nm, a Second, Indirect Emission Peak Appears and Grows With Thickness
For crystals thinner than about 70 nm, only the direct E1ex/E2ex excitonic emissions appear. Past that thickness, an additional lower-energy peak — labeled I — emerges and strengthens as the crystal gets thicker, reaching its strongest, most resonant form in bulk crystals over 1 μm. Both direct and indirect emission are present simultaneously in these thicker crystals, meaning the same piece of ReS231Se3 can be probed either way depending on where the laser excites.
Transmittance Confirms the Extra Peak Is a True Indirect Optical Gap
Comparing μPL emission to transmittance absorption spectra in three bulk samples (ReSe2, ReS0․1Se1․2, and ReS2) shows the absorption edge lines up with the new I emission peak, not with the direct Eex peak — independent confirmation, by a completely different optical method, that the lower-energy feature really is an indirect band-edge transition and not an artifact of the PL measurement itself.
Why ProTrusTech Is Featuring This Research
Every exciton energy, every polarization ratio, and every direct/indirect assignment in this paper comes from a single measurement technique — polarized micro-photoluminescence, performed with a RAMaker system — applied consistently across eleven compositions and a thickness range spanning two orders of magnitude. It’s a demonstration of how much a well-controlled polarization-resolved PL measurement can carry on its own: the same instrument and method that identifies exciton energies at 1.5 eV in a 23 nm flake also resolves an indirect emission feature in a 1.2 μm bulk crystal, without switching to a different characterization technique in between.
Engineering the Solution
The measurement approach here is deliberately simple to reproduce: samples are exfoliated from bulk crystals with ordinary tape, transferred to a SiO2/Si substrate, and their crystallographic b-axis identified from AFM images showing the characteristic anisotropic edge. From there, polarized μPL with a rotatable linear polarizer, at room temperature, is sufficient to extract every result in the paper — no cryogenic cooling, no specialized sample preparation beyond standard mechanical exfoliation.
Featured Instruments:
Why This Matters
This paper is a useful reference point for the site’s broader body of polarized and angle-resolved work: it shows that polarization sensitivity in a layered crystal isn’t just a curiosity to be noted in passing — used systematically across an entire compositional series, it becomes the tool that resolves a genuinely open structural question (direct vs. indirect band gap) that absorption measurements alone had left ambiguous. The general lesson extends beyond ReX2: in any low-symmetry, anisotropic 2D material, a single polarization-resolved PL or Raman measurement, applied across composition and thickness, can carry far more diagnostic weight than an unpolarized spectrum from the same instrument.
What’s Next
- Low-temperature polarized μPL. The current measurements are at room temperature; cooling could sharpen the exciton features and potentially resolve the still-uncertain origin of the ESex emission series.
- Device-level polarization-sensitive light emitters. The authors point to flexible, near-infrared (1.2–1.6 eV) polarized light-emitting devices as the natural application target for this material family.
- Deeper theoretical treatment of the ESex transition. Density-of-states calculations offer a tentative origin, but the authors flag this as needing further theoretical and low-temperature experimental confirmation.
Working on polarization-resolved photoluminescence, anisotropic 2D material characterization, or micro-PL mapping of your own? Get in touch with our team.
Key Takeaways
- Polarized micro-photoluminescence, performed on a RAMaker system, cleanly separates the E1 and E2 band-edge excitons of the full ReS231Se3 (x = 0–2) alloy series by polarization direction.
- Exciton energies shift smoothly and predictably with selenium content, following a quadratic composition dependence tied directly to the crystal’s changing lattice constant.
- Below ~70 nm thickness, only direct exciton emission appears; above it, a second, indirect emission peak emerges and strengthens with increasing thickness.
- Transmittance measurements independently confirm that the thickness-dependent peak is a genuine indirect optical transition, not a PL artifact.
- Direct and indirect light emission coexist in the same bulk crystal — a distinguishing feature the authors attribute to the material’s uniquely weak interlayer coupling and in-plane anisotropy.
