Access Official Technical PDF

Share Summary:LinkedInEmail

InP nanowire gate control offers a way to change semiconductor carrier populations electrostatically while observing their optical response. A transparent indium-tin-oxide (ITO) gate makes that combination possible: researchers can apply bias to nanowires and measure photoluminescence (PL) without blocking the probing light. Measurements on gated InP nanowires show, however, that gate response depends strongly on doping, diameter and the condition of the nanowire surface.

These findings matter when designing nanoscale optoelectronic devices or experiments that rely on controlled carrier populations. In structures with a high surface-to-volume ratio, interfaces are not merely packaging details: they can shape the electrostatics and determine whether radiative emission is observed.

Why use a transparent wrap gate?

A wrap gate surrounds a nanowire, giving the gate access to more of its circumference than a planar electrode would. The structures examined included vertical InP nanowire arrays with an all-around ITO gate and individual lateral nanowires with an omega-shaped gate. In both cases, the transparent electrode was intended to provide electrical control while allowing optical measurements.

The array structures included an approximately 10 nm HfO2 gate dielectric between the InP and an approximately 30 nm ITO layer. InP is a direct-bandgap III-V semiconductor with useful optical properties. Its nanowire geometry also makes it possible to study electrical and optical behavior in very small volumes, although the large surface area relative to volume makes surface conditions especially influential.

The broader experimental goal was to control electrons and holes with gate bias and investigate the formation of spatially separated excitons—bound electron-hole pairs. The PL studies instead highlighted a more immediate challenge: the gate and its interfaces can alter the nanowire’s emission and carrier distribution.

What photoluminescence reveals

In PL spectroscopy, a laser excites electrons across the semiconductor bandgap, creating electron-hole pairs. As carriers relax, some recombine radiatively and emit photons. The emitted spectrum provides information about optical transition energies, while its intensity can help indicate changes in carrier populations and recombination.

At sufficiently high excitation, available states can become occupied. This state-filling effect can shift emission toward higher energy, a behavior related to the Burstein-Moss effect. The measurements reported a substantially larger excitation-power-related blueshift in p-doped samples than in n-doped samples: about 110 meV compared with about 10 meV. A proposed explanation considers the different effective masses of electrons and holes in InP. Because the conduction-band electrons have a lower effective mass than the holes, adding minority carriers can spread them across a different energy range, producing different shifts in the two doping conditions.

This interpretation is a model for the observed difference, not a universal rule for every nanowire device. Band bending, surface states and carrier recombination also influence the measured spectrum.

Surface depletion and interface traps

At a semiconductor surface, the crystal structure terminates and can introduce electronic states within the bandgap. These interface states can capture charge and affect the local electrostatic potential. The resulting band bending may deplete a region of mobile carriers near the surface. In a nanowire, that depleted region can extend across much or all of the cross-section because the surface is close to the core.

The impact depends on nanowire radius, doping and interface-trap density, commonly denoted as Dit. If depletion removes many of the carriers needed for radiative recombination, PL can weaken. The modelling described in the study examined how radius, surface defects and gate bias could change band bending and depletion. It also suggested that surface effects may become strong enough in smaller wires to limit the influence of applied gate bias.

Measurements of ungated arrays showed that PL intensity varied with nanowire diameter. Undoped and p-doped wires had a pronounced increase as diameter grew, while the n-doped wires showed a much weaker change. The diameter trend was not explained simply by a proportional increase in nanowire volume. Surface depletion and interface states were proposed as factors that could account for the radius dependence.

Gate response depends on doping and structure

N-doped and intrinsic nanowire arrays showed changes in PL intensity with gate bias. For n-doped arrays, emission increased after a bias threshold, consistent with the expected behavior of carrier accumulation in a gated semiconductor structure. At higher bias, leakage rose and the gate dielectric ultimately experienced electrical breakdown. This illustrates why gate-response measurements need to be interpreted alongside leakage behavior and device integrity.

The p-doped arrays behaved differently: their PL was suppressed after gate processing, and they did not show a clear response to bias. A single lateral p-doped nanowire showed a possible bias-dependent response, but the measurement was noisy and came from only one device. The evidence therefore does not establish a consistent p-type response across structures.

Diameter also affected the observed response. In the n-doped arrays, the relative increase in PL with applied bias was small for the thinnest wires and larger for wider ones. This runs counter to a simple expectation that thinner wires should be easier to influence because of their greater surface-to-volume ratio. One possible explanation is that surface depletion and interface traps already dominate the electrostatics of the thinner wires, leaving less scope for the gate to change their carrier population.

Practical implications for nanowire device design

For engineers and researchers developing gated nanowire structures, these results emphasize the need to treat the interface as part of the device design. A transparent gate can enable simultaneous optical access and electrostatic control, but transparency alone does not guarantee that the gate leaves the semiconductor’s optical behavior unchanged.

  • Characterize the surface and interface: Interface traps can contribute to band bending, depletion and changes in emission.
  • Account for diameter and doping: Gate response and PL trends may differ between n-type, p-type and undoped wires, as well as across nanowire radii.
  • Monitor leakage and breakdown: Changes in emission at high bias should be considered together with dielectric leakage and device condition.
  • Separate optical and electrostatic effects: Comparing gated and ungated structures, and considering possible processing residues, can help identify why PL changes.

The modelling also has limits. The electrostatic calculations considered interface-trap effects and gate bias, while the separate semiconductor simulation did not include interface-state pinning or all possible contact effects. Models are therefore useful for framing hypotheses, but they do not remove the need to validate behavior in fabricated devices.

Conclusion

InP nanowire gate control can support optical measurements while changing the electrostatic environment, but its effectiveness is governed by more than gate geometry. Photoluminescence depends on carrier state filling, doping, diameter, surface depletion and interface quality. For practical device development, careful characterization of the gate-semiconductor interface—and evaluation of emission alongside bias and leakage—is essential to distinguish useful electrostatic control from surface-related limitations.

Access Official Technical PDF

About The Author