Developing efficient and cost-effective ultraviolet light-emitting diodes (LEDs), especially in the deepultraviolet (UV-C) is a long-standing goal of the photonics community. UV-C LEDs are of particular interest and could replace gas discharge lamps, revolutionizing disinfection systems, provide water purification, and facilitate non-line-of-sight communication. In light of the on-going global pandemic due to COVID-19, implementation of a rapid surface disinfection method in which chemical cleaning is impractical or not as effective is of critical importance.
Aside from the materials challenge of creating semiconductor-based LEDs that operate in the UV-C portion of the electromagnetic spectrum, these LEDs have been traditionally limited in their power and thermal efficiencies. The wall-plug efficiency, or power-conversion efficiency, is a key figure of merit and is a product of the external quantum efficiency and voltage efficiency. The external quantum efficiency is the product of the light-extraction efficiency and the internal quantum efficiency, which characterizes the efficiency of charge transport, carrier confinement, and carrier radiative recombination in the active region and the voltage efficiency characterizes the energy loss due to the electrical resistance of the metal contacts and the semiconductor layers.
Much of the commercially available 200-280 nm UV LEDs, the portion of the electromagnetic spectrum highlighted in Figure 1, consist of AlN or AlGaN materials grown on sapphire or other exotic substrates by molecular beam epitaxy or chemical vapor deposition methods. There are significant hurdles to these devices being scalable and manufactured at a price that would achieve wide-scale market adoption. These III-V materials also have poor light extraction efficiencies which limit their performance. Additionally, sapphire and most semiconductor-based substrates have large lattice mismatches.
An attractive alternative that circumvents many of the limitations discussed above is the use of semiconductor quantum dot (QD) based LEDs. QDs are semiconductor nanoparticles, scaling between 2 nm and 20 nm, that display unique electronic and optical properties. Their small size (smaller than the exciton Bohr radius) confines the motion of electrons and holes in all three spatial directions, which then creates atomic-like energy level spacing that is inversely proportional to the square of its size.
By changing the quantum dot particle size, engineers can fine-tune the optical properties, including the wavelength of an LED. QDs offer high color purity with a narrow emission wavelength band (typical FWHM < 40 nm) and intense luminescence(~200,000 cd/m2). Furthermore, the quantum confinement, which dictates the optical and electronic properties of the material, enhances tunability from the ultraviolet (which can support COVID-19 decontamination efforts) to the infrared.
The ability to control wavelength (Figure 2), color, and luminescence makes QDs ideal for a number of different industries. The healthcare industry in particular, along with municipalities, transportation and others that have to frequently provide sanitation and decontamination, stand to benefit from the use of QDs in UV-C LED lighting. Coupled with the lowered cost and ability to be thin, optically transparent and flexible, QDs are bringing us closer to mass market adoption of LEDs as the preferred light source.
A material of particular interest is silicon which would have a lower threading dislocation density compared to AlN and a higher internal quantum efficiency. The search for a suitable silicon precursor that could provide lower capex and lower operating costs is highly desirable. One such material is cyclohexasilane (CHS), which The Coretec Group is commercializing. The market needs cost-effective silicon nanocrystal precursors that can be readily chemically functionalized and has found promise in CHS (Si6H12). CHS is a silicon that maintains a liquid state at room temperature, enabling safer and easier storage and handling. It has a moderate boiling point, 80°C at 15 torr, and is thermally stable for long periods of time when stored at 0°C. It has long been considered to be the preeminent silicon precursor for a variety of applications including semiconductor devices, LEDs, and optoelectronics.
Silicon QD LEDs are continuously advancing in flexibility and efficiency, offering the capability to produce “single-wavelength” devices and white light emitters through an ordered distribution of QD sizes. Coupled with the stability and efficiency of silicon nanocrystals, these materials are appropriate for use as light spectrum generators for blue or near-UV high brightness LEDs.
In addition to its efficiency improvements, silicon has proven to be a stable material, showing promise over the last three decades in its longevity for consumer electronics. With the ability to establish a longer lifetime for the LED using a single semiconductor, end consumers for large installations can feel confident in their investments.
Furthermore, silicon QDs can be easily passivated, making them a suitable colloidal suspension in solutions. This achieves scalable deposition methods such as inkjet printing which reduces capital and convenience barriers to processing and commercialization. In addition to the passivation, these Si QDs can be readily functionalized when derived from CHS to contain organic dye molecules which can facilitate up-conversion to produce UV light with high quantum efficiencies.
Creating QDs for UV LEDs suitable for COVID-19 decontamination is entirely reliant on the creation of commercially scalable methods to produce nanostructured materials with well-defined size and morphology below 5 nm in size. Until recently, manufacturers were faced with two options: a traditional bottom-down method (e.g. ball milling) that was inexpensive but would produce what you paid for with high degrees of agglomeration, large particle sizes, and poorly defined morphologies and surfaces, and alternatively, bottom-up options, (e.g. chemical vapor deposition (CVD)) that were able to create well-defined particle sizes and an array of nanostructure articles from nanoparticles and nanowires to thin films, but these come at a high cost and produce low yields and are hazardous due to the use of silane, SiH4.
To find a more balanced middle ground, liquid-silicon nanomaterial replacements are emerging for SiH 4, with CHS leading the charge. CHS requires only mild conditions for the formation of tailored silicon materials and offers a competitive advantage with lower temperature deposition and faster deposition rates compared to SiH4. This is due to its low activation energy (0.3 eV), which is 5-8 times lower than SiH4. This approach is simpler, more robust and far more efficient from both a production and capital standpoint for Si thin-films and nanomaterials development (Figure 4). Cyclohexasilane thus provides an easy avenue to manufacture highly tailored and efficient LEDs based on silicon QDs.
Instead the lattice mismatch issues posed above may be addressed by using an amorphous silicon layer and/or a thin oxide layer between the dissimilar semiconductor materials. The amorphous silicon may be grown in a strain-free method using cyclohexasilane as a reagent. Either of these strategies, or a combination of them, will lead to avoiding lattice mismatch due to interface defects and crystal disorientation. An insulating oxide may also provide an electron and hole tunneling layer. This oxide layer may also be derived from cyclohexasilane with its thickness controlled to act as a passivation and as a tunneling layer. This approach to date has been very limited.
The use of cyclohexasilane to manufacture a highly p-doped membrane on an LED anode represents a novel method to generate these materials. This p-doped silicon membrane material should alleviate some of the previous limits of being too resistive, having poor ohmic contact, and not being amenable to standard semiconductor fabrication. Cyclohexasilane derived p-doped silicon (Figure 5) should have higher metallic character and a lower resulting contact resistance to p-layers, leading to a more electrically efficient LED. This innovation also avoids more costly and less efficient approaches such as polarization doping and tunnel junctions.
One additional advantage that CHS provides is a lower temperature method to produce silicon carbide (SiC). SiC and doped-SiC materials have recently been shown to be promising substrates for optoelectronic devices and would have much lower lattice mismatch issues than sapphire or other semiconductor materials, have a lower threading dislocation density, and would be more cost-effective.
Silicon QD-based LEDs created using CHS can impact a number of different industries but in response to the COVID-19 crisis, their scalability and efficiency makes them suitable for larger installations and public spaces where contaminated droplets present serious public health risks. These LEDs offer higher efficiency and tunable light output without the cost barrier that other solutions create.
In hospital and healthcare facilities specifically, QD LEDs offer the expanded range of customization across a vast color spectrum – a feature critical for decontamination – that specifically requires high-efficiency UV-C LEDs. While UV-C LEDs can decontaminate pathogens and other microorganisms, the ability to also create UV-A, UV-B, blue, green, red and other customized lighting outputs can enhance patient comfort in a way that is cost-effective and energy efficient.
QD LEDs are a serious contender in the UV-LED space, and the incorporation of new materials for QD development, including CHS, are helping scale these innovations and bring them to market faster. Creating next-generation LEDs for novel efforts, including COVID-19 decontamination requires LED manufacturers to reevaluate the materials they’re using and make the switch to options like CHS that provide cost effectiveness with efficacy, convenience and efficiency.
This article was written by Ramez Elgammal, VP of Technology, The Coretec Group (Ann Arbor, MI). For more information, visit here .
- N. Shirahata , Phys. Chem. Chem. Phys., 2011, 13, 7284-7294.