Direct growth technique could produce low-cost, high-efficiency graphene-on-silicon Schottky junction solar cells.

2018 was a record year for renewable energy, and predictions point to another big year in 2019. But with a rising demand for solar energy comes wider questions on the sustainability of our current photovoltaic devices. The search for alternative materials, innovative architectures and new fabrication techniques has been ongoing in research labs for decades. Dye solar cells can be made using a low-energy manufacturing process, but have a limited maximum efficiency. For solid state perovskite solar cells, efficiencies above 22 % have been achieved, but they are not yet stable enough to replace conventional silicon solar cells. Silicon also still dominates the commercial market, so, it’s perhaps unsurprising that many researchers are looking for ways to redesign the traditional p–n junction-based cell.

One approach is to combine Schottky junction solar cells, which can be fabricated at relatively low temperatures, with graphene; a material with a unique combination of properties, including tuneable work function, flexibility, mechanical strength and optical transparency. In a new Carbon paper [DOI: 10.1016/j.carbon.2019.03.079] a team from Sejong University, Korea, report on their latest development – a low cost fabrication technique, which allows different thicknesses of graphene to be grown directly onto bare silicon.

The team started with a standard phosphorous doped n-type silicon wafer, which was cleaned, etched and annealed to remove oxides and any impurities. Graphene was grown on the surface via a continuous flow of hydrogen and methane in a plasma-enhanced CVD chamber, and the thickness of the layer was controlled via the growth time. Samples obtained after 2.5, 3.5 and 4.5 hours had a graphene layer measuring ~2 nm, ~4 nm and ~8 nm, respectively, as determined by atomic force microscopy, and in all cases, graphene covered an area of 0.3cm-2. Raman spectroscopy was used to confirm the presence and p-type nature of the graphene, and energy dispersive X-ray (EDX) analysis mapped the carbon as the layers grew thicker.

Contacts of copper (on the graphene side) and aluminium (on the silicon side) allowed the device to be electrically characterised, and the researchers found that there was an optimal thickness of the graphene, ~4 nm. Those samples displayed the highest average power conversion efficiency, 5.51 %. Increasing the thickness beyond that point reduced the layer’s transparency, negatively impacting the cell’s performance.

The efficiency of the cell was further increased to 9.18 % by adding and doping a polymer (PMMA) layer. The authors attribute this improvement to a reduction in the transfer of electrons from the n-type silicon to the p-type graphene, reducing the leakage current. This addition also seemed to make the cell more stable over time. The authors say that their direct growth technique is “compatible for industrial-level applications,” and suggest that it offers a simpler and more reliable alternative to manually transferring CVD-grown graphene onto a silicon surface.


Malik Abdul Rehman, Sanjib Baran Roy, Imtisal Akhtar, Muhammad Fahad Bhopal, Woosuk Choi, Ghazanfar Nazir, Muhammad Farooq Khan, Sunil Kumar, Jonghwa Eom, Seung-Hyun Chun, Yongho Seo . “Thickness-dependent efficiency of directly grown graphene based solar cells”, Carbon 148(2019) 187-195. DOI: 10.1016/j.carbon.2019.03.079