The spectral energy distribution of solar light has a maxima in the visible portion.This is at around 1.5 eV and hence the semiconductor having band gap near 1.5 eV is preferred for solar cells. Since maxima in intensity occurs in the visible portion of sun light, only this portion is  useful. The other portion of spectrum is not useful. It produces heat if absorbed or attempts are made to get it reflected. This is one of the reasons of low efficiency of solar cells as only 46 percent is visible portion.

The efficiency depends on the solar irradiation spectrum and the absorption spectrum of the photovoltaic absorber. For an inorganic semiconductor absorber all the photons with energy above the energy gap are able to pump an electron from the valence band to the conduction band. So with a low energy gap you have high current. But the higher the energy gap, the higher the theoretical photovoltage. So the power output is a trade off of high current (small gap) and high voltage (large gap). For semiconductors and AM1.5 irradiation (solar spectrum at 1.5 air mass i.e. in Europe), the optimum energy gap is in the range 1.1 to 1.7 eV with a theoretical efficiency above 25%.

More than 90% of solar light falls in visible and IR region and both are getting almost equal proportion of light. So visible and IR region is more useful.

Dear Dr Gopal

Light having energy less than band gap of semiconductor does not produce electron hole pair and hence can not be useful for photovoltage.If the used semiconductor has band gap in the energy range of visible light, infrared portion will not be useful for photovoltaic application.

I think the most useful is still in the visible. I think so because if your PV material had its bandgap absorbing in the UV, current will be low and if infrared, voltage will be low. As said above about the ideal solar cell calculated to have about 1.4eV bandgap which clearly falls in the visible.  However, I think there are people working on how to effectively harness these portions

I am perfectly in line with those of the others. I'd like just to add that is the material employed that limit the part of the solar spectrum useful for PV. In principle the radiation down in the near IR can be exploited  keeping in mind that the muximum photon flux in the visible range.

Multijunction solar cells (such as used in CPV) try to use as much of the spectrum as possible, including the NIR up to 2 µm or more.

For high latitudes, and therefore large air masses most of the time, the actual spectrum is red shifted and richer in NIR. PV material with a good response in the NIR would be best for those locations.

Chris, you said: "For high latitudes, and therefore large air masses most of the time, the actual spectrum is red shifted and richer in NIR. PV material with a good response in the NIR would be best for those locations."

This is an useful viewpoint. I'd like to mention that this rule is opposite to the thickness of the troposphere (that contains approximately 80% of the atmosphere's mass and 99% of its water vapour and aerosols). The average depth of the troposphere is approximately 17 km in the middle latitudes. It is deeper in the tropics, up to 20 km, and shallower near the polar regions, approximately 7 km in winter. 

I wonder to what extent this affects the spectrum?

The air mass, m, I talk about is just the photon pathlength relative to a zenith sun. So m=1 for a zenith sun and m=2 for a sun altitude of 30°. The rating of solar cells considers m=1.5. This is not directly related to the altitude of the tropopause.

830nm or 1.5ev

For a standard silicon solar cell, the maximum quantum efficiency is from 800 to 850nm. Depending on the technology, we can make the best use of more parts of the spectrum efficiently for example by using tandem structures

To clarify things, here's a figure I made comparing the spectral response of various PV materials...

Dear Dhanaji Mahadeo Kale ,

welcome!

Single junction solar cell from one absorber material absorbs radiation with Eph>= Eg. But the losses increases with the increase of Eph because of the thermalization loss and the dead zone loss at UV. The theoretical efficeincy amounts to about 33 percent.

The absorption of light will be enhanced for multijunction cells with multiple absorbing materials and the conversion efficacy will appreciably increasing.

Form more information please refer to the book chapter: Chapter Solar cells and arrays: Principles, analysis and design

Best wishes

The solar spectrum that reaches the Earth's surface covers a wide spectrum of wavelengths, from 290 nanometers (ultraviolet, UV) to 3,790 nm (infrared, IR). Photons from the entire electromagnetic spectrum reach photovoltaic panels, but not all photons are absorbed by the materials that make up these panels. Achieving that they are would dramatically increase their efficiency. And that's what Ephocell is looking for.

In an ideal situation, the spectral response (number of photons per unit wavelength) of photovoltaic materials should perfectly match the sunlight, to convert the maximum number of photons into electricity. However, the range of wavelengths where photovoltaic materials generally absorb is between 400 and 1,200 nm (visible range) depending on the photovoltaic technology in question.

"Ephocell intends to develop an easy-to-implement device that allows modulating the wavelengths of the solar spectrum to maximize the absorption of photons by photovoltaic panels of different technologies (a-Si: H, InGaP, DSSC, organic polymer cells, etc). For this, two known processes are combined: Down-Shifting (DS), which allows the transformation of high-energy photons to low energy, and Up-Conversion (UC), which performs the reverse process. Therefore, the Ephocell device will use the DS process to convert photons from UV to visible while and the UC to transform photons from IR to visible ”, they explain at the Spanish technology center Leitat, which coordinates the project.

The consortium is made up of high-level research centers and companies involved in the development of new materials: Max Planck Institute for Polymer Research (Germany), Sofia University (Bulgaria), ICTP-CNR (Italy), UPC (Spain), DIT (Ireland), Cidete Ingenieros (Spain), Daren Labs (Israel), MPBata (Spain) and Leitat himself. “During the first stages of the project, great progress has been made: such as the development of new Up-Conversion systems that allow modulating the wavelength of photons from the near-visible IR, and the coupling of these Up-Conversion systems. with their Down-Shifting counterparts that allow efficient energy transformation. In parallel, different materials that house both the Down-Shifting and Up-Conversion systems have been studied, also studying the stability of the same fronts under different environmental conditions ”.

The UC process has been observed directly with the sun, in the same atmospheric conditions in which the solar panels are found. These advances will be presented on October 2 and 3 at the technology center, located in the Terrassa technology park, together with the advances of another eight high-level European projects that study different ways to improve efficiency in photovoltaic technology based mainly in nanotechnology. The image that illustrates this information shows the Up-Conversion (UC) process in a laboratory-scale system that can be coupled to photovoltaic cells. In this case, the system transforms "green" light of low energy into "blue" of higher energy.

Dear Dhanaji Mahadeo Kale

From my practical experience, the best spectrum is the visible spectrum