Technical Article

Selecting and Implementing Photovoltaic Cells for Solar-Powered Devices

May 31, 2018 by Robert Keim

This article discusses details that will help you choose the right kind of solar cell and successfully incorporate it into your next project.

This article discusses details that will help you choose the right kind of solar cell and successfully incorporate it into your next project.

I recently wrote an article, The Circuit Designer’s Guide to Photovoltaic Cells in Solar-Powered Devices, that explains important solar-cell characteristics from the perspective of circuit design. In this article, I want to expand upon that information by discussing the different types of photovoltaic cells and exploring additional implementation details.


Evaluating a Solar Cell

When comparing the performance of different types of PV cells, the discussion usually revolves around two concepts: efficiency and spectral response.



The word “efficiency” seems to dominate most discussions of photovoltaic technology, and with good reason. It refers to the ability of a particular solar cell to convert electromagnetic energy into electrical energy. Whether you’re designing a tiny IoT device or an off-the-grid mansion, solar cells are not exactly cheap and not exactly convenient. Thus, we often want to get as much power as possible from whatever sunlight will be available to the cell.

However, this article is about solar power from the perspective of low-voltage electronic design, and in this context, efficiency is sometimes not very important. A small solar cell might provide more than enough power for the device and, consequently, other features (such as cost or bendability) will be more prominent.


Spectral Response

Any normal indoor or outdoor lighting environment will provide a solar cell with electromagnetic radiation covering a wide range of wavelengths. Thus, the amount of power generated by the cell will be influenced by which wavelengths it can convert into electrical energy. The following highly informative plot shows the spectral content of sunlight, the spectral content of fluorescent light, the spectral response of the human eye, the spectral response of crystalline silicon (c-Si), and the spectral response of amorphous silicon (a-Si). If you don’t know what crystalline silicon and amorphous silicon are, you soon will (if you keep reading the article).


Plot taken from this datasheet (PDF).

Choosing Your Solar Cell

The AAC textbook provides fairly extensive information on the various types of solar cells. My objective here is to cover this same topic in a way that is a bit more concise and approachable, and I’m going to limit the discussion to three general solar-cell categories: monocrystalline, polycrystalline, and amorphous.


Monocrystalline Solar Cells

As the name implies, a monocrystalline PV cell is formed from single-crystal semiconductor material (usually silicon). The silicon is “grown” in such a way as to result in a continuous crystalline structure, without edges or boundaries or impurities. Producing semiconductor material in this way is not cheap; also, the resulting “ingot” of silicon is circular, and since solar panels are rectangular, there is an additional cost associated with “squaring” the ingot. Thus, the higher cost is one of the disadvantages of monocrystalline cells.

It’s usually safe to assume that the presence of impurities or any other imperfections will negatively affect functionality, so it’s not surprising that monocrystalline cells are the leaders when it comes to performance. They offer higher efficiency, less efficiency degradation over time, and wider spectral response.


This plot of “external quantum efficiency” (EQE) conveys the spectral response of a monocrystalline solar cell manufactured by IXYS. Taken from this datasheet (PDF).


Polycrystalline Solar Cells

Polycrystalline silicon doesn’t require the same “growing” procedure; instead, molten silicon is placed in rectangular molds and allowed to cool. This doesn’t result in a single, continuous crystalline structure, and that’s why it’s called “polycrystalline.” Solar cells made from this material are less expensive, but the efficiency is lower and they are more susceptible to efficiency degradation over time. The spectral response is wide but not as wide as that of monocrystalline cells.

Amorphous Solar Cells

The word “crystal” evokes rigidity and fragility. The sound of the word “amorphous,” though not exactly common in daily conversation, gives an opposing impression: vagueness, shapelessness, susceptibility to alteration. And this is exactly what distinguishes an amorphous solar cell from a crystalline solar cell.

Amorphous cells are made from amorphous silicon. The atoms in this material have an irregular structure, in contrast to the extremely regular structure found in a crystalline substance. This increases the amount of interaction between photons and atoms, such that even extremely thin amorphous cells can generate enough power to be worthwhile.


A comparison of the allotropic forms of silicon. Image courtesy of Cdang [CC BY-SA 4.0]


There is an interesting mixture of pros and cons associated with amorphous solar cells.



The efficiency is low. Individual products can vary from the norm, and I make no effort to keep track of the constant research and development occurring in the world of solar power; nonetheless, I think it’s safe to say that in terms of efficiency amorphous cells are significantly worse than both monocrystalline and polycrystalline cells.


Spectral Response

Another significant limitation is that the spectral range is rather narrow:


Plot taken from this datasheet (PDF).


You can see here that the responsivity of amorphous cells is restricted more or less to visible light. This means that they’re not a good choice for outdoor applications because the energy in the non-visible portions of solar radiation won’t contribute to the power available from the cell. You might want to avoid them even for indoor applications if illumination is provided by incandescent bulbs (which generate large amounts of energy that is not within the visible spectrum) or by windows.

An interesting side effect of this narrow spectral range is that an amorphous cell makes a good ambient-light detector. As you can see in the plot, the spectral response is quite similar to that of the human eye, and it’s a much better match than the response of crystalline silicon. So if you put an amorphous solar cell in a room and monitor the open-circuit voltage, it will vary according to the amount of illumination that would be perceived by a person (rather than by a snake, for example, or any other creature that can see infrared).



To manufacture an amorphous solar cell you need both the semiconductor material and a substrate. If the substrate is glass, you can bet that the device will be just as rigid as a crystalline cell. However, stainless steel or a film material can be used as the substrate, and this means that amorphous cells offer an interesting new feature: flexibility. I don’t know how exactly this might be used in the context of small, solar-powered electronic devices, but surely there are creative people out there who have some clever ideas. Maybe a PV armband that supplies power for a wearable medical device?



I would say that in general, the cost of amorphous cells is moderate or on the low side, but it’s not quite as simple as that because the low efficiency means that more cells might be needed, and also because the choice of substrate affects the price.


We’ve discussed efficiency and spectral response in the context of solar power, and we looked at salient characteristics of monocrystalline, polycrystalline, and amorphous photovoltaic cells. If you have experience with the design of solar-powered electronic devices and have any thoughts on the pros and cons of different types of cells, feel free to leave a comment and let us know what you’ve learned.