Are solar cells operated under any kind of bias or not?

[nashsth]

Hello all,

I am having a very hard time understanding the operating principles of solar cells. I'm mostly confused about what external biasing conditions a solar cell operates under. I have tried googling this question and I get different answers (see the links below). For the TL;DR version, skip to the last two paragraphs.

I understand that they are similar to p-n junctions when there's no illumination. So, the I-V curves of a solar cell will be identical to the I-V curves of a p-n junction when there's no illumination. I will first attempt to explain my understanding of p-n junctions:

Now, for a p-n junction, if you apply a negative bias, then you will have a reverse saturation current due to the drift of minority carriers (the majority carriers diffuse and the strong electric field present in the depletion region opposes the diffusion of majority carriers). If you apply a positive (forward) bias, then the majority carriers will have a much easier time diffusing across the depletion region due to the weakened electric field (since the built-in electric field and the electric field due to the external forward bias are in opposite directions and will partially cancel each other out). The stronger the forward bias, the larger the majority carrier diffusion. In fact, the current is exponentially dependent on the forward bias voltage, as seen in the Shockley Diode Equation.

From what I understand, the V in the I-V curve is the external voltage applied to the p-n junction so that if V < 0, then we are applying an external reverse bias; if V = 0 then we are not applying any bias; and if V > 0 then we are applying an external forward bias. The I in the I-V curve is the resulting current flowing through the p-n junction under the given bias (reverse, zero or forward). If I is negative (as is the case in reverse bias) then this means that minority holes are flowing from n-type semiconductor to p-type semiconductor, and that minority electrons are flowing from p-type semiconductor to n-type semiconductor (so that the conventional current is from n-type to p-type). If I is positive (as in forward bias), then this means that the majority holes are flowing from p-type semiconductor to n-type semiconductor and that the majority electrons are flowing from n-type to p-type (so that the conventional current is from p-type to n-type).

When there's no illumination, the solar cell will follow the exact same logic as described above (or so I hear) and so the solar cell will have the same I-V curve as the p-n junction in this condition.

Now, when there IS illumination, the I-V curves for a solar cell is shifted down from the I-V curve of the solar cell under no illumination. To me, this suggests that the V in this I-V curve is still the external bias applied to the solar cell. So, this shifted I-V curve is saying that when there's illumination AND the solar cell is reverse biased, the reverse saturation current that flows is higher than for the non-illuminated, reverse biased solar cell. This is because under illumination, there is optical generation going on in the solar cell IN ADDITION to the thermal generation of carriers. Under a reverse bias, the minority electrons go from p to n-type and minority holes go from n to p-type. But, because of optical generation, there's more of these carriers flowing and so the reverse current is higher.

Now another confusion of mine is why under a weak forward bias, the I-V curve for an illuminated solar cell still shows that the current is negative (i.e. the 4th quadrant of the I-V curve for an illuminated solar cell). I have a feeling that it has to do with the optically generated carriers but I can't seem to make a connection.

At any rate, the I-V curve for an illuminated solar cell is shifted down and I have read that solar cells operate in the 4th quadrant of the I-V curve. The 4th quadrant has V > 0 and I < 0. From this, I would assume that solar cells operate under a forward bias. However, I would also think that they operate under no bias because I thought the entire point of a solar cell is to produce electricity from solar radiation. I thought that they wouldn't need "any other help" other than solar radiation because that would defeat the purpose of the solar cell. Since there is an external voltage source to bias the cell, then would this mean that the solar cell uses some of its generated energy to power up that voltage source? Finally, I would also think that reverse biasing a solar cell would be beneficial because reverse biasing causes the depletion region to increase so that more area is available for the generation of carriers which would lead to a higher current. So my main confusion is, what biasing conditions are solar cells operated under?

Links:
https://ecee.colorado.edu/~bart/book/solar.htm
(States that solar cells are operated under a forward bias)

https://courses.engr.illinois.edu/ece110/sp2018/content/courseNotes/files/?photodiodes
(Figure 7 in the above link shows no external voltage source to bias the solar cell, and so I assumed that there is zero bias used to power up the cell)

https://www.quora.com/Are-solar-cells-operated-under-zero-bias-or-forward-bias-see-comment#
(Steve Noskowicz's answer states 0 bias used)

https://www.quora.com/Why-is-photo-diode-reverse-bias-while-LED-is-forward-bias
(Chandan Mishra's answer states that reverse bias is used)

https://www.reddit.com/r/Engineerin...elp_clarifying_the_iv_curves_of_a_solar_cell/
(This answer suggests forward bias is used)

 

[berkeman]

For the TL;DR version, skip to the last two paragraphs.

I'm no expert in solar cells, but I do know that it's important to operate the solar cells at their Maximum Power Point. As shown in your first link, that will be when there is almost a forward bias on the cell, but not quite. If you let it go too close to forward conduction, you lose power. The "bias" is not really a bias, it is just a voltage that is caused by the photocurrent creating a voltage drop across the power converter circuit (a resistor or a switching power supply input).

https://ecee.colorado.edu/~bart/book/solar.gif

https://ecee.colorado.edu/~bart/book/solar1.gif

 

[ngonyama]

To study the properties of your system you might want to apply a bias, but means that you have to supply it.

The idea of a solar cell in real operation is to obtain energy from it, not supply to it. So there you generally don't want to have to apply a bias.

 

[CWatters]

+1

Typically you bias a photo diode but not a solar cell. For the reason ngonyama and berkeman state.

Photo diodes are typically used to detect a light signal and turn it into a voltage signal. Biasing moves the operating point to where the light vs voltage curve is steeper. That makes it more sensitive. The photo diode consumes power from the bias network.

You don't bias a solar cell in the same way but you may want to control the load on it for the reason berkeman gives. In AC circuits transformers are used to "match" a load to a supply (example 12V load to 110V supply). Solar panels are DC and their voltage and current varies with both load and illumination. So a more complicated matching device is needed.

In something like a solar powered calculator it might not be essential to capture every last bit of power so you could just use a linear regulator or even a resistor to match the output of the solar cell to the battery and waste any excess power as heat. In other applications it's more important and you use a controller that behaves a bit like a transformer but for DC and with variable turns ratio. It automatically adjusts itself to keep the output of the panels at the max possible power point for the amount of light currently arriving.