SOLAR Frequently Asked Questions -- Section 1

[Picture of panel]

Solar panel circa 1973, Photo from NREL

1) Basic Questions (some maybe too basic)

1a) What are photovoltaics?

Photovoltaics are semiconductor devices that convert light into electrical energy. The basic unit is a photovoltaic "cell" which is a large slice of semiconductor material. Usually, multiple cells are joined into a single self-contained unit called a "panel" and these are usually joined (in an installation of any size) into an "array." A collection of arrays is called a "field."

1b) How does a photovoltaic cell work?

1b1) What is a semi-conductor?

Since a photovoltaic cell is a special type of semi-conductor diode, we need to do a little "semiconductor physics 101" to get to our working understanding.

I want to strongly thank Dr. Thomas Mowles for innumerable e-mail "talks" on these topics. Any correct understanding conveyed here comes from his contribution. Any erroneous content comes from me and me alone!

A semi-conductor is a material that has the electrical properties both of a conductor and an insulator. To understand this, we must talk about electrons and how they behave. All matter is made of atoms. Atoms have a positively charged nucleus and are surrounded by a cloud of negatively charged electrons. In many materials, these electrons are shared between pairs of atoms in "bonds" that make molecules. If you put enough energy into an electron, you can break it free of its bond. If a freed electron loses kinetic energy, the electron will "give back" the energy it got to break the bond and it will "fall" back into a bond (although not necessarily the same one it left, but it must be a bond lacking an electron).

In a conductor, there are some electrons that are not part of any bonds and thus are always free to move. They can occupy a wide variety of energy states and are free to move in any of those energy states.

In an insulator, there are no free electrons because only a very large amount of energy can lift an electron out of a bond. In some cases, electrons are either in a bond or stripped completely away, leaving the material with a static charge. Even when charged in this way, an electric current still will not flow through the material.

A semi-conductor lies somewhere between these two extremes. In a semi-conductor, the electrons are in bonds, but once you give the electron enough energy to get out of its bond, it can occupy a number of energy states and it is free to move.

In semi-conductors, when you put in energy and free an electron, you also leave a "hole." This hole is the second half of a pair, which physicists call an "EHP" for "Electron-Hole Pair." This requires some imagination (how can a hole be a particle?), but it is real. It happens. Think of it this way. You've freed the electron to move. You've also made a hole. The "hole" can also move. How? Well, the hole really wants an electron. A neighboring atom's electron might get "temporarily" freed (The phenomenon is called "quantum tunneling"). When that happens, the second electron must fall into the hole left by the first electron. In this way the hole has moved along a course totally different from that taken by the first freed electron. There is still just one free electron and one free hole (the second electron was never free, it just "fell" into the first hole). In this way, the electron and the hole wander randomly through the material until they meet up again and join together ("recombination" in the lingo).

While this is very interesting, it doesn't accomplish very much. One other thing has to be done to the material to make a useful device out of the semi-conductor phenomenon. That's the next question.

1b2) What is a diode (or what is a junction)?

When you have a uniform semi-conductor material you have a scenario like that described in the previous question where energy is put in and the electrons and holes wander randomly around.

Suppose you could introduce atoms of some other material into the semi-conductor material, perhaps a material with one extra valence (bonding) electron? This would make a material that is "hungry" for holes. Suppose further that you could also introduce atoms of yet another material that has a deficit (one fewer) of valence electrons compared to the semi-conductor material? We call the side with extra valence electrons "n-type" material (for negative) and the side with a deficit of valence electrons "p-type" (for positive). Note that these materials are NOT electrically charged! The electrons of these "other atoms" are matched with the same number of protons in the nucleus of the atom.

Something very special happens at the boundary between two such regions. Remember that semi conductors have some EHPs in them all the time (due to thermal energy mostly). These electrons and holes wander around the material randomly. Some holes will "fall into" the extra electrons in the n-type material. Some electrons will fall into the "holes" (missing valence electrons) in the p-type material.

Slowly, then, a tiny space charge is built up, the extra electrons that have "fallen" into the holes in the p-type material give it a tiny negative charge while the holes that have joined with the extra electrons in the n-type material give it a positive charge. Eventually, this process, called "diffusion," is stopped by the built up charge itself. Remember that like charges repel. The "pile-up" of electrons on one side will begin to repel any electrons that wander over. If an electron is energetic enough to get over anyway, it will lose a lot of energy in the process and will likely recombine, adding to the space charge.

So, the n-type material has a positive charge concentrated at the boundary with the p-type material. The p-type material has a negative charge concentrated at the boundary with the n-type material.

The line between them is called the "junction." (We are talking about structures on the scale of atoms here. Because of this, the junction is not a nice neat straight line. Instead it is a band of several atomic widths. When you look that closely, the junction is referred to as the "depletion region," which is more descriptive at the atomic level. For those of us who are quite a bit bigger than atoms, the term "junction" works just fine).

The following description is an analog, intended to make the idea of the junction clear. The actual physics of the process is tied to some rather nasty math which is rather too advanced to get into here (which is a cop out that means "It's beyond me."). I found this description of the idea seemed to clear things up for me, so I hope it will help you. My apologies to the physicists, chemists, and engineers who wince! (If you have a better description that remains comprehensible to the layman, PLEASE feel free to send it to me!)

Consider an electron wandering around the p-type material. It has some amount of kinetic energy (velocity). It approaches the junction. It begins to feel repulsion from the negative charge at the junction. That repulsion slows it down. It keeps moving toward the junction. If it has enough velocity to make it a little over halfway through the negative charges near the junction, it begins to feel the repulsion of the electrons it has already passed pushing it forward now and it begins to feel the attraction of the positive charges on the other side. Now the slowed electron begins to accelerate towards the n-type material. It starts to gain velocity. It makes it across the junction. If the electron did not have enough kinetic energy, it would have been reflected back by the charge, whereupon it would wander around until it recombines with a hole. Now consider an electron with the same amount of energy as our first electron, only this one is wandering towards the junction from the n-type material. It begins to feel the attraction of the positive charges and so accelerates towards the junction. But wait! At about halfway through the positive charges, it feels the positive charges it has passed pulling it back and it starts to feel the minus charges on the other side pushing it away. The electron can only cross the junction in one direction!

You can take the above paragraph and substitute "hole" for electron and "p-type" for "n-type" and things would work the same. In other words, the junction is an "energy barrier" that electrons can cross in one direction and holes can cross in the other.



        
           - +
         - - + +
       - - - + + +
     - - - - + + + +
   - - - - - + + + + +
   p-type       n-type

Such a device is called a "diode" and it has an important property. If you put wires into each side of the junction, you have a device where electricity can flow in one direction (electrons can move from the p-type to the n-type), but not in the other (electrons cannot move from the n-type to the p-type).

(Again, this is a simplification -- If you apply a strong enough voltage, you can force electrons and holes the "wrong way" over the junction. It's rather like salmon jumping UP a waterfall!)

This is the simplest semi-conductor device.

1b3) What is the photovoltaic effect?

In the above example, the diffusion of electrons and holes creates a junction. The application of an electromotive force (EMF) like a battery dumped a bunch of electrons into the device. If you dumped them in on the P-side, then they can easily cross the device and a current will flow. If you dump them in on the N-side, then they cannot easily cross and no current will flow. In any case, the current is due to the battery and not to the separation of electrons and holes by the junction. The EHPs that are culled to make the junction were created primarily by heat energy. There are other ways to make EHPs, however. If you have a material where the energy difference between a bound electron and a free conducting electron is about the energy of a photon of visible light, then you get the photovoltaic effect. Electrons and holes are created in such a material merely by shining light on the material.

(Actually, a critical energy input in both cases is heat energy -- room temperature is much warmer than absolute zero. Electrons are already at a fairly high energy due to heat energy. We don't usually think about that because we aren't used to thinking about room temperature as "hot," but even freezing winter cold is "hot" on the scale we're using here.)

1b4) What is a PV cell?

If we combine what we've seen in the answers to all the previous questions, we have enough to understand what a PV cell is and how it works. Imagine now a diode made of material where electrons and holes are made by shining light on the material. When a hole and an electron are separated by the junction, a difference in charge is created between the two regions. If you connect a wire between the P region and the N region, this charge difference will cause a current to flow through the wire until the charges are again balanced. If you keep shining light on the semi-conductor, a current will flow continually. The voltage of this current is a property of the materials used to make it, while the amperage (current) is a function of the amount of light falling on the device.

A PV cell is a large, flat diode. This is because the amount of energy picked up from sunlight is a function of area. The greater the surface area of the junction exposed to light, the greater the current obtained from the cell.

1b5) What else matters?

All of the above is the merest gloss on the topic. We did not discuss electronic structures, absorption spectra, band-gap energies, direct and indirect absorption, and a host of other topics all of relevance to PV and semi-conductor theory. Fortunately, there is a great deal of scientific and engineering literature on semi-conductors since this science and technology are the entire foundation of so-called "high-tech." There is also a great deal of material out there on semi-conductors aimed at the layman, and while very little of this is specific to PV (most of it aims at diodes, transistors, and integrated circuits), all of the theory applies to both.

Be aware that the above discussion is almost entirely based on the operation of mono and polycrystalline silicon photovoltaic cells. There are a number of other types, some of which have much more complex junctions and layers.

Cells are available based on the following technologies:

[Image of boule growth]

Growing hexagonal boules of pure crystal silicon, photo from NREL

monocrystalline silicon
This is the most common type. These are made from wafers of highly pure single crystal silicon. These have the advantage of high efficiency and stable output. They have the drawback of using a large amount of expensive pure silicon. This is somewhat offset by the fact that most of the current production is from boule ends and rejects from the IC industry.
polycrystalline silicon
This is also fairly common. It suffers from the same "thickness" problem as monocrystalline, but here instead of a single, perfectly grown crystal the wafers are cut from pressed ingots of silicon. This allows faster and lower energy production, but, of course there no free lunch. Cells made from polycrystalline silicon are slightly less efficient.
Both of the above technologies depend on wafers cut from silicon, so there is also a fairly substantial loss in the kerf (a term for the width of the saw cut, as you carpenters and wood shop students well know!).
amorphous silicon
Amorphous silicon is a "thin film" technology. This means that an amorphous silicon cell is only a tiny fraction of the thickness of a wafer cell. This means the cost of the materials in the cell is much lower. They can also be produced much faster. They also can capture more sunlight in the much lower thickness. Sounds ideal, doesn't it? Well, again, there's no free lunch. Because the material does not have a crystal lattice structure, recombination occurs much more readily in this material. With all the trade-offs, amorphous cell efficiency is quite a bit lower than either of the crystalline cell technologies. The low manufacturing costs help offset this.
There's more. Part of the process in achieving reasonable efficiency from these cells is a process that imbues hydrogen into the film to "heal" the dangling bonds that cause recombination. Unfortunately, many of these bonds get released when sunlight breaks the electron bonds between the silicon and the hydrogen. This lets the hydrogen escape to the air and the recombination center is back. What this means is that amorphous silicon cells degrade in sunlight. The efficiency loss can be as much as 50%.
Amorphous silicon, then, tends to be used most heavily in applications where this degradation is not a serious problem. Solar calculators and watches and so on tend to use these cells because they have very low power requirements.
When one talks about powering a home, the large surface area required to allow for degradation might very well offset any cost savings over the long haul. I have not seen enough data to do any sort of comprehensive analysis of this.
copper indium di-selenide (CIS)
This is a "bleeding edge" thin film material. I haven't seen cells of this type for sale from common "home power" sources. This is the best performer in the thin-film arena right now. The only problem I'm aware of with this technology is that the global production of indium is insufficient to supply CIS panel production on a scale that can meet a significant fraction of the world's energy needs.
cadmium tellurium (CdTe)
This is also a thin film technology with reasonable efficiencies. The chief problem with this material is that cadmium is a toxic heavy metal.

There are other cell technologies that are currently in laboratory scale production. Gallium arsenide cells exist. I know of some research into other semiconductor materials as well. Some of these technologies will never see mass production (due to rare or expensive materials), and some will never leave the lab (due to technical impediments). There are a lot of exciting possibilities, but as the opening of this FAQ states, we are concerned with what you can buy right now.

All of the thin films also have various highly toxic gases involved in their production. There are also toxic chemicals used in the production of the silicon wafer cells, but these are usually liquid solvents that pose far less of a danger than the gases used for thin films. I normally wouldn't mention this since a great many manufacturing processes used to create things we buy every day use toxic chemicals and solvents, but since PV lays claim to environmental "niceness," (and rightly so) the claim requires a higher standard of honesty. The (silicon) cells are nice and safe when the leave the plant, but some of the stuff in the plant is toxic and highly reactive. This is a side of the technology that deserves a close look.

Dr. Mowles has provided most of the list given in the Bibliography. He recommended (with only minor reservations) the Zweibel book as a good introduction to the field of PV.

1b6) There's more, isn't there?

There is a lot more to this topic. It took me a lot of hunting, asking, and re-asking just to learn what is found above. There are a host of details about the making of semiconductor materials, the construction of junctions, their assembly into cells, their construction into modules or panels, and much more that help one to understand the present potentials and limitations of both the technology and manufacture of PV devices. The Zweibel book (see the listings below in the FAQ) gives an excellent layman's summary of the present state of both.

1c) I know there's AC and DC, but what is the difference and why do I care?

Electricity is the name we give to the force that appears when matter has an imbalance of charge. An atom is made up of a nucleus composed of positively charged protons and neutral neutrons. Normally, the positive charge of the nucleus is offset by a cloud of negatively charged electrons near the nucleus. When you shuffle your shoes over a carpet on a dry day, you pick up electrons from the carpet. That means that you have an excess of electrons and the carpet has a deficit. This charge is called "static" because it is not moving. You are simply negatively charged. When you bring your fingertip near a light switch, or better, a younger sibling, who is not negatively charged (relative to you), a spark will jump the gap between you and the surprised younger sibling. For a very brief moment, there is a "current" flowing from your body to the sibling until the charge is equalized.

We don't use static electricity very often because it is really not useful for accomplishing work. Instead we use a continuous flow of charge called an "electric current."

An electric current requires a "circuit." A circuit is a path between a positive charge and a negative charge. A current can only flow so long as there is a closed path from the one to the other. If you break the path (which is what a switch does), the current will stop.

You will often see the terms AC and DC when talking about electricity. AC means "alternating current" and DC means "direct current." You get DC from a battery. A battery has a positive pole (or anode) and a negative pole (or cathode). If you connect a wire between them, a current will flow through the wire continuously at one voltage until the battery is exhausted. (Not an experiment I would recommend, since the wire can get very hot!) AC is the type of current that we get from our wall receptacles. In AC current, the voltage is not constant. Rather it begins at zero volts, rises to a positive peak, falls away to zero again, then reverses falling to a negative peak, then rises again to zero. The shape is a sine wave and it is a result of producing electricity by spinning magnets past coils (or coils past magnets). In the USA, this cycle is 60 Hz. AC is used primarily because it allows the use of "transformers" to change voltage.

In the context of solar energy you care because solar cells are low voltage DC producers and most of your appliances consume "mid-voltage" AC. That means you've got to convert either your electricity (using an inverter) or your appliances. There are tradeoffs to both routes.

1d) What's all this stuff about volts/amps/ohms/watts. (volts X amps = watts, etc.).

There are a number of units used to describe electricity.

VOLT
Named for Alessandro Volta, this is the unit of electrical force. In other words, it is a measure of the charge difference between the two ends of a circuit or between two charged objects. It is therefore a measure of energy of each electron. The higher the voltage, the more energetic the electron. Typically represented by the letter "E"
AMPERE
This is the unit of electric "current." It represents the number of electrons passing a point in a circuit each second. Typically represented by the letter "I".
WATT
The product of force and current. The watt is a unit of power. One watt is equal to one volt at one ampere. Typically represented by the letter "P". Power relates to voltage and current in the following relationship:
P=EI
OHM
The unit of electrical resistance. Typically represented by the letter "R". Voltage, current, and resistance are related in a simple and direct equation known as "Ohm's Law:"
E=IR
The ohm is also the unit of impedance, which becomes important in AC circuits. We won't go into impedance here.
WATT HOUR / KILOWATT HOUR
The product of power and time. If you use a certain amount of power for a given amount of time, the product of those is the amount of ENERGY used. If you use half as much power for twice as long, you've used the same amount of energy. Your electric bill almost certainly charges you for the number of "kilowatt hours" you use. A kilowatt hour is the amount of energy used to produce 1000 watts of power for one hour. So, a kilowatt hour (abbreviated kWh) would be enough energy to run ten 100 watt light bulbs for one hour or to run one 100 watt light bulb for 10 hours.

There are other units commonly used in electrical work including:

FARAD Unit of capacitance
HENRY Unit of inductance

And some not so often used (by electricians, anyway):

COULOMB Unit of charge (an ampere is 1 coulomb per second)
OERSTED
GILBERT
MAXWELL
WEBER All related to magnetic force, inductance, flux, etc. (I confess that in my ham radio work, I hardly ever use these and I don't know which one is exactly what!)

1d1) What units describe solar PV products?

(Much of the information in this section was gleaned from Steven Strong's "The Solar Electric House." See the "Books" section for information)

Well, when you are looking for a PV panel (integrated and packaged group of cells), you are primarily interested in the total power you will get from them. Remember that the voltage of a cell is fixed, so all of the electrons you get will have the same energy. What can vary, then, is the current (in amperes) which is how many of those electrons of a fixed energy you can get per unit time. The product of voltage (energy of an electron) and amperage (current, or number of electrons per second) is the power.

[Image of NREL Test Equipment] Unfortunately, the output of a given PV device is not constant. Obviously, current varies with the amount of light falling on it. If that were the only factor there would be no problem. Unfortunately, the voltage varies with the temperature of the device. PV cells produce a higher voltage when cold.

{temperature/voltage explanation is being re-written and will go right here!}

So, when a PV manufacturer says "The Solar Sun-Smasher Six thousand Solar Cell (tm) puts out 350 watts!" how do you know what this means? Is this 350 watts when the cell is at -40 degrees C? The change is not small enough to write off either. The variation can be as much a 40% from lowest operating temperature to highest.

Fortunately, the DOE (U.S. Department of Energy) has defined measurement standards that allow all PV devices to be measured on an equal basis. The most important of the standards is the Wp (Watts peak), which is the wattage of the device at noon under direct sun on a clear cold day. That's still vague (define "cold!"). Okay, the DOE Standard Test Conditions (STC) are:

Cell temperature (NOT ambient air temp!):  25 degrees C
Air mass (see below):                      1.5
Irradiation:                               1,000 watts/sq. meter

Air mass is a measure of the interference caused by the atmosphere. In space, the AM would be zero. Clear air at the equator at noon on the equinox at sea level would be 1.0. The AM is actually the secant of the sun's angle from the zenith (straight overhead). It is a measure of how much atmosphere the sunlight used has passed through.

These standard test conditions basically don't exist anywhere on the earth's surface. For whatever reason, somewhat less energy will be obtained from your PV device. Most places on earth get somewhat less solar radiation than the standard, but the biggest factor is that most cells get hotter than 25 degrees C under most conditions. Still, by using this standard, the relative performance of different devices can be reliably assessed.

Also, watch for SOC and NOCT. These stand for Standard Operating Conditions and Nominal Operating Cell Temperature. These are two more standards that are a little more realistic, but are also dependent on the system used.

SOC is defined as:

Insolation (actual solar energy reaching device):  800 watts/sq. meter
Ambient air temperature:                           20 C (68 F)
Wind velocity:                                     1 meter/sec

Further, the device must be oriented towards solar noon (tilted and aimed so that the sun is exactly perpendicular to the plane of the device) and the device must be "open" (not electrically connected so that there is no current flowing through the device).

NOCT (Nominal Operating Cell Temperature) is defined as the temperature of a cell in the device when it is operated in the above conditions.

These give you a more believable performance figure, but since these are actually dependent on parts of the configuration other than the device itself, these figures are often not available.

But wait! There's more!

There is a useful graph you can draw for any DC electricity source. This graph plots current on one axis and voltage on the other as you vary the load across the device from a short circuit (zero resistance) to an open circuit (infinite resistance).

For the extremes we use the following names for the values: I sc The maximum current the device can supply. The "I" means current, and the "sc" means short-circuit. V oc The maximum voltage the device can supply. The "V" means voltage and the "oc" means open circuit.

The graph you draw for a PV cell changes shape as incident light changes and the temperature of the device changes. You must know the conditions under which the graph was produced. This will usually be STC or SOC.

There's a few more values we care about from the I-V curve (as it's called). There is a point on the curve "near the knee," as Steve Strong puts it, where the product of current and voltage is maximum. In other words, this is where the PV reaches maximum power, or wattage. This point is the "maximum," abbreviated as "m" and applied thus:

P m Maximum power in watts. This is the wattage the device produces when load is at optimum. V m Voltage at maximum power. (NOT maximum voltage!) I m Current at maximum power. (NOT maximum current!)

These are the common measures you will see.

1e) What is "RMS" voltage?

In DC where the voltage is constant there is no doubt or uncertainty. 24 volts is 24 volts! In AC, though, things are a little more muddled. Most of the time, when an AC voltage is given, that voltage is an "RMS" voltage. RMS stands for Root Mean Squares and is a way to calculate the "effective" voltage of an AC current. Consider figure 1e-1:

------XXXX------------------------------------ +170 volts
    XX    XX
---X--------X--------------------------------- +120 volts (RMS)
  X          X
 X            X
 X            X
X              X
X--------------X--------------X---------------  0 volts
	       X              X
	       X              X
		X            X
		X            X
		 X          X
------------------X--------X------------------ -120 volts (RMS)
		   XX    XX
---------------------XXXX--------------------- -170 volts

	    figure 1e-1

(Forgive the freehand waveform!). Your house current might very well have a peak voltage of 170 volts and still we refer to it as 120 volts AC. Why? The answer is that because the voltage drops away to zero and builds back up to peak, the voltage that you get over time is more like DC at 120 volts. You can't use the mean voltage. The mean voltage of an AC waveform is zero volts! That's why we use RMS volts. This figure is often more useful than the peak voltage, since voltage is only at peak for a very small period of time. Whenever you see "x volts AC" the "X" is more than likely the RMS voltage unless the word "peak" is present.

The calculation method is implied by the name. If you take the square of each instantaneous voltage, calculate the arithmetic mean of those squared voltages, then the square root of that mean is the RMS voltage.

Go to top of FAQ

Go to previous FAQ section

Go to next FAQ section

Go to the author's home page

Send mail to author

$Id: sf1.html,v 1.1.1.1 2008-01-25 16:36:20 mschwarz Exp $