Why wouldn't Silicon cells last forever?

Art

Hi Guys,

Another silly question that came to mind when considering warranty periods.

I know panels will eventually become physically damaged by hail and other elements
that break down the sealants, etc.

Having said that, why wouldn't a poly/monocrystalline solar panel last forever (in theory)?
Does the cell itself break down somehow during use?
Do the conductive traces allong the surface of each cell break down?
Cheers, Art.

Solar Guppy

Re: Why wouldn't Silicon cells last forever?

In theory, yes, cells will last forever. There will be some degrading overtime but its measured in decades and frankly, cells haven't been around long enough to measure greater than 50 years.

I think its more likely the glass will have a great reduction in panel output due to abrasion and other environmental effects

Dave Angelini

Re: Why wouldn't Silicon cells last forever?

Solar Guppy wrote: »

In theory, yes, cells will last forever. There will be some degrading overtime but its measured in decades and frankly, cells haven't been around long enough to measure greater than 50 years.

I think its more likely the glass will have a great reduction in panel output due to abrasion and other environmental effects

A little short on the life. 51 years ago explorer 6 took panels into space. I had some of the old panels from Spectra Lab at an earth station for HP in the hills above the Silly Valley. They are still there I am told!

The Sun will eventually ruin anything left to it's exposure.

GreenPowerManiac

Re: Why wouldn't Silicon cells last forever?

Odds are that they'll last depending on their environment. Most broken cells are caused by something hitting it like a stone. Nowadays I see more solar panels controlling yellow flashing lights, warning signs and road related stuff. Any time where there is traffic, eventually a stone or sometimes vandalism strikes. It's pretty hard to destroy panels when they're on your rooftop. Newer panels are equipped with tempered glass, vacuum sealed and resist a lot of weather related abrasion. Nothing lasts forever, but I can guarantee most things will outlast ourselves. Other components break down faster/slower than others.

GreenerPower

Re: Why wouldn't Silicon cells last forever?

It is semi-conductor, the P-N doping of the cell would degrade over-time and reduce its efficiency if it survives other mishaps. The UV accelerates this degradation.
GP

westbranch

Re: Why wouldn't Silicon cells last forever?

makes one think, If it can last in space where there is no 'air' around to ameliorate the UV and other radiation effects, why cant it last on the surface of the earth? not including heat as space is damn cold so they should get better performance from the panels...

Thus, how long have those space bound panels lasted? any one know?

eric

icarus

Re: Why wouldn't Silicon cells last forever?

Define "forever".

Reminds me of the time I was assured that I had "life time tech support for a computer hardware piece. I said "life time huh?" He responded with, "absolutely, lifetime, two years!"

So if you define lifetime as two years, perhaps "forever" is ten!

Tony

GreenerPower

Re: Why wouldn't Silicon cells last forever?

westbranch wrote: »

makes one think, If it can last in space where there is no 'air' around to ameliorate the UV and other radiation effects, why cant it last on the surface of the earth? not including heat as space is damn cold so they should get better performance from the panels...

Thus, how long have those space bound panels lasted? any one know?

eric

It won't last in space either, its life is just long enough to be useful for the space trip.

http://www3.interscience.wiley.com/journal/78504453/abstract?CRETRY=1&SRETRY=0
http://www.mrs.org/s_mrs/sec_subscribe.asp?CID=5725&DID=166363&action=detail

GP

Dave Angelini

Re: Why wouldn't Silicon cells last forever?

It won't last in space either, its life is just long enough to be useful for the space trip.

http://www3.interscience.wiley.com/j...TRY=1&SRETRY=0
http://www.mrs.org/s_mrs/sec_subscri...&action=detail

GP
__________________
Be greener and may the force be with you

Not sure I get you GP but solar panels last in space quite a bit longer than the trip. The same ones are on the International Space Station. Can't go there on how long the spy satellites have been lasting. I know the Xray sat for NASA is still at full power now for 12 years because one of our sensors just failed.

The AGM's and Gel's that are inside are the expendables that keep getting changed out.

Solar Guppy

Re: Why wouldn't Silicon cells last forever?

I haven't found any real science behind PV cells and eventual operational life decline. The two links GP had are broken and the ones I did look at didn't have and data or peer reviewed publications

even the panels from the 80's they were cooked at 3X the sun are still going reported by people that owned them. The technology of the cells, purity, sealing and build quality and such has only improved

The solar I own for sure will out live me ... 80% in 25 years is the warranty, not the expected life of the panels

RCinFLA

Re: Why wouldn't Silicon cells last forever?

Lot of talk about PN junction degradation which accelerates with temperature and higher energy radiation like UV rays. Semiconductors are essentially stained glass that will evenually diffuse over time degrading PN junction.

The weakest link is probably the metalization and connections. The metalization will eventually corrode.

The daily heating and cooling will stress the tabbing connections and lead to a catastrophic connection failure. This last item will likely be the fate of most of the PV panels, if they don't get hit by a hail storm or baseball first.

BB.

Re: Why wouldn't Silicon cells last forever?

RCinFLA wrote: »

Lot of talk about PN junction degradation which accelerates with temperature and higher energy radiation like UV rays. Semiconductors are essentially stained glass that will eventually diffuse over time degrading PN junction...

For anyone not in the electronics industry--PN Junctions are the classic building blocks (plus silicon) for transistors and diodes used in solid state electronics.

Metalization is done after you have built your diodes, transistors, etc. on the surface of your silicon die, then metal is placed on the surface of the silicon and eteched in to "wires" to conduct the electricity (obviously, there are a lot of complex steps involved--but the ideas are pretty straight forward--"build the components" then connect them together with wires and insulation as needed; then package into an IC integrated circuit for easy assembly onto a PCB printed circuit board). Silicon solar cells are just a a subset of the Integrated Circuit with a glass top. ).

Some links to learn a bit more:

The basic Junction Transistor/Diode
Making a solar cell from Silicon
Semiconductor Doping Details
History of Integrated Circuits
Metalization

-Bill

GreenerPower

Re: Why wouldn't Silicon cells last forever?

Hmm ! Those links are not valid any more. Try google for "solar cell degradation" or "solar cell life", there are plenty of references. I haven't seen specific number in term of years of life but from these, sure it has some limit life (probably in the order of 50 years where the efficiency decays to a point not to be useful anymore), not indefinite.

From one of the google link wrote:

Modeling solar cell degradation in space: A comparison of the NRL displacement damage dose and the JPL equivalent fluence approaches.

Abstract
The method for predicting solar cell degradation in space radiation environments developed recently at the US Naval Research Laboratory (NRL) is compared in detail with the earlier method developed at the US Jet Propulsion Laboratory (JPL). Although both methods are similar, the key difference is that in the NRL approach, the energy dependence of the damage coefficients is determined from a calculation of the nonionizing energy loss (NIEL) and requires relatively few experimental measurements, whereas in the JPL method the damage coefficients have to be determined using an extensive set of experimental measurements. The end result of the NRL approach is a determination of a single characteristic degradation curve for a cell technology, which is measured against displacement damage dose rather than fluence. The end-of-life (EOL) cell performance for a particular mission can be read from the characteristic curve once the displacement damage dose for the mission has been determined. In the JPL method, the end result is a determination of the equivalent 1 MeV electron fluence, which would cause the same level of degradation as the actual space environment. The two approaches give similar results for GaAs/Ge solar cells, for which a large database exists. Because the NRL method requires far less experimental data than the JPL method, it is more readily applied to emerging cell technologies for which extensive radiation measurements are not available. The NRL approach is being incorporated into a code named SAVANT by researchers at NASA Glenn Research Center. The predictions of SAVANT are shown to agree closely with actual space data for GaAs/Ge and CuInSe2 cells flown on the Equator-S mission. Published in 2001 by John Wiley & Sons, Ltd.

This one is old data compared to today's solar cell technology but it those solar cells in the spacecrafts in the past probably won't last long.

IEEE Transactions on Electron Devices wrote:

Publication Date: Aug 1971
Volume: 18, Issue: 8
On page(s): 507- 511
ISSN: 0018-9383
Current Version Published: 2005-08-09
Abstract
The Lincoln Laboratory satellites LES-4 and LES-5 each carry solar cell experiments consisting of the following. VocMeasurement of 10 Ω . cm silicon cell with 30-mil cover slide. IscMeasurement of 10 Ω . cm silicon cell with 30-mil cover slide. IscMeasurement of 10 Ω . cm silicon cell with 6-mil cover slide. IscMeasurement of two CdTe thin-film cells (LES-4 only). IscMeasurement of two CdS thin-film cells (LES-5 only). LES-4 was orbited in December 1965 in a highly elliptical orbit with an 18 000-mi apogee and a 100-mi perigee; LES-5 was injected into a quasi-synchronous orbit in July 1967. In the LES-5 experiment, the Si cells exhibit an Iscdegradation of eight percent per year plus an initial short term degradation of four percent; Vocis relatively unaffected. The CdS cells have an Iscdegradation of 20 percent per year plus an initial degradation of five percent. In the LES-4 experiment, the Si cell with the 6-mil cover slide shows two rates of degradation, with the break point occurring at about 100 days; the cell with the 30-mil cover slide shows substantially less degradation. After 700 days, the short-circuit currents of these two cells are 60 percent and 78 percent of their initial AMO values. One CdTe cell has decayed to 38 percent of its initial AMO value after 700 days; the second sample gives anomalous results. In each experiment AM0 to AM1 short-circuit current ratios of approximately 1.09 were noted.

GP

RCinFLA

Re: Why wouldn't Silicon cells last forever?

Space rated solar cells and arrays are in a whole different class then typical PV panels used by us mere financial mortals.

Dave Angelini

Re: Why wouldn't Silicon cells last forever?

RCinFLA wrote: »

Space rated solar cells and arrays are in a whole different class then typical PV panels used by us mere financial mortals.

Yes the space rated cells are probably now just dropping below $1000 per watt!
Don't forget to add the development cost and the qual units which end up in backyards eking out those end of life watts! Think Stimulus!

Sheldon

Re: Why wouldn't Silicon cells last forever?

Space degradation is a bit different from terrestrial.

Semiconductors in Space fail primarily due to integrated radiation dose, which causes oxide charging, and field threshold inversion, which manifests as leakage.

A secondary failure mode is temperature cycling. If you take any crystalline material and cycle it over a wide temperature range, it fatigues, faster if it has mismatch of strain due to mounting and metallizations.

Space temperature extremes can be hundreds of degrees C. see Mil-PRF-19500 and associated documents for some idea of environmental constraints for Mil (and space) usage.

Here's a good reference on Mil semiconductors in general
http://biancaweb.cnes.fr/Standard_CNES_public/Presentation/MIL_system_overview%20.pdf

In terrestrial use, a more likely failure mode will be electromigration, when the cumulative "electron wind" from all that DC current moves the metal atoms down the conductor paths to the point of local high resistance. Its a trade off of long life (wide/thick) conductors vs high performance (narrow/thin).

Your particular panel's wiring, soft solder mounting, and cell to cell interconnect is probably more of an issue than the actual silicon as well as the gradual decay of the clarity of the panel cover.

azrc

Re: Why wouldn't Silicon cells last forever?

I'm guessing here because it has been a while, but I think the physics of cell degradation has to do with the diffusion of material at the junction. This junction (gap) must be very narrow, see diagram, for the device to operate.

Cells are made up of 2 parts silicon, one infused (doped) with extra electrons and the other with extra protons (holes). Not exactly, but it's highly mathematical. Anyway, incident light causes the (electrons or holes, I forget which) to cross this gap and you get current. Over time the + and - doped silicon dissolves (yes, solids dissolve into other solids, just like sugar into water..it just takes decades) into each other widening this gap and making it more difficult to push electrons through it.

RCinFLA

Re: Why wouldn't Silicon cells last forever?

Yes there is a degradation over time of the PN junction but it will be twenty or thirty years before you will notice the degradation. Plenty of time for one of the forementioned items to cause a catastrophic failure.

Just last night, on Discovery channel there was a segement on measuring age and erosion rate of Himalaya mountains by measuring the amount of 'sunburn' the rocks get from UV radiation.