It’s been quite a while since I attempted a useful, content-filled post, but as a new grad student I’m learning a lot…and I’m completely inundated with data.  What to do with all the I-V curves I generated while being trained on device fabrication?  Wouldn’t want to publish them…P3HT/PCBM is a very thoroughly studied system, so more than enough people have already done that.  I was only using that system to confirm that I could make fairly consistently functional solar cells.  So: been there, done that, have now moved on to more interesting (read: unstudied, unoptimized and commercially unavailable) systems.

So: carbon-based photovoltaic materials: hot field now, right?  There’s plenty of room for synthetic chemists to get in on that action, right?[1]  But there are numbers involved with that research.  Big, scary numbers.[2]  What are they?  What should you do with them?  (“Hide from them until they go away” is not the correct answer.)

The first one that’s important is the V_oc, or open-circuit potential.  This is the potential difference between the two electrodes (usually ITO for the anode and often aluminum for the cathode) when the current is zero.  For most organics, this is between 0.5 and 1 V.  The V_oc depends largely on the energy levels of the donor and acceptor materials–commonly simplified: HOMO of donor, LUMO of acceptor (though there are a lot of things this doesn’t take into account).  However, actually seeing a reasonable V_oc out of your devices requires that the film morphology does not suck.[3]

Then there’s J_sc, or I_sc as some people abbreviate it–the short-circuit current density.  This is the current at zero bias, divided by the area of the device.  This is the tricky part!  If you don’t accurately measure the device area, J_sc is wrong…and so is your calculated efficiency.   J_sc around 10 mA/cm² is nothing to cry about…of course, the higher the better.

Here are a couple of I-V curves for your enjoyment:[4]

If your I-V curve looks like a straight line, your device is shorted–you probably scratched the film at some point.  This means you fail at life.  Resistors have straight-line I-V curves…solar cells should not.

This is more like it!  You can see J_sc and V_oc nicely circled for your viewing pleasure…but what’s that yellow rectangle?  The fill factor is also very important.  In order to calculate it, you need to know the maximum power point of your solar cell.  Despair not, mathophobes…just multiply the J column and the V column of whatever spreadsheet you have the data in, and see where it maxes out.  (Current times voltage is power.  Not real ultimate ninja power, but power nonetheless.)  Once you have that, just divide it by V_oc multiplied by J_sc.  The higher, the better–a fill factor above 0.6 results in much rejoicing.  (A fill factor above 1 means you fail at math.)

The final and perhaps most important number you get out of this data is the power conversion efficiency, η.  There is an equation for calculating this that’s pretty much everywhere, so finding and using it is left as an exercise to the reader.[5]  The highest reported efficiency for an organic bulk heterojunction solar cell is a little over 6%.  (Under 1% is not terribly impressive; over 3% and you should celebrate.)

[1] by which I mean there’s plenty of room to make stuff for device people.  We either have no idea how to do anything synthetic, or have no time.  Keep in mind: it must be super clean and in sufficient quantity (if you can’t scrape together at least 100 mg, give up).  You must also be patient, because optimization can take a long freaking time.

[2] big = >8 (perhaps >18 for organometallicists)

[3] on the long list of things I have learned the hard way lately.

[4] Igor FTW!  I heart Igor.  Origin can DIAF for all I care.

[5] I am too lazy to type it out. 

This entry was posted on Sunday, August 30th, 2009 at 10:34 pm and is filed under FAIL, ninja, solar cells, unsolicited advice. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.