Nepal Human Hair Solar Panel Hoax

"My PhD is in Organic Materials for Photovoltaics... I am horrified that [news agencies] have published this article verbatim, without even questioning the basic science behind this 'invention'... Melanin is really bad at charge separation, a basic requirement for solar cells." ~~Caesar_Dia

Introduction


An article appeared in the Daily Mail claiming that an 18-year-old Nepalese studentMilan Karki and colleagues invented a solar panel capable of generating 9VDC/18W with human hair as the main component. However, test results and analysis of quotes from the inventor show this invention to be a hoax. The idea that human hair can be used as a solar cell has long been debunked but continues to be circulated on so-called "free energy" or "zero point energy" blogs and eco blogs. Claims on these blogs are based on confusion about the nature of static electricity and misinterpretation of research on melanin's electro-optical and semiconductor properties. The solar panel design is based on the students' misinterpretation of preliminary unpublished results they found on the internet. The students make the false claim that hair is a conductor and generates a voltage when exposed to light.

What has really happened? The Trinity College students made a cuprous oxide solar cell that is commonly used for demonstrations in school laboratories. This cell does indeed generate a small voltage and current, but is completely impractical to scale to a larger size, and the cell would work just fine without human hair. A reporter with no training in science or engineering witnessed the cuprous oxide cell working but didn't realize it wasn't generating the claimed power levels. The story was broken without being reviewed by experts, however now most news outlets are retracting the story based on feedback from scientists and engineers.
 
In the remainder of this article, we examine this story in detail, starting with how the story spread and why it spread.  Then, we discuss the technical details of this invention and how it measures up to current science.  Then we elaborate reasons for calling the invention a "hoax" based on information published so far and quotes from the inventor himself.  The article ends with remarks directed at the inventor and some suggestions for more fruitful research areas.  The article provides an alternative and perhaps more feasible solution in the form of a small thermoelectric generator that can be powered either by the sun or captured heat from cooking coals.
  

How the Story Spread 

The first news article I can find in English is the Rebublica article from July 2009, though the September 2009 Daily Mail article seems to have sparked interest in the West.  The Mail article has pictures of the smiling boy and his solar panel, proclaiming him as the "next Edison".  One article states, "A new type of solar panel using human hair could provide the world with cheap, green electricity, believes its teenage inventor. Milan Karki, 18, who comes from a village in rural Nepal, believes he has found the solution to the developing world's energy needs. The young inventor says hair is easy to use as a conductor in solar panels and could revolutionise renewable energy."
 
The article was immediately picked up and circulated world-wide on ecology and technology blogs. It is rather sad that nobody bothered to check the facts. Even quotes from the boy indicate a problem from the outset, as he seems to claim a relationship between static electricity and the photovoltaic effect -- two completely unrelated phenomena. Anyone who completed a high school physics course ought to have heard alarm bells on reading, "He was originally inspired after reading a book by physicist Stephen Hawking, which discussed ways of creating static energy from hair." and "I realised that melanin was one of the factors in conversion of energy."
 
The young man claims he has sent several units out for evaluation which, on the face of it, lends credibility to his claim: "I'm trying to produce commercially and distribute to the districts. We've already sent a couple out to the districts to test for feasibility." On the other hand, this means that he has built prototypes capable of producing 9V/18W. Based on the analysis below, this seems highly unlikely and, unfortunately, seems to indicate this is a deliberate hoax. As discussed below, the claimed output of this device does not agree with the published properties of photoelectric organic dyes, making it likely that a conventional solar cell is concealed inside the panel.
 
Furthermore, the article states, "Half a kilo of hair can be bought for only 16p in Nepal and lasts a few months, whereas a pack of batteries would cost 50p and last a few nights. People can replace the hair easily themselves, says Milan, meaning his solar panels need little servicing" and "The young inventor says that human hair due to the presence of melanin is sensitive to light and also acts as a type of conductor". These statements indicate that the device uses human hair directly, not purified, extracted melanin which further invalidates the claim. The melanin can't be electrically active because keratin is an insulator. Human hair is non-conductive and not photochemically active as published articles and my own experiments show.
 
In other online articles (July 2009) on this topic, we find additional details that explain exactly what is going on: "A group of five students of the Trinity Science and Technology Guild led by Milan Karki and Harihar Adhikari demonstrated Tuesday how they succeeded in devising such a module by making use of copper, iodine, black hair, silicon dioxide and black chart paper at a minimum cost of Rs 3,000... "Unlike the commercial solar cell, we used moist black hair as a main component in our solar cell. We replaced the silicon bridge with a natural black hair where it serves as a main agent to convert the solar radiation into electrical energy,” Milan, the main architect of the module, said while addressing a press conference in the capital, Tuesday. The hair fiber has to be immersed in salt water (NaCl solution) whereby the hair gets moist as dry hair is a bad conductor of electricity and when moist it acts as semi-conductor."  A more recent Gizmag article and my response (October 2009) provides even more detail and some close-up photos of the device.
 

Technical Overview

This turns out to be a variation of DIY solar cells such as the cuprous oxide cell, the cuprous oxide/iodide cell, or dye-sensitized cell. A simple DIY flat cell is very similar to the student design, except it just uses salt water without the hair. In the case of the hair solar panel, the hair is serving as a sponge to replace the electrolyte container. The student's explanation is completely wrong, and I don't know how the faculty advisors didn't catch these obvious errors: the cuprous oxide is the semiconductor responsible for producing electricity from sunlight, and the salt water contained in the hair sponge is merely an electrical conductor (not a semiconductor, as claimed).  Saying that hair becomes a conductor when immersed in salt  water is like saying a brick becomes a conductor when you cover it with tin foil.  The basic science behind this so-called invention is simply wrong.
 
The melanin in the hair is immaterial, and the black hair has absolutely nothing to do with converting solar radiation into electrical energy as claimed - aside from the fact that it is black and, therefore, doesn't reflect light. The hair itself is not conducting electricity whether wet or dry; the salt water is the conductor, and keratin doesn't have the atomic properties to act as a semiconductor. In other words, the students could have used any dark colored synthetic or natural fiber to hold the salt solution or simply immersed the copper plate in a bowl of salt water and eliminated the hair altogether. The user of the solar panel will need to keep the hair mat wet at all times, or the cell will stop working.
 
I suspect the idea for a solar cell based on human hair comes from a mis-reading of some information the students found on the internet. A paper entitled Photoelectrochemical Properties of Melanin was posted in 2007. This is preliminary work that was never published in a peer-reviewed journal, not to mention that it's nonsense, as the experimenters don't seem to know the difference between a battery and a photocell. Here is a brief excerpt from the paper (I have corrected some English errors for clarity): "We started with a very simple cell where the electrolyte was a 1.3% solution of melanin in distilled water, copper and aluminum electrodes 2.5 cm apart, and copper wires (covered with silicone) attached to the electrodes with glue. We noticed that any kind of soldering affected the melanin's behavior. The cell started to generate electricity just a few minutes after being assembled."
 
To me, the use of dissimilar electrodes in an electrolyte solution in the Nature Precedings paper suggests, not a solar cell, but an electrochemical cell of the type used for simple demonstrations. To ensure that electricity was being generated photovoltaically, the Nature Precedings authors should have used the same metal for both electrodes. What the original paper authors probably meant to write was "covered in silicone" (final "e"), meaning simply that the wires attached to the electrodes in their cell were typical laboratory silicone insulated test leads. I believe the students who constructed the hair solar cell misunderstood "covered with silicon" from the original paper and that's why they tried to use silicon (no final "e") in their hair cell design. The referenced article states that authorities are waiting for the students to build a prototype without silicon, and "The [hair cell] experimenters claimed that silicon dioxide was used only at the edge as electrodes. They said it can be replaced with lead."  The students apparently borrowed concepts from the Nature Precedings paper without fully understanding them.
 
Furthermore, the Nature Precedings authors used purified synthetic melanin for their experiments, and the students naively believed they could substitute human hair since it contains melanin. The final bit of misinterpretation is that melanin only enhances the photoelectric effect of organic dyes. It does not produce electricity when used by itself, but the students claim, "We replaced the silicon bridge with a natural black hair where it serves as a main agent to convert the solar radiation into electrical energy." In summary, the students found some preliminary (and faulty) research on the internet, misunderstood the research, and based their design on this misinterpretation. Direct quotes from the students describing their invention contradict well-established principles of physics and chemistry.

The Gizmag article includes a schematic provided by Milan Karki.  This schematic drawing makes absolutely no
sense and indicates a complete lack of technical understanding.  First, the blocking diode is installed backwards -- the band (cathode) should be connected to the postive terminal of the battery, not the solar cell, as pictured.  Though it might be a typo and they meant "battery", I have a feeling that the comment "a diode used to prevent discharging of the cell" means the students thought that the solar cell could discharge, which just underscores their lack of understanding.  The next problem is that the negative terminal of the solar cell connects through a metallic pin to silicon dioxide "stain" on a glass substrate, and silicon dioxide is an excellent insulator (I have a feeling the teens meant to use titanium dioxide here, but just didn't know what they were doing).  The negative lead cannot possibly conduct as shown.  The positive terminal is cuprous oxide also connected with a metallic pin.  Cuprous oxide is a semiconductor and can be used to make a solar cell, but not with an electrode connected directly to it as shown.  The electrode should be connected to the underlying copper, not the cuprous oxide layer as labeled in the diagram.  Finally, the hair fibers interconnecting the anode and cathode make no sense at all.  Hair doesn't conduct an electric current and, even if it were generating electricity as claimed, the physical arrangement makes no sense in terms of collecting the emitted electrons.  Kaeki says the hair is saturated with "organic salts" to make it conductive. Elsewhere he says the "salts" are NaCl, so I assume he means table salt, not something like sodium acetate.  In any event, since he claims the hair's job is to "conduct electrons," then why not use strands of copper wire?  The drawing submitted by Milan Karki is complete rubbish and his explanations are nonsensical.
 
Using organic materials or readily available waste products such as human hair to generate solar energy has its romantic appeal, but the cost of extraction, purification, and end-product stability must be considered. For example, separating melanin from hair requires the use of harsh chemical solvents and heat, and the disposal of these chemical agents as well as the energy required to heat the solution must be considered. Also, melanin itself isn't photo-reactive, but is typically used to enhance the current-generating capability of organic dyes, so a device would require organic dyes in addition to melanin. In short, we must consider not just the initial cost of raw materials, but all factors involved in manufacturing and using any particular solution. It may be an over-simplification, but consider that the element silicon, used in solar cells, is one of the three most common elements on earth and is essentially purified sand, making silicon a cheap, stable, and relatively inexpensive solution as compared to organics such as melanin in combination with organic dyes.
 

Melanin's Properties

While it is true that melanin can enhance the photovoltaic properties of organic dyes, the isolation of purified melanin, production of the accompanying dyes, and instability of these products makes the use of melanin for solar power generation infeasible at present - although research in the area of organics is ongoing. As regards the proposed solar panel based on human hair, it appears from photos that strands of human hair are matted beneath the front conductive cover which would require the hair itself to be a conductor or semiconductor. After my own testing and referenced tests, even hair processed in an electrolyte such as a salt solution remains, itself, nonconductive. It is possible to extract melanin from hair by dissolving the keratin (hair's structural protein) by boiling the hair in a strong acid or alkalai solution. The problem is that such harsh methods tend to destroy the pigment or alter its chemical properties. There are more concentrated sources of melanin (eumelanin) than hair available in nature, e.g. cuttlefish ink and the uveal tract and retinal pigment epithelium of bovine eyes. The most economical way to obtain melanin is to simply purchase the synthetic compound or its chemical precursors. The electrical, semiconductor, electroluminescent, and optical properties of melanin films are well-studied. The properties, which are highly variable, are a function of heat, presence of water, and other environmental factors. The use of isolated melanin to enhance the photovoltaic properties of organic dyes has been studied, for example, Dye Synthesized Photocells, Hanyz & Wrobel (2002) and especially in Solis, et al (2007). A discussion of melanin chemistry is found in Nordland, Boissy, et al., The Pigmentary System: Physiology and Pathophysiology, 2nd ed, Wiley-Blackwell, 2006, pp. 289-290. The overall conclusion is that, at present, the use of melanin for solar energy production is impractical because it requires the presence of additional expensive organic dyes and because such dyes have been shown to be unstable.
 

Verification of Prototype Solar Panels

  • In a darkened room, measure the resistance of the solar panel output leads from positive to negative and then reverse the leads. If the device is a true solar cell, the resistance will be high in one direction and low in the reverse direction.
  • If possible, disassemble and inspect the panel. Do a physical inspection for hidden solar cells or batteries. Solar cells are usually shiny and dark blue with silver metalization on the surface. Hold the panel up to a strong light and look for cells behind a thin layer of hair.
  • Expose the device to sunlight and ensure that it is generating voltage. Observe first if there is a change as the panel moves from darkness to light and a change when the panel is partially and then fully covered up while in full sun. Obviously, the power output should vary in proportion to how much of the solar panel's area is shaded.Measure and note the voltage and current produced using a voltmeter and ammeter in full sun. The inventors claim the solar panel produces 2,000 mA. At a bare minimum, the panel should be producing 200 to 500 mA in full sun. If the panel is producing significant current, then suspect that silicon cells are hidden inside, as mentioned above. If the solar panel is producing something like 20 to 50mA, that is not really a practical amount of current and would be expected of a typical homemade demonstration cell made in school laboratories.
  • Be sure to distinguish demonstrations operating and powered completely by the solar panel, versus demonstrations powered by batteries charged by the solar panel and do not mix these up in reporting results. A proper report would include the claimed voltage and current output of the solar panel alone in full sun, the types and sizes of rechargeable batteries being used, the claimed voltage, current, and Ah rating of the battery pack, how long it takes the battery pack to charge in full sun, and how long it takes the battery pack to discharge under a typical load at night.
 

Why The Human Hair Solar Generator Is A Hoax

  • The first question that comes to mind is this: If the melanin in hair is a significant solar electric generator, then why are all the other chemicals in this project necessary? Various articles on the internet (see links below) mention copper, graphite, salt water, cuprous oxide, silicon dioxide, and iodine as components of the cell. It is interesting that all of these components happen to be found in different student demonstration solar experiments. It's as if the hair solar inventors combined all of these experiments into a single invention; something that really makes no sense. If hair were the main generating component, as claimed in direct quotes from the inventor, then the solar cell would consist only of hair and electrodes. More importantly, in order to take advantage of melanin's semiconducting properties, it would need to be isolated and purified chemically. Since hair is a non-conducting medium, the melanin embedded in it cannot generate solar power. In fact, melanin merely converts UV to heat and cannot generate power unless it is purified and paired with organic dyes.
  • Hair is a non-conductor as illustrated by this controlled experiment. Furthermore, keratin lacks the atomic properties to act as a semiconductor. Bound to keratin, melanin acts neither as a semiconductor nor a photoelectric material. As far as I can tell, the hair in this experiment merely serves to contain the salt water solution, and the salt solution is the conductor, not the hair. Even if the solution somehow doped the hair fibers and made them a semiconductor (highly unlikely) then the claimed output would still be too low for the area of the panel. I expect the experimenters got different results for light and dark hair fibers and attributed better performance to darker hair, leading to the false conclusion that melanin was responsible. In fact, the different results were related to variations in reflectivity. If the students had used proper controls (i.e. other fibers besides hair), this would have been evident.
  • The claimed voltage and power levels of 9V/18W are far too high. There is not enough photovoltaic material present to generate these levels, insufficient surface area, insufficient energy in the pictured light source, and the diameter of the thin conductors interconnecting the cells is too small to carry the claimed current of 1 to 2 amps without burning up. Since a solar panel is typically dark blue, it seems reasonable to me that it could be concealed behind a thin layer of dark hair and the plastic tape or sheet that covers the front of the device, especially since the whole affair is then enclosed in glass. Since the students claim to have sent out samples, and one would expect that these samples would have to generate close to 9V/18W, I can't think of any other logical explanation. A concealed battery wouldn't last long enough, for example. There are numerous PV companies and programs in Nepal, making it realtively easy to obtain a 9V/18W panel: SolarBuzz, Solar Tukis, Swogun, Vajra, Seman, WWF, Himalayan Lights, SELF, and many others. Using the photograph shown with the article of the boy holding up the solar panel, I held up a tape measure to a mirror to estimate the dimensions of his panel and estimated it to be 2ft on a side. The advertised dimensions of the commercial solar panel listed above are 638mm x 278mm or about 2.09ft wide. In other words, the panel shown in the photo is roughly the right size to conceal a typical commercial 9V/18W solar panel.
  • Even if the students have produced a cuprous oxide solar cell using a hair mat saturated with salt water, I doubt the cell could provide 9V/18W as claimed. For this reason, I still suspect the unit contains a hidden silicon solar panel. A cuprous oxide cell produces roughly 0.25 volts at 50 uA or about 12.5 uW per 0.01 square meter. To get 9 volts at 50 uA, you need 36 cells in series, bringing the area up to 0.36 square meters. To get 9V at 2A you need 40,000 of the 9V assemblies or 14,400 square meters which is a panel 120 meters long and 120 meters wide.
  • By way of contrast, I have a silicon solar cell in my junk box. This cell is 2cm x 4cm and is rated at 0.55 volts at 300mA. Rounding down to 0.5V you would need 18 of these cells to generate 9V at 300mA. To generate 9V at 2A you would need 7 x 18 or a total of 126 of these cells. Assuming a 12 x 10 grid of 120 cells, this works out to 24cm x 40cm total for the silicon cell, as compared to the 14,400 square meters needed for the cuprous oxide cell. Note that even a bank of silicon cells like this would take literally all day in full sun to charge a 2400mAh 9V battery pack with enough juice to power a single low-wattage fluorescent light using a voltage inverter for just a few hours. This feat would not be possible with the students' hair solar panel as they describe it. One other point is that the solar panel would have to deliver 9VDC in order to charge a 7.2VDC bank of 6 rechargeable cells, as claimed, because each cell is rated at 1.2VDC not 1.5VDC. That means the inverter would have to be running on only 7.2VDC from the charged batteries, and that seems rather odd to me. In any event, a typical bank of 6 cells delivers probably 6VDC under load at 1 or 2Ah, so the claim of generating 9V/18W is misleading, as it confuses the panel output with usable energy from charged batteries.
  • Then, there are practical considerations. Ignore for the moment the claim that the panel generates 9V at 2A and let's say we just need 9V at enough current to charge a bank of batteries at a reasonable rate. Let's say we scale the solar panel up to 2m x 2m (4 m^2). Even if we break that down into four 1 m^2 panels, we still need four 1 m^2 pieces of sheet class to cover the panels. Then, we need: Enough metal to make the frames, enough chemicals (copper, cuprous oxide, silicon dioxide, iodine, and salt) to cover the panels, enough wire and tacks to wire up the panels, enough plastic to cover the hair, enough material for the panel backing, and enough hair to cover 4 m^2. I expect that the price is going to be much, much higher than $40 (just the sheet glass alone costs roughly $20 USD per square foot).  This doesn't even consider that the hair is going to rot rather quickly, since it has to be kept wet with salt water at all times. That means the panel will have to be re-built every 4 to 6 weeks and the hair replaced. Finally, a person will be required to re-wet the hair constantly, since dry hair won't produce any electric current. The technical expression for this is that the solar panel prototype "doesn't scale up."
  • I personally conducted an experiment to test whether human hair was conductive or was able to generate electricity when exposed to the sun or to artificial UV (generated by a UV LED). My test shows that human hair is non-conductive and does not generate electricity under any circumstances. I used a sample of jet black human hair.Of course, this is just a demonstration, not a controlled experiment, however, it's simple to replicate. If you have any doubts try it yourself. The photo shows strands of hair fixed to small white card. Interleaved loops of bare wire sit on top of the strands, and the whole fixture is secured with clear tape. The leads are attached to the test leads of a digital voltmeter. In this particular test, I illuminated the sample with UV from a UV LED and there was no voltage produced, even with the meter set to maximum sensitivity (capable of detecting a fraction of a mV). I tried the same test outside in full sunlight with no results. Finally, I measured the resistance of the sample at the most sensitive range, and the resistance was infinite, indicating that human hair is a non-conductor. You can closely duplicate the students' cell by modifying this flat cell. Saturate some translucent gauze with salt water and cover the cell with it. Dye the gauze black and see what effect it has. Replace the gauze with hair fibers and see if it makes a significant difference. You will see that the cuprous oxide solar cell doesn't produce nearly the power claimed by these students - if my calculations are right, 9V/18W would require a panel 120m x 120m! At the same time, you can convince yourself that black hair and melanin have absolutely nothing to do with producing electricity from sunlight in this application. Important note: be careful about using fluorescent lights as light sources for your experiments, as these produce enough EMI to register on your meter. The safest bet is to experiment using sunlight far away from sources of electrical interference, since you are looking for very small voltage and current effects. Also, be careful to keep your hands away from any of the circuits during the experiment, since your body can introduce capacitive effects, change resistance readings, and transfer EMI.
Conclusion:  It is not possible to use human hair in any configuration to generate electrical energy when exposed to light.
 

Why Perpetuate a Hoax?

  • The students involved here are fairly young and inexperienced. It may be that the students became overly-excited when their prototypes worked, and extrapolated to a solar panel that can generate 9V/18W using their technology. They held a press conference and the news coverage got out of hand, backing them into a corner.
  • I can tell you from personal experience that it is very easy to fool yourself when doing an experiment. For example, when I first did my test (above) it looked like I was registering a minute voltage. However, I noticed that the voltage didn't follow the level of illumination, and it disappeared when I went outside away from electrical fields. Also, if the students get caught up in the idea that melanin in hair is a source of energy, they may forget to do a simple control check by using other kinds of fibers that don't contain melanin. Perhaps the darker hair worked better because it got hotter and reduced the resistance of the circuit or triggered a thermoelectric effect having nothing to do with melanin. A control experiment would have sorted this out.
  • I've spoken to several reporters about this news story. I asked them why stories like this get printed without being fact-checked. The answer is that news agencies are under tremendous pressure to produce instantaneous results, because that's what the public demands. Faced with missing a deadline and risking their jobs, reporters often skimp on fact-checking.
  • On the surface, the story is appealing and fires the imagination. Some headlines claim the invention will "solve the world's energy problems." Folks like the idea of an obscure inventor detached from corporate, mainstream scientific research coming up with a radically new idea. It is an attractive prospect that people who don't have high-technology solutions at hand can solve their own problems. Somehow, the idea of using human hair -- a universally available resource -- seems attractive, especially given our feelings that anything "organic" is bound to be better than anything "industrial" like silicon (note, silicon isn't organic, but it is basically sand and is one of the most abundant elements on Earth). Folks use the word "sustainable" for this idea without really thinking it through; for example, the claimed design depends on additional chemicals that might not be so "sustainable."
  • Last, but not least, there's a commercial reason for perpetuating this story. Selling DIY (do it yourself) solar panel instructions is a thriving business. These companies like to spread any story that implies that an individual can make a homemade solar panel, because it lends credibility to their sales pitch. You will notice them posting spam for their products on blogs discussing this story, for example, Sakriya Neupane, a friend of Milan Karki, has posted the hair solar panel invention all over these DIY scam sites. These companies have web pages or videos about making cuprous oxide cells and trick people into thinking they can "scale up" such cells and power their home with them -- a completely bogus and irresponsible claim as shown by some calculations on SciToys. Here is an awesome DIY solar panel debunking site.

Epilogue

A case like this saddens me on several levels, and I feel the need to preach a little. Just skip this part if you don't want to hear it. A news agency picks up an interesting story without having it fact-checked by a subject matter expert. The story spreads all over the internet and people believe it because it has a romantic element and also because the story has a kernel of truth to it. What a waste of resources just because a reporter couldn't be bothered to do a bit of checking. Even a reporter who is not a science geek should know that hair is an insulator and that static electricity is unrelated to the photoelectric effect. Just the barest amount of research would have shown that melanin must be isolated to be used in organic solar cells. If I were a reporter, I'd at least do some minimal investigation of the solar panel to verify it. Simple stuff. Does it have a battery hidden inside? Does it have a silicon solar cell hidden inside? Does the amount of power it's generating make sense? That is just good reporting.
 
The same logic extends to other news sources picking this story up without fact checking it. It amazes me that some of the technology editors who propagated this story didn't know better. Again, you don't have to be a science geek to do a little research and at least question the claims instead of just publishing them as fact. Some of the comments on the blogs concern me a bit, too. I can't believe the number of people who seem to lack a basic familiarity with science. I hate to come off as preachy, and maybe because it's important to me personally I expect others to have the same priorities. I was just disappointed at the lack of critical thinking and analysis I saw on some of the blogs.
 
Then we come to the kids. I blame their teachers for a lack of guidance and for not teaching the kids the basics of the scientific method. For example, did they even test the panel with other fibers besides human hair to see the effect? Why didn't the advisor teach the students to use experimental controls? Did the advisor not question the student's claim about human hair that "when moist it acts as a semiconductor".
 
And finally, there's scientific fraud. I know that in some countries the pressure to succeed in school is enormous. Students compete on exams to make the top couple of percent in order to gain admittance to a university. Professional scientists and academics all over the world commit fraud because of the pressure to publish results. I don't know for sure if these students committed fraud, though I strongly suspect they did because they claim to have sent out working prototypes. As I have shown above, it is not possible for a device of that size and technology to produce the claimed power levels. Again, I blame the faculty advisors for not vetting the design. The advisors should have questioned whether a home made solar panel could produce the kind of voltage and current claimed by the students. If the students were caught cheating, it was the adult advisor's responsibility to detect the fraud and discipline the students.
 
I have only kind words for Milan Karki and his colleagues. Mr. Karki, if what you have produced isn't valid, then drop it and take another path. That's how science works. Dye sensitized cells are a hot topic. Start with this source which discusses the use of natural organic dyes (berry juices) in solar cells. Order some lab grade melanin (or synthesize it) and combine it in a controlled manner to see what results you get. You need to start with pure melanin to do controlled studies, because extracting melanin from hair will produce impurities and variable quality unsuitable for research. Bear in mind that melanin alone won't generate electricity because melanin only acts as a "frequency shifter" for light. You will need to combine melanin with a large variety of organic dyes and vary the concentration of melanin. To determine the starting melanin concentration, consult current publications starting with the Gratzel Cell and working your way forward. It may take you hundreds or even thousands of trials, but that's how Edison worked! You might also consider thermo-electric generation, using wind to turn a generator, a solar stirling engine generator, and fuel cells. If you need electricity during load shedding, investigate energy storage methods such as batteries in addition to power generation. If you need light, investigate lighting methods like luminescence or pyrognomic materials.
 
In your application, thermoelectric seems to be a good choice because the generator can be made out of junk: scrap metal for a heat collector/radiator, wire (copper, aluminum, steel, nichrome from old toasters) and stones or glass for insulators. Some examples here seem to provide enough energy to charge batteries, but it would take a lot of experimentation to build something practical. Here's a rough design I came up with to get you started. If you try it, drop me an email and let me know what happens.  Here's a young man in Malawi who's the real deal:  he builds windmills to power pumps and generators.
 
My message is this: Don't worry about glamour. Focus on practicality, maintainability, and using what is inexpensive and available to you. Contact me for advice and guidance. I will be happy to assist you. Correct your mistakes. Educate yourself. Keep working. You are a big fan of Thomas Edison and so am I. His words still ring true:
 
"Genius is one percent inspiration and ninety-nine percent perspiration." -Thomas Edison 

Notes

  • Republica: (July 2009, hair solar cell demo/press conference, comments by experts) However, experts pointed that the true success of the experiment can be determined only without the use of silicon. Speaking at the function, Dev Kumar Shah of NAST (Nepal Academy of Science and Technology) said, “It is a good effort, but NAST will certify it as a viable alternative energy only when it is experimented without silicon coat.”
  • Gizmodo: My PhD is in Organic Materials for Photovoltaics... I am horrified that Gizmodo has published this article verbatim, without even questioning the basic science behind this "invention"... [M]elanin is really bad at charge separation, a basic requirement for solar cells.
  • ScienceGeekGirl: That's where this seems to bring in the bogus-icity of the science. I'm not sure if [melanin] being bound to keratin will make a difference, but it's certainly insulated by other materials in hair.
  • Jim Hall: Human hair is not suitable for any use in a PV array. Shockingly bad science.
  • The Nature Conservancy: OK, so it turns out that you can’t get solar power from human hair, ...
  • Bright Green Blog: A Nepalese solar panel made from human hair? We’re not convinced.
  • Examiner: Engineers who have commented on this story say it is a HOAX,
  • Geek: Did someone say shenanigans? It appears the solar panel created from human hair is among many things in this world too good to be true.
  • Treehugger: This smells like a hoax... [R]eading the Daily Mail piece, things don't quite add up... I'm filing this into the "probably too good to be true" drawer...
  • Fast Company: [I]t's unlikely that the panel can generate enough juice to charge a phone with that much surface area
  • Museum of Hoaxes: "...doesn't look like it would be enough to generate the electricity they claim to generate" (Treehugger)
  • Mythbusters: It's rubbish. And I'm quite sure this came up 5 or 6 years ago.
  • CEMA: Previously reported human hair solar panel most likely a hoax
  • Eric Busboom: Daily Mail sucked in by science hoax.
  • Gingers Have More Energy: New research from the La Salle University in Bangladesh, by Louis de Larreviere claims it can prove ginger headed people have more energy than any other coloured hair type (satire).
  • PESWiki: Complete bunk; surface area doesn't compute; a scam; can't make electricity from hair; a hoax; some science in melanin as a semiconductor.
  • SciToys: (commenting on cuprous oxide cell lack of scalability) "To run a 1,000 watt stove, you would need 800,000 square meters of cuprous oxide, and another 800,000 square meters of plain copper, or 1,600,000 square meters all together. If this were to form the roof of a home, each home would be 282 meters long and 282 meters wide, assuming all they needed electricity for was one stove. There are 17,222,256.7 square feet in 1,600,000 square meters. If copper sheeting costs $5 per square foot, the copper alone would cost $86,110,283.50 USD.)"
  • Clean Energy News: The prototype has produced eight volts of electricity and successfully lighted a three-watt CFL bulb. Unlike the commercial solar cell, SHM has used hair as the main component in generating electric energy. "We replaced the silicon bridge with natural black hair, which serves as the main agent to convert the solar radiation into electrical energy. It is based on the recreation of artificial photo synthesis principle," Karki said. Karki, whose idol is Einstein, said the power potentiality would depend on the varying size of SHM. The cost of producing eight volt electricity was around Rs 3,000; against Rs 20,000 by commercial solar panel.
  • Olivia Lang: (email exchange with the Daily Mail article's author) "[Y]es, of course I saw the solar panel and read the voltage."
  • WordPress: (Daily Mail author's breathless assessment of Milan Karki's intellectual gifts) "The eighteen year old is clearly a genius. ... Personally, I would definitely put bets on him being the next Edison."
  • Reddit.com:  a funny T-shirt.
  • Gizmag: Article from October 2009 providing more details and a schematic and Craig's response.
  • YouTube:  Movie claiming to demonstrate hair generating electricity under a light source.  Unfortunately, the meter is just picking up EMF from the nearby fluorescent light with the wires going to the "solar cell" acting as antennas.  A controlled experiment would use a 30cm wooden stick with a card attached to the end to alternately block/unblock the light source with the meter always connected and keeping the hands well away from the wiring.  This would demonstrate that the solar cell was not generating any voltage.  Another control would be to simply omit the hair and see what results were obtained.  This video is a perfect example of how easy it is to fool yourself and why proper scientific controls are necessary.
  • GettyImages:  Search for more detailed closeup photos on Getty Images.
  • Solar Energy Tech: List of solar energy companies in Nepal, two of which are connected with the name Karki: Energy and Construction Company Pvt. Ltd., Kalanki, Kathmandu, (Bhuwan Karki); C-ZONE HVAC Incorporate Pvt. Ltd., New Baneswor,  Kathmandu-34, (N. R Karki).
  • edisonmilan92:  Milan Karki's "inventor" blog.
  • Prime Minister Madhav Kumar Nepal on Friday...gets punked.

Contact and Press

 
Contact me at craighyatt@live.com with your comments.  Comments from subject-matter experts are especially welcome, and if you find mistakes, I will gladly fix them.
 
 
Credits:  Photo of Milan Karki above is (c)Tom van Cakenberghe/Barcroft Media; solar cell schematic courtesy of Milan Karki via Gizmag;  T-shirt image courtesy of Reddit.com.