Can a Photomultiplier Detect Single Photons for a DIY Double Slit Experiment?

[Edgewood11]

I'm trying to perform young's double slit experiment at home and I need assistance. Several problems arise such as what should be used as the light source (laser vs low watt bulb) and the photon detector (photomultiplier vs phototransistor). My goal is to replicate the scenario where one photon at a time exist at any certain moment. This is problematic because finding a light source that acts this way is difficult. I considered a very low wattage light bulb that would emitt only 1000 photons a second. Sounds like a lot but light takes time to travel. The next problem that arises is how to detect a single photon? The only logical answer is to use a sensitive photomultiplier. Quite frankly I am not familiar with the device but I understand how it works. Just let me know what you guys think. Pitch me a whole new experiment set up if you please! I have researched this extensively so I feel I am on the right track. (And yes I know you can simply shine a laser at a small double slit and get the interference pattern on the wall. I want to replicate the single photon at a time idea used in Young's real experiment.)
Thanks, Cam

 

[Simon Bridge]

http://qubit.nist.gov/qiset-PDF/Nam.QISET2004.pdf - (slides)

Suggests HBT Interferometer to detect, and a several possible commercially available sources - including laser+attenuator.

A 1000 photon per sec bulb? 10^-15W (or so) bulb? You can get those?
A bulb has the problem of having a broad spectrum of light and not very directional - for good results, and since you want to count individual photons, you want the light as close to monochromatic as you can get and highly collimated.

Light travels at approx 300000kmps - so it would cross an average 3m room in a bit over 100th of a millisecond. So one every millisecond would be good to aim for sure.

The trick then would be catching them. You'd have to do a run, wait for the detector to accumulate counts, stop, move the detector, rinse, repeat. Ideally you want an array of detectors.

But if your aim is to show the interference spectrum from a diffuse source, then a CCD array would be good to collect counts from a lot of different places at once.

Basically - work out your aim, your constraints, and see what other people have done.

 

[sheaf]

I considered a very low wattage light bulb that would emitt only 1000 photons a second.

That's an extremely low light level. Are you sure you can make a room dark enough so as not to completely swamp this source ?

Also, there is some discussion to the effect that it is not possible to generate single photons from a multi-atom source (such as a bulb).

 

[Edgewood11]

Which detector and light source do you see as cheapest and most effective because I obviously do not have access to some of the expensive detectors?

 

[Cthugha]

Which properties are you willing to sacrifice first? Attenuating a light bulb or laser can get you to a situation where you have one photon or less ON AVERAGE in your setup, but you can never assure that there is only one photon at most in your system. You would need non-classical light sources for that. Those are usually lab-only or not cheap.

If you insist on using very low light levels, you need single photon sensitive detectors. This will usually mean avalanche photo diodes or the likes. However these have high costs, too. Also you might need strong ambient light suppression. If you skip the low light levels a common laser pointer, a sheet of paper as the detector and your eyes will do. This will of course kill the cool effect.

 

[Simon Bridge]

It's not that exciting to see anyway - it's not as if you can see the photon traverse the system or watch the counts accumulate into a pattern like you see in the animations.

In fact, with one photon per millisecond, wouldn't it look pretty much the same as you get off the usual optics experiment - only really really dim? (i.e. the distribution of counts still accumulates faster than human time-scales?)

Anyway, I know there are people who have done well out of college contacts and discarded lab equipment so it may be possible to do something good. It is impossible to advise properly without knowing the resources available. Provide an idea of what you have available, what you can spend, and where you are, and it is possible a member may be able to give you a hand. You know: since you are very very keen.

 

[Edgewood11]

Yeah I think before going any farther I must solve the single photon detector problem as I think the light source will come easier. Have you heard of photomultiplier tubes? I can find ones on eBay for 20 bucks. It's worth a shot.

 

[Edgewood11]

The experiment would of course take place in complete darkness. Secondly, maybe a variable voltage power supply would do the trick? As far as the detector goes I am lost. The more I research the topic the more confused and doubtful I become.

 

[Cthugha]

It's not that exciting to see anyway - it's not as if you can see the photon traverse the system or watch the counts accumulate into a pattern like you see in the animations.

Oh, in principle you could see the pattern build up if you feed the signal of your detector to some software which shows you an accumulated detection pattern. You would most likely fall asleep while doing that because it can take pretty long, but in principle it is possible.

Yeah I think before going any farther I must solve the single photon detector problem as I think the light source will come easier. Have you heard of photomultiplier tubes? I can find ones on eBay for 20 bucks. It's worth a shot.
[...]
The experiment would of course take place in complete darkness. Secondly, maybe a variable voltage power supply would do the trick? As far as the detector goes I am lost. The more I research the topic the more confused and doubtful I become.

PMTs are pretty sensitive, but usually have problems with afterpulsing. I am not sure whether that is a problem for you.

As regarding complete darkness, you would be surprised how really low light levels get caught by sensitive detectors. That keyhole in the door can be enough. When I use SPADs, I routinely place black tape over every even so tiny LED even at the far end of the lab.

 

[Edgewood11]

Using software to get that accumulation pattern is exactly what I'm trying to do! I mean are PMT's sensitive enough to detect individual photons? Plus how would that work because a PMT can't determine the exact location where the photon hit. It simply sends an electric impulse when a photon triggers the photocathode and then electrons bounce off dynodes and then finally hit an anode. As far as making it completely dark might be hard but I don't foresee this as being a major problem. Maybe just time consuming. Going back to the PMT, the sensor that collects the photons is only like the size of a cell phone battery so I don't know have I would do the accumulation pattern. But that pattern is my goal. I wish there was a detector device that was simply a screen the size of a piece of paper that sends signals that plot points on a graph.