It may be somewhat easier (everything is relative!) to construct than those other lasers in terms of the required chemical/gas supplies (only a drop of mercury and helium) and ease of mirror alignment (since it uses a wide-bore tube). However, a medium vacuum system and some glass work will still be needed.
Another Helium-Metal-Vapor laser is the Helium-Cadmium (HeCd) laser which is available commercially and operates in the blue (442 nm) and near-UV (325 nm) wavelengths. See the chapter: Helium-Cadmium Lasers for more info.
And yet another, the Helium-Selenium (HeSe) laser, can generate wavelengths all the way from red to UV depending on the mirrors used. While I don't know of commercial versions of this laser, it appears to be an interesting candidate for a home-built laser once you master the HeHg laser. See the sections: Mark's Helium-Selenium (HeSe) Laser and Discussion of Other Helium-Metal-Vapor Lasers.
WARNING: Mercury, selenium, and cadmium are ALL classified as toxic heavy metals though I don't know which kills you the fastest. Once a laser tube is sealed off, there is no danger unless it breaks. However, safe handling procedures must be followed for toxic heavy metals when dealing with the initial filling of the laser tube and disposal of any excess material. Where the laser tube is not sealed gaseous and liquid mercury, or gaseous cadmium or selenium (depending on what you are using) may be sucked up by the vacuum system so a trap must be provided on its inlet port.
Avoid eye contact with the direct or reflected beam. This includes the 4 pairs of beams reflecting off the Brewster windows which may be quite strong.
More information on the hazards of mercury and other heavy metals can be found from the Material Safety Data Sheets (MSDS). One source is the Cornell Material Safety Data Sheets Page.
Provide proper warning signs for both the laser radiation and high voltage. Keep pets and small children out of the area and make sure everyone present is instructed as to the dangers. The use of proper laser safety goggles for the specific wavelength(s) of your laser are highly recommended.
For more information, see the chapter: Laser Safety. Sample safety labels which can be edited for this laser can be found in the section: Laser Safety Labels and Signs.
See the Amateur Scientist article in Scientific American (October, 1980). Also:
Refer to Typical Home-Built Helium-Mercury Laser Assembly for a simplified diagram of the overall glasswork and power supply electronics.
D1
H o--------+ T1 T2 +--------+--|>|--+--+-------+---o HV+
)|| ||=||( | | | |
Variac )<--------------+ || ||( | D2 | | |
0-115V )|| )|| ||( +--|--|>|--+ | /
3A )|| Neon Sign )|| ||( | | _|_ C1 \ R1
)|| Transformer )|| || +--+ | | --- 10nF / 100M
+--+ 9kV,20mA )|| ||( | | | D3 | 15kV \ 5W
| )|| ||( | | +--|<|--+ | | 20kV
N o-----+-------------------+ || ||( | | | | |
||=||( | | D4 | | |
| | +--|--+-----|<|--+--+-------+--o HV-
| | |
G o---------------------------+---+----+
_|_
////
No other aspect of the laser tube assembly itself is as important as the quality of the Brewster windows to the ultimate outcome of this project! While, certain types of distortion won't prevent lasing (some may even make it more exciting with complex mode structures), this is a complicating factor your first laser can do without.
CAUTION: Apparently, it is possible for an electron beam to be produced from the positive electrode during the high current bake-out step which can quickly melt a hole through the tube wall opposite the side-arm if left running for more than a few seconds. The visual effect will first be a spot of bright yellow sodium light from the point of impact. Use lower current and/or make that area of the tube (the outside of the turn) much thicker.
WARNING: Mercury is considered a toxic heavy metal so appropriate precautions should be taken in its handling and disposal. Vapors are poisonous and can build up in a confined space. Mercury combines readily with many substances resulting in hazardous compounds. See the section: Home-Built HeHg Laser Safety.
You could try a chemical supply house but there are lower cost alternatives to buying a kilogram of mercury (assuming they will even sell it to you) when all that is needed are a couple of grams at most.
Your local friendly dentist has plenty - that's what he mixes with silver to make those wonderful amalgam fillings you love to have installed. What a great excuse for a checkup! If you can convince him/her that you know what you are doing and provide a suitable container, it should be a simple matter to donate a couple drops to a good cause.
Don't want to go near dentists? I don't blame you!
OK, you probably have one or more silent switches in your house or apartment. These are the kind with absolutely no detents or snap action whatsoever. They are most likely mercury switches. Find an excuse to change one. "Ma, there are sparks, smoke, and 10 foot flames coming out of the wall..... I think this switch is bad". ;-) Inside a mercury switch will be one or more metal/ceramic capsules - they sort of look like tall flying saucers and can be cracked open in a vice (make sure you do it in something to catch the mercury!). There is ample mercury in a single one of these for any reasonable HeHg or similar laser or mercury vapor discharge experiments.
Of course, tilt switches out of thermostats, water level sensors in sump pumps, and similar devices may have even more mercury. Since they are usually made of glass, getting at it will be somewhat easier. But don't take them from where they will be missed: "Tommy, it's roasting in here and why is the basement flooded?". :)
Note: Fluorescent and high intensity discharge lamps also contain a drop of metallic mercury but it isn't worth the trouble to get at it. And, smashing these light bulbs may release other toxic substances! With the pressure to prevent environmental contamination, only very small quantities of mercury are used in modern lamps (maybe a few milligrams for a 4 foot tube). Thus, not recommended.
Refer to Typical Home-Built Helium-Mercury Laser Assembly for a simplified diagram of the overall glasswork and power supply electronics which should be similar except for the dimensions.
(Specifications from: David Hansen (gldhansen@rcn.com).)
D1 Spark Gap
H o--------+ T1 T2 +--------+--|>|--+--+-------+--> <---o HV+
)|| ||=||( | | | |
Variac )<--------------+ || ||( | D2 | | |
0-115V )|| )|| ||( +--|--|>|--+ | /
15A )|| Neon Sign )|| ||( | | _|_ C1 \ R1
)|| Transformer )|| || +--+ | | --- 20nF / 100M
+--+ 15kV,60mA )|| ||( | | | D3 | 35kV \ 25W
| )|| ||( | | +--|<|--+ | | 35kV
N o-----+-------------------+ || ||( | | | | |
||=||( | | D4 | | |
| | +--|--+-----|<|--+--+-------+--------o HV-
| | | D1-D4: HVC1200
G o---------------------------+---+----+
_|_
////
I am just about to finally complete a mercury vapor laser as per the SciAm plans. All that I have left to do is polish flat the brewster angle cuts at either end of the plasma tube so my quartz windows will fit snugly. I am looking forward to many hours of exciting experimentation with this rare type of laser. I will get some pictures of the completed system to you as soon as I can.
A Helium-Cadmium (HeCd) laser could probably be built using the same basic design as the Hese laser given below.
This HeSe laser was built by Mark Dinsmore (dinsmore@ma.ultranet.com).)
Refer to Mark's Home-Built Helium-Selenium Laser Tube for a simplified diagram of the overall glasswork and electrode structure.
In addition to a vacuum gauge, there should be at least one trap between the pump(s) and the laser tube to prevent any mercury from contaminating the vacuum system and/or escaping to the atmosphere. The exhaust from the mechanical pump should be passed through another trap if possible before being vented outside and away from open windows.
D1 R1
H o--------+ T1 T2 +--------+--|>|--+--/\/\---o HV+ (Main)
)|| ||( | |
Variac )<--------------+ ||( | D2 | R2
0-115V )|| )||( +--|--|>|--+--/\/\---o HV+ (Cat)
10A )|| Copier )||( | |
)|| Transformer )||( | |
+--+ 4kV,250mA )||( | | D3
| )||( | +--|<|--+
N o-----+-------------------+ ||( | |
||( | D4 |
| +-----+-----|<|--+-------o HV-
|
G o---------------------------+
_|_
////
An Oudin coil (tesla coil type vacuum tester) is used to start the discharge eliminating the need for a much higher voltage transformer or separate starting circuit. A pair of power ballast resistors distributes the current in approximately the ratio 10:1 between the main anode and the anti-cataphoresis electrode.
I gather from the "Light and its Uses" HeHg laser description that while the laser produced 'intense pulses of green and red light', the average output power wasn't more than a mW or so. Even a 1 mW of 567 nm green would appear quite intense - 4 or 5 times brighter than 1 mW of 632.8 red.
Also see the section: Comments on HeHg Lasers.
However, a tube full of mercury vapor at a couple of Torr is not much material so a drop of mercury should go a long way. In addition to losses out the vacuum system and by condensation on the tube walls, some may form an amalgom with the metal of the electrodes or their lead-in wires (depending on the metals involved), be adsorbed by the tube walls or Epoxy seals (if used), or be buried under sputtered electrode material.
Also see the section: Comments on HeHg Lasers.
Refer to Home-Built Pulsed HeHg Laser Using HeNe Laser Tube for a simplified diagram of a possible system.
The idea would be to take a long certifiably dead but undamaged HeNe tube, break the seal, add a drop of mercury, and attach it to your vacuum/gas supply system. I would recommend a tube of at 18" to 24" in length to provide enough gain to have a chance of lasing. The mirrors on a red (632.8 nm) HeNe tube should work for to the 615 nm HeHg wavelength; the mirrors on a green (543.5 nm) HeNe tube could work for the 567 nm HeHg wavelength. In both cases, the mirror reflectivity will be quite close to the value at their design wavelengths. I don't know how the narrow bore of the HeNe tube will affect gain - usually a narrow bore has higher gain than a wide bore. The optimal gas pressure and fill ratio (He to vapor pressure of Hg) will also likely be considerable different for this shorter and narrow bore tube.
Actually, adding a gas inlet port at the opposite end from the exhaust tube (probably the anode-end and the exhaust is usually done at the cathode-end), would be better than just injecting a glob of mercury. To do this would require drilling a tiny hole in the side of the mirror mount and installing a thin piece of tubing to connect to your helium supply and mercury reservoir. The advantage would be in being able to better control the temperature of the mercury and thus it vapor pressure. Vacuum requirements are also less stringent with a flowing gas system. However, since the high voltage will then connect to the same place as the gas/mercury feed, some means will have to be provided to prevent the discharge from taking place via the this route (e.g., a long thin plastic capillary). And, of course, everything there will be floating at a high voltage! CAUTION: If you are using what was once a polarized HeNe tube, there will be a Brewster plate inside at (usually) the HR-end which might be damaged. Also, drilling may leave metal particles and/or a raised ridge that could be difficult to remove.
In addition to a vacuum gauge, there should be at least one trap between the pump and the HeNe tube to prevent mercury from contaminating the vacuum system and/or escaping to the atmosphere. The exhaust from the pump should be passed through another trap if possible before being vented outside and away from open windows.
One of the concerns with respect to tube life would be possible condensation of metallic mercury on the mirrors. However, some (low temperature) heating tape wrapped around each one would probably provide adequate protection given the low vaporization temperature of mercury.
Another problem with using an existing HeNe (or any other type) laser tube is that the metallurgy will not have been selected with mercury in mind. Mercury will combine (amalgamate) with many metals including of course silver (remember all those wonderful fillings?!) and copper. What implications this has for the life of a home-built HeHg laser, I do not know. Since only small quantities of Hg will be will be introduced to the tube in the form of a low pressure vapor, this may not be a serious short term problem at least. I did some simple tests with a drop of mercury and the metal of a HeNe tube without any obvious effects but didn't let it soak for very long!
A power supply similar to the basic neon sign transformer/rectifier/HV capacitor specified for the SciAm home-built HeHg laser should work here as well, though some 'adjustments' may be needed for the narrow bore tube since the breakdown voltage is likely to be much higher than even for the much longer, but wide bore SciAm design.
Whether this will even work - or for how long - I cannot say at present. However, it's something to try without risking a lot of time or money. The cost of intact, but very likely unusable tubes (as HeNe lasers) should be modest or free: Laser and Parts Sources. The optics are internal and probably aligned correctly (though this should be confirmed) so much of the difficult work and cost is gone. If anyone attempts or succeeds at an endeavor such as this, please send me mail via the Sci.Electronics.Repair FAQ Email Links Page and I will add your experiences to the collective knowledge base on the home-built HeHg laser!
CAUTION: If you have any fantasies or desire to resurrect the original HeNe tube as a HeNe tube, DO NOT attempt this lunacy as there will be no practical way of purging it of mercury contamination after you are done playing.
Just off the top of my head, I'd say the problem with it is the mercury. That is toxic stuff, and it goes right through your skin. Make it a vapor and you would breath it. The less you fool with it, the better off you are.
Another problem might be that mercury is expensive. A pint jar of mercury would be worth hundreds or thousands of dollars (well, a lot, anyway). The same amount of copper would be worth only a few dollars.
(From: David Demmer (ddemmer@physics.utoronto.ca).)
Nah, in an earlier life I was a chemist, and demonstrated a physical chemistry lab that used a Toppler pump. For the uninitiated, this consists of some ingenious glassware and about 500 mL of mercury. The stuff is really pretty cheap when bought in quantity: actually about $100/kg for reagent grade from Aldrich. Of course a kg is a smallish volume because it's surprisingly dense stuff.
As for a laser like a helium-mercury vapour laser, the amount of mercury necessary is actually pretty small. And besides, expense doesn't stop anybody using a gold vapour laser. The cost of the fill is trivial compared with capital and other operating costs.
Toxicity, too, never stopped anybody from using HF, F2, etc. in their excimer lasers.
So, while not knowing the answer to the fellow's question, I guess I'm saying that I don't agree with yours.
(From: Richard Alexander (RAlexan290@gnn.com).)
Well, it looks like you pretty well stopped my argument for the expense, but I think toxicity is still in play. In defense of that argument, I present my belief that excimer lasers are used mostly in industrial settings, away from most people (ok, I know there are medical excimers that offer some interesting promise).
Maybe it's not too late for me to agree with the other poster that the wavelength put out by a mercury vapor laser is better produced by another type of laser. (I've had this discussion before, several years ago. It's just taking time for me to remember what I was told.) It seems like it isn't worth the bother of making the laser when HeNe and Argon Ion cover whatever the Hg vapor laser will do (ok, I admit I don't remember what lasers cover the Hg vapor laser wavelengths. You get the gist of my hazy memories.)
(From: Herman de Jong (h.m.m.dejong@phys.tue.nl).)
Mercury is used in home used thermometers, in thermostat switches for heating the home, in tilting switches, in high power reed switches, in TL end related light tubes etc.
The mercury laser uses maybe a few milligrams/year and it could be trapped in an amalgam with silver, copper, zinc, not iron, about the only exception. this effectively blocks mercury from reaching the pump.
A thermometer contains maybe a few hundred milligrams this is no problem.
(From: Herman de Jong (h.m.m.dejong@phys.tue.nl).)
I agree with you David.
If I remember right the CW laser wasn't peculiarly intense but it had low (single ended) optical feedback and the length of the vapour was a few feet. and it was a DC discharge at a few kV with a neon transformer. It needs a Helium flushing bottle and a vacuum pump. At that wavelength it probably can't compete with the more powerful Ar+ lasers (514 nm, some are sealed) and the much simpler green HeNe(sealed). The Neon transformer (6 kV, 100 mA?) suggests also very poor efficiency. But for your home built laser it is a nice project and even the vacuum pump can be very cheap by using the compressor from a discarded air conditioner, refrigerator, or freezer.
The expensive setup would only be acceptable it the power could be high. I suspect the population of the concerning state densities to be low although they have very high absorption/stimulated emission probabilities. This means the Sc.Am. is probably operating near saturation and it is hard to increase output energy. More length will have a linear effect, higher vapour pressure (T) is complicated. Maybe a different kind of discharge favors the population of concerning states but it could be a lot more complicated then this.
I did a bit of a literature search on the helium-mercury and other mercuary based lasers.
The Scientific American design is gain limited, but the power scales with tube diameter up to say 2 inches, and gain scales pretty much proportionally to length as well. The laser not only emits in the green and red, but has many lines in the IR around 1.2 to 1.8 microns. It has lased continuously on the visible lines in a hollow cathode type tube. It is also super-radient. It lases after the excitation pulse has almost died out (lases in afterglow). For all intents and purposes it seems like a ideal low cost laser and would probably be useful to someone.
However I've noted two things, No one ever seems to have measured the output power, save for one citation listed below and two, papers stopped appearing on it after 1970 or so. One wonders if it does not have some major fault, as it was discovered by one of Spectra-Physics chief scientists, and yet nobody makes it commercially?
The above book has plans for a two meter long Scientific American style design and a hollow cathode design excited by a thyratron based pulser, it does 10 W pulsed on the red line.
This still leaves the question, Why is it not in Use?
Here are some papers to check out:
Summary states: "Quasi CW operation of 3 laser lines, 650.1 nm, 521.0 nm 479.7 nm, was obtained in a pure Hg, very low pressure discharge. These lines are the Hg-III 5d86s2-5d96p transitions. A per pulse output power of 40 W at 650.1, 25 w at 479.7 and 5w at 521.0 was observed."
Summary reads: Using the 202 isotope of Hg (202Hg), gain as high as 120 mW/cm was obtained in a 2 to 5 cm diameter tube. However, gain varies over an order of magnitude as it is very sensitive to the temperature profile along the tube. Used 30 mg of Hg. Entire tube must be strongly heated. Used cathode of 20 cm long and 2.5 cm in diameter with ring anodes one cm away. Notes problems with fringe fields from any sharp edge of glass within 2.5 cm of cathode. Cathode must be closely fitted (1 mm gap) to outer glass wall to transfer heat.
Now we know a few reasons why mercury based lasers aren't commercial: The temperature variation, plus 202Hg isn't cheap. Along with a thyratron, it shoots itself in the foot. Xenon is looking a lot better then Hg, except that a turbo or diffusion pump is needed to get there. Most stuff I'm reading now says 10 to 20 microns for xenon III-IV ions, to get the really screaming lines.
(From: David Demmer (ddemmer@physics.utoronto.ca).)
Well done!
The large bore size and scaling sound a lot like a copper or gold vapour laser, i.e. unstable or planar multimode resonator. This would make it lose out to, say a Kr ion laser for beam quality, and a copper/gold vapour for raw power in the green or red.
On a different note, I don't know that the phrases "ideal low cost" and "thyratron based" should appear in the same discussion ;->.
(From: Sam.)
Aside from the cost issue of a thyratron pulsar, the need to keep fiddling with the gas pressure/fill to maintain laser action in an efficient and stable manner could be a factor in the lack of commercial viability. Unlike a fluorescent or high pressure discharge lamp using mercury vapor, the HeHg laser would appear to operate at a very low pressure (1 or 2 Torr) where a stable gas fill could not easily be achieved. Therefore, some possibly complex and costly feedback mechanism would be needed to produce a viable product (possibly along the lines of a low flow CO2 laser). This may push it over the cliff so to speak when compared to alternatives like those mentioned above. However, it should be noted, that the helium-cadmium laser has similar requirements but DOES fill a niche in wavelength where suitable alternatives have not been readily available.
Then again, perhaps it was a simple business decision to go with Ar/Kr ion rather than HeHg and now that any patents have run out, they are just keeping its potential benefits and ease of construction quiet. :-)
(From: David Hansen: (gldhansen@rcn.com).)
If anyone does build or has built a HeHg laser, many experiments can be done with the gas fill and cavity optics. Chris Chagaris (pyro@grolen.com) told me of a Hg vapor laser that used a small (1.5 mm) bore and used argon instead of helium. It put out many wavelengths in the blue and into the UV. Anyone up to trying all the noble gases in a Hg vapor laser tube? How about radon? :) If someone did do this, it would be easier to use white light optics, rather than constantly interchanging your mirrors and having to re-align them. Ball-and-socket joints might be nice for the Brewster windows as well since this would permit their angles to be optimized for each wavelength.
The HeCd laser is the only one of this class I know of to be readily available commercially operating at blue (442 nm) and near-UV (325 nm) wavelengths. The basic design is similar to other gas lasers with the addition of some means of heating Cd pellets contained in a reservoir inside the sealed tube. Typical power requirements are 800 to 2,000 VDC at 10 to 25 mA or more. The vapor pressure of the Cd, total tube pressure, tube current/voltage and temperature all affect output power and several control loops may be present to maintain one or more of these constant.
A description in standard format of the Helium-Selenium laser discussed below can be found in the section: Mark's Home-Built HeSe Laser.
The major contributors are:
(From: Chris.)
I agree that the typical "how do I build a powerful laser" posters are crude hobbyists who really have no idea what is involved in building a working laser. They are therefore likely to not know much about laser or high voltage/power safety precautions either.
For myself, I have built a CO2 laser to serve in a (heavily supervised) instructional laboratory session for undergraduate physics majors. Think about it: Is a physics major who really doesn't know how to build (or at least align the cavity) of a laser really deserving of a B.S. degree when he graduates?
My next project: A HeCd laser.
(From: Thomas C. Sefranek (tcs@cmcorp.com).)
Wow! Your getting a VERY blue output. (If my memory serves me.) Do you plan a vapor trap?
(From: John Woodgate (jmw@jmwa.demon.co.uk).)
Even the light will be toxic! :)
(From: Chris.)
A sealed HeCd tube poses no Cd hazard, unless you break it. There is typically some facility to trap Cd vapor within the tube to keep it from depositing on the optical windows. That is about the only bad thing that can happen, aside from significant difficulties getting the vapor pressure of Cd correct while at the same time getting the He partial pressure correct at operating temp.
Since it is difficult to properly evacuate and seal a sealed tube without sophisticated vacuum equipment and knowledge of glass to metal seals, annealing, bake-out, etc., it may be easier to build a flowing gas tube. I would do this with a helium flow manifold, preheater, a heated (tube heating is of course microprocessor controlled for accurate control of pressures) Cd pool, and a cold trap for Cd vapor at the tube's exhaust port before the optical window and before the vacuum pump.
I don't know if a flowing gas HeCd is more difficult than a sealed tube. It may be worth learning the glass to metal seal techniques (or contracting a glassblower that knows it), and investing in the UHV equipment.
Certainly a flowing gas CO2 is much easier than a sealed tube, the sealed variety requiring water cooling and a catalytic cathode that can regenerate CO2 from dissociation fragments. But HeCd may be the other way around.
Anyone got any experience with HeCd engineering they'd like to share?
(From: Steve.)
You might want to start with pulsed oxygen or xenon first to learn the glassblowing mechanics, as those are more tolerant of dirt in the tube and outgassing. Xenon, oxygen and even neon, all lase in pulsed systems at visible wavelengths. Xenon and neon are green, and used to be used in pulsed resistor trimmers. Oxygen lases in the blue and yellow. Just about any gas element has laser transitions. You never know what may work till you try. See the section: Home-Built Pulsed Multiple Gas (PMG) Laser which presents a design similar to the one described in the paper: "A Cold Cathode Pulsed Gas Laser" by R. K. Lomnes and J. C. Taylor, The Review of Scientific Instruments (RSI), Vol. 42, No. 6, June, 1971, pp. 766. This generic gas laser may be used with several different gases such as Ar, Kr, and Xe, but also with oxygen (O2).
The O2 laser lases in the afterglow, in a medium bore diameter tube, mainly in the yellow (559.2 nm), with much higher gain then mercury. It likes a long tube and needs pulsed excitation. The researchers who built this laser claimed much higher power then pulsed argon using the same tube, even with sub-optimal argon optics. It is a simple design using glass tubing, aluminum electrodes, and a Plexiglas box for cooling. Like small CO2 lasers, it was pumped in real time with a slow gas flow. The authors opted for a thyratron pulser to gain a lot more peak power but the basic neon sign power supply charging a HV capacitor should work as well.
After you have mastered the pulsed gas lasers, maybe then move on to HeSn or HeIn or HePb.
My only problem with the RSI lasers and the Hg design is nobody ever specs the required transmission for the output couplers, which really is what determines the power emitted by the laser. Too much transmission and it won't lase; too little and you don't get any power out and the tube saturates, making it look like it's not lasing well. The problem is we usually only have the money for one set of optics to try.
I used to own a long positive column self heated HeCd, a Liconix 401. It only lased in a range 5 mA wide at a tube current of about 250 mA tube current and had the tube lagged with asbestos for a certain thermal time constant. You cut holes in a piece of fireproof paper to adjust the air flow over one end of the tube to adjust the pumping of the cadmium from one end to the other via cataphoresis as the tube aged. It took 25 to 30 minutes to start lasing and was very tricky to adjust at first.
Modern tubes have a multisegmeted bore with multiple Cd sources, a glass frit helium pump and a bunch of other things to condition things right. So if you do it, having an oven on the tube helps as well as active pressure control. Helium pressure is probably going to be in the 7 torr range.
The toughest thing you will face is cutting the Brewster stems, the old thing about using abrasives and a brass blade in a hacksaw doesn't work well. You'll need stress free windows and stems. Lasing power more then anything else is dependent on the window quality and axial angle tolerances. A quartz window not properly cut will have birefringence properties that may prevent lasing entirely so you have to get good windows. Fused silica will work for a lab toy, but you'll have long term lifetime problems with them from solarization and the output power will be lower. You might wish to consider buying a few glass to vacuum flange transitions to make your life easier. And use a hot cathode - older tubes used off the shelf fluorescent lamp cathodes.
If your willing to use Pyrex, the seals are inexpensive and off the shelf. The actual glassblowing for a straight tube is easy enought for a beginner to do. There are a few good books on amateur glassblowing and a few days practice should get you all the skills you need. Look for a glassblowing book by C. L. Stong.
(From: Mark.)
There are lots of people who will tell you not to try a project like a HeCd or other Helium-Metal-Vapor laser, or that it cannot be done. Don't listen!!. I very successfully built a Helium-Selenium (HeSe) laser 5 or 6 years ago. It still runs great - I fire it up in my basement every once in a while. I used a 1 meter confocal cavity with argon ion laser mirrors and got 100 mW CW of multimode, multiwavelength light with lines from 540 nm to around 460 nm. If I had had a broadband cavity, I could have gotten 24 wavelengths from red to UV! It was a difficult project with a very satisfactory conclusion. Just use caution around the chemicals, HV, and laser light. (BTW, selenium is actually good for you in small quantities. :) Good luck!
(From: Steve.)
Mark, care to share your bore diameter and length? Partial pressures? Tube currents?
(From: Brian.)
Yes, please tell us more. This sounds like a very fascinating laser. I'd like to know more about what wavelengths are possible and at what relative powers, plus anything else you care to share as well.
(From: Chris.)
Very interesting. I don't know about HeSe. Only one question at this point:
I have figured that to control the metal vapor partial pressure you need only know a little about its thermodynamics, then by controlling the tube temperature, you control the metal vapor pressure. Of course, avoiding temperature gradients is important. Is this on the right track?
Is yours a sealed tube or flowing?
(From: Mark.)
The laser I built is based on cataphoresis for controlling the vapor pressure of the selenium. Selenium vapor is introduce at the anode end, is swept along the plasma tube by cataphoresis, and deposited on the walls of the tube at the cathode end. The cataphoresis/deposition keeps the partial pressure of Se constant in the gain region, allowing the laser gain to be achieved. The temperature of the Se source is quite critical, approximately 270 °C plus or minus a few degrees, but it is very easy to tune. The temperature in the bore needs to be hotter than that, but the plasma current self-heats this section and is not critical at all as long as it is hotter than 270 °C.
(From: Chris.)
Ok, I understand that cataphoresis will maintain a constant pressure over the tube length, but isn't the Se pressure still controlled by the temperature?
(From: Mark.)
Absolutely. The selenium is located in a sidearm of the tube, isolated from the discharge. An external, temperature controlled heater generates the proper vapor pressure, the vapor expands up into the discharge, and gets swept down the bore by cataphoresis.
(From: Chris.)
Did you do any cataphoresis rate and thermodynamics calculations, or did you just approximate someone else's construction? It is funny how much you don't need to know to make a laser. Mechanical innovation is the primary prerequisite, it seems, plus some EE wits.
(From: Mark.)
Absolutely, again!
I did them a long time ago for a class in electrooptics in college, but the reality of this kind of laser is that you just adjust the parameters to maximize the gain. A reasonably good way of measuring the potential gain without having the laser actually oscillating is to measure the sidelight at the max gain wavelength (520 nm) with a filter and a photodiode. That way, you can adjust the helium pressure, the selenium temperature, and the discharge current and monitor the effects.
(From: Chris.)
Since a plasma tube cathode is typically pretty hot, I would expect that end of the tube to be hotter than the rest, and so unlikely to condense the Se. Does your Se condense *after* the cathode? Or perhaps you cool the tube walls around the cathode? I guess you don't have to worry about deposition on the cathode, huh. But perhaps it wouldn't even matter if it happened.
(From: Mark.)
After the capillary bore, I allow the diameter of the discharge tube to expand. This allows a cooler region where the selenium can deposit before getting into the cathode area. As you say, the cathode runs quite hot, so I don't think much Se lands there. one of my prototype tubes did exactly what you say - the Se was deposited behind the cathode in a cool area near the getter.
The big problem is keeping the Se off the Brewster window on the cathode end of the tube. I elaborated on someone else's idea and run a reverse current of about 10% discharge current to keep the Se off the window. All will be clear when I get my report done.
If the source temperature is too low, the plasma is the pink color of He. If it is too high, it turns a vivid dark blue. If it is just right, it is bright white! The helium pressure is about 10 Torr, not very critical.
I have about 100 hours on the tube, and have seen almost no depletion of the Se in my reservoir. The construction is a sealed pyrex glass envelope with a 1 mm bore plasma tube approx 0.8 meter long. It has been sealed off for >6 years with epoxy sealed brewster windows, and it still works!
(From: Chris.)
Now that is impressive. I have done a few glass/epoxy/metal vacuum seals, and have found them remarkably good for the ease of performing them. From what I have read so far, glass/metal seals are not trivial. Good to hear about a sealed tube holding up with epoxy seals. I wonder, just hardware store epoxy, or something special?
(From: Mark.)
Special. I was using Epotek 305, and it worked quite well. I have recently discovered another substance, Vac-Seal, which has an incredibly low vapor pressure and is silicone based, and they promote it for general purpose vacuum sealing and sealing laser windows. Works great on some x-ray tubes I build at work - several years of pinch-off with no degradation yet.
The laser tube is about 50% more complicated than a HeNe system, but gives much more output. The gain of the brightest lines is about 5%/meter, so surplus argon laser mirrors work very well. As far as I can tell, the only reason the Se laser never became commercially viable is that the lines are so close together that it is difficult to select just one with a prism selector. It has a really nice green line.
(From: Steve.)
I've got something to cure the separation problem, a PCAOM. That stands for Polychromatic Acousto-Optic Modulator, it's really a sort of AOTF that can amplitude modulate up to 8 lines from a given laser. We use them for color control on laser shows. With a little work, it could be reprogrammed for the HeSe lines. BTW, it keeps all the lines colinear so you don't need huge arrays of mirror mounts and dichroic optics to do color separation and recombination. You shoot a white beam in and get controlled colors out and a waste beam. It can control lines as close together as 3 nm. Cost is about $1,800 for the 4 channel and $3,500 for the 8 channel units.
(From: Mark.)
I have been planning to write up the project but have been very busy lately. There has been enough interest that I will try to get a complete description and photos on Sam Goldwasser's laser FAQ site as soon as I can. I will alert this newsgroup when I have it completed. Wish I could get alt.laser - my ISP doesn't provide it.
(From: Chris.)
Now your making me feel guilty. I've got some CO2 photos, but I don't know when I can get them scanned. So damn busy. This HeSe is very interesting. You are making me think about doing that instead of HeCd.
What kind of total output power are you getting? I envision 10 to 50 mW, right?
(From: Mark.)
I got a little over 100 mW, multi-line, multi-mode. It was enough that focussed with a lens, I could melt holes in my black Swiss Army knife. A new standard for laser output?? :)
There is a good survey article on metal vapor ion lasers in Scientific American back in the early 70's. They talk quite a bit about the HeSe laser, and give journal references for some of the important work.
(From: Chris.)
Thanks, I'll be checking that out some time.
(From: Keith (Thallium204@aol.com).)
I'm sure you know that thallium is named after its strong green spectral line. It just happens that it has a very low melting/vapor point, and like mercury needs no buffer gas to lase, it just needs a vacuum. It is not very critical of vapor pressure as in the HeCd and HeSe. The use of argon or neon as a buffer does make it more efficient but is not absolutely needed. The mirrors from a dead green HeNe laser tube work fine (as long as their radius of curvature is consistent with stability for the resonator).
Without knowing the required reflectance and radius of curvature (RoC), the following comments can only be very general.
Nothing off-the-shelf is close in the required diameter, RoC, transmission, and wavelength except maybe some doubled YAG optics for the green line and a few argon ion high reflectors. Dye laser OCs cut for the 620 nm region might be a option but they usually have a very short radius and a 5% transmission which may be too much. You might try two Newport or Thorlabs flats with BB or BD coatings (both broad-band high reflectors) to get going as a test and take your beam off the Brewster reflection. That would set you back $80 to $150 for two flats at 1" diameter. Flat to Flat is very hard to align. At least one mirror should have say a 2 to 4 meter RoC. Even a BD1 coating is going to leak some small amount of beam out if your lasing and the back of the optic is not frosted. But it would let you see how much gain you have. The Chinese and the Bulgarians will run custom coating lots at around $1000 to $2000 a lot if you supply the blanks and your specs are not too tight. US coating runs are something like 2 to 5 times that when I've looked into it.
There was even a low cost ($500) "educational version" of such a laser. It had a peak output power of less than 1 mW but at the time (pre-1990), was by far the cheapest green laser available. See Photo of QE Technology EL-100 Hg:N2 Laser (Courtesy of Chad Andersen). As can be seen, there isn't much in there besides the laser tube with its mercury reservoir, a pair of UV lamps with a couple of electronic components to run them, and the cavity reflector. Apparently, the rear mirror (under the black thing at the rear?) could be adjusted to get all sorts of higher order spatial modes. :) Although these are 20 year old lasers with a soft-seal tube, the mercury will oxidize and combine with any oxygen that should leak in, so if the lamps work, the laser may still lase.
(From Steve Roberts (osteven@akrobiz.com).)
The relevant patent is: U.S. Patent #3,992,683: Optically pumped collision laser in Hg at 546.1 nm.
There have been a number of journal articles including:
Abstract: The kinetic model, output power, and frequency spectrum are described for the 546.1-nm at. Hg laser. The laser is optically pumped by a Hg arc lamp at 2 different wavelengths and N2 is used both to link the pump stages and to drain the lower laser level by inelastic collisions. The laser construction and cycle are described.
Abstract: The operating characteristics of continuous-wave (CW) at. Hg lasers are described. Operating at 546.1 nm, the laser is optically pumped by d.c. Hg discharge lamps. The laser has high perceptual brightness and potentially low noise and high stability. Experiment and theory aspects of the laser and its performance are discussed, and the effectiveness of various optical coupling schemes is considered. Evidence for absorption loss due to Hg dimers is presented, and some results are given for direct measurement of this loss by tunable dye laser.
You can buy this from the U.S. National Technical Information Service. It will be the final report of the research project, and probably quite detailed instructions on how to build it.
