Homemade Manganese Vapor Laser

In this 10th anniversary article, I will describe what is a logical consequence, a continuation of my independent laser industry. After constructing a power source that is suitable for pumping pulsed metal vapor lasers and gaining experience with ready-made active elements of copper vapor lasers and its compounds, it remained only to fabricate the active laser element (hereinafter - AE) completely independently, with a new working medium.



The following requirements were imposed on the choice of a working medium: radiation in the visible region, relatively high attainable power, acceptable efficiency, easy accessibility and low cost, not too high operating temperature.

Laser generation was obtained for almost all possible chemical elements, but only a few units meet the above conditions. Oddly enough, the best choice is all the same copper and its compounds, but for me this is already a largely passed stage. There are still a number of metals as candidates that have a more or less decent laser generation efficiency under the excitation regime similar to that for copper vapor lasers: barium, strontium, manganese, lead, gold. Barium and strontium immediately disappear, because their generation lines lie in the IR range, which is boring and ugly. Remains manganese, lead and gold. Gold also drops out due to the high cost and too high working temperature (1600+ degrees Celsius). Only 2 candidates remained - lead and manganese.Both metals are cheap and should be relatively affordable - lead can be found literally under your feet, gutting a battery or a piece of armored high-voltage cable selected in the trash, manganese is widely used in ferrous metallurgy and is produced in colossal volumes.

Manganese is able to generate at a wavelength of 534 nm (the main line) and several more strong lines in IR and about 3-4 weak green ones. The achievable power for a well-made AE ​​and a carefully tuned power supply exceeds 10 W in total for all radiation lines, the operating temperature is in the region of 1000-1100 degrees, which is significantly lower than that of copper. The optimal pulse repetition rate varies depending on other experimental conditions and must be selected. In any case, it fits into the “standard” range of 5-15 kHz. In lead, the main generation line lies in the borderline with the IR range of 722 nm and there is another transition with a wavelength of about 405 nm, an achievable power of units of W for the 722 nm line. A feature of this environment is a record gain of 600 dB / m.The operating temperature is 800-900 degrees, and the optimal chisi according to various literature data below 10 kHz. The optimal buffer gas for both metals is helium, although argon can also be used.

I decided to stop my choice on manganese based on the above data - the generation efficiency is clearly higher, and the wavelength is clearly more pleasant, very close to the usual "reference" 532 nm. After that, it was necessary to determine the design of the active element. A lot of materials on manganese lasers were found in scientific journals like “Quantum Electronics” and “Instruments for Scientific Research”, which are available on the Internet or directly, as in the case of CE or via the hub (for the English version of PNI). From the reviewed articles, it became clear that generation is possible for almost any size of the discharge channel and in a sufficiently wide range of excitation conditions, while the optimal excitation mode and buffer gas pressure must be selected for a specific AE. After which I drew a rough sketch of my future AE,based on the available parts, and proceeded to manufacture.



Looking ahead, I’ll say that it was unexpectedly difficult to find manganese for this laser, and I will describe all my ordeals in this area in a separate spinoff article.

So, the basis of the active element of the laser is a ceramic tube with an inner diameter of 14 mm and a length of 800 mm, which is the discharge channel. Using bushings made of aerated concrete, it is fixed in a large diameter quartz tube. The space between the walls of the tubes is filled with thermal insulation. As the first option for such thermal insulation, I decided to try using a thin corundum sand, as in the early copper vapor lasers.







After assembling the discharge channel with the casing and high-temperature thermal insulation, such a device was obtained.





At the ends of the outer pipe-casing, electrode assemblies are fixed due to the rubber seals squeezed by the flanges. The electrode assemblies are flange heads machined from aluminum. And for their manufacture, it was first necessary to cast the blanks.



After turning, these are the details.





From the inside of the heads, an anode in the form of a tube of small diameter and a cone-shaped cathode are screwed onto the thread. Both electrodes were machined from a non-magnetic stainless steel





Aluminum heads are equipped with fins for air cooling. The resonator mirrors are attached to the ends of the electrode heads by small flanges. Rubber rings between the mirrors and the electrode head act as a vacuum seal. At the same time, they provide some mobility of the mirrors for their alignment. From the sides on the electrode heads there are fittings for pumping and gas inlet. Gas dosage is performed using the same insulin needle glued into one of the fittings. Thus, a completely coaxial AE design resembling a factory one was obtained. Dimensions of parts can be seen on the sketches that I drew before turning.





After manufacturing all the parts, I performed a test assembly of the AE, which immediately revealed "childhood diseases." Here, the electrodes themselves are not yet installed.







First of all, it was necessary to reject sand as insulation. At the first pumping out of the tube, the air enclosed in the voids inside the backfill began to loosen it, throwing large amounts of sand into those places where it should not be, including drawing it into a vacuum pump, which did not benefit him in any way. The solution was found as a replacement for sand with ceramic wool. The air was already leaving the cotton unhindered during pumping.



Another unexpected difficulty was the extreme fragility of the entire structure. 2 assembling and disassembling procedures out of 3 ended with chipping the edge of the quartz tube at the time of tightening the flanges of the electrode heads, although the wall of the tube seemed to be thick. A solution to this problem would require a radical alteration of the heads and a method of attachment, which so far has been decided not to do, since the design is experimental. During the assembly process, I added another element - an aluminum tube around the entire AE, which acts as a reverse current path and reduces the parasitic inductance of the AE. So that nothing would break under the weight of this pipe, he added support from the opposite end.





Thus, the AE was ready for test pumping and the first inclusion of the discharge in it. But at this point I did not have enough connecting cable. For its manufacture, I used the same type of coaxial cable as for the copper vapor laser and a similar connector from LGI21, which had to be modified. After that, it was possible for the first time to turn on the discharge in it and try to enter the operating temperature regime while idling. As a working gas, I used argon at a pressure of the order of 10 Torr. Another flaw came up here - a large amount of discharge energy flowed down through the support of the return current conductor to the ground through the pipe wall, causing it to heat up strongly, and less energy than could be delivered to the “target discharge”.





In addition, one internal insulation of the discharge channel was not enough. I had to disassemble the AE again and move this support in the opposite direction, and fill the space between the quartz case and the return conductor with additional thermal insulation from the same ceramic wool. In place of the metal support, where it was before, installed a sleeve of aerated concrete. Thus, energy leakage and heat loss were eliminated.





When assembling, the edge of the pipe was traditionally chopped, as a result of which, with each assembly, the entire laser AE was slightly shortened. In the end, I managed to assemble it, and I was able to start a full-scale training without working substance. The main objective of such a training is to allow the AE to warm up to operating temperature and above, to completely burn out all the remains of volatile impurities which are literally teeming with both the ceramic tube of the discharge channel and the internal high-temperature insulation, especially after inadvertently the oil from the vacuum pump gets into the tube. To prevent this from happening again, I put an oil trap in the form of a classic Drexel flushing into the gap in the vacuum hose. Oddly enough, there is no vacuum leak on a conventional chemical section. The whitish color of the discharge indicates the release of impurities.The training had to be extended to several hours, so that all the remaining oil evaporated and burned and the discharge acquired a normal pinkish-blue color, characteristic of argon. During the training, the AE was able to be heated to operating temperature, and the heating of the electrode heads was unexpectedly moderate, despite the fact that the anode and cathode were red-hot.
Training process:





By the end of the training, the color of the discharge became the characteristic color for argon.



The discharge channel heated very viciously, like the electrodes. The glow is visible even through thermal insulation.





After training, the tube was filled with argon to atmospheric pressure, the fittings were closed, and it remained in this form.

In parallel, I was looking for the actual working environment of the laser - manganese. This turned out to be unexpected difficulties, all the familiar chemists made a helpless gesture and did not know where to get manganese, Internet merchants who had manganese on sale, sold only in barrels-wagon-ships, and its purity was far from “laser” - 95% of the main substances. And also on the Internet there were unique people trying to sell 200 grams of 99% manganese at the price of a kilogram of silver, and even with the demand "wait a month until they deliver from Germany." There was no metal manganese in the local chemical reagents store either, but there were salts of it, from which I made an attempt to obtain metal manganese myself. More on this will be in the corresponding article. All my tribulations suddenly endedwhen a small king of high-purity manganese came from the creator of the website PeriodicTable by mail, for which he is very grateful.



After that, it remained to make quartz boats for the working substance and push them into the discharge channel. The boats had to be ordered in a glass-blowing workshop. When they were ready, I separated small pieces from the king, placed them in the boats, and the boats themselves easily slid into the discharge channel through the holes of the laser mirrors.



Then the mirrors were returned to the place and tests began.

After the laser tube was pumped out to the maximum vacuum, I aligned the mirrors, they must be aligned under vacuum, since otherwise the alignment will fail when the seals are deformed. By the way, I forgot to say that aluminized deaf mirror and plane-parallel quartz plate as an output window are used as mirrors. The mirrors were adjusted using the LGN-109 school helium-neon laser.

Laser before starting.



Gas system.



Then I launched a small stream of argon at a pressure of 10 Torr. I decided to “accelerate” the laser with argon, since there was quite a bit of helium. At this pressure, the discharge ignited easily and the laser started to warm up. He warmed up quite quickly, and when the discharge channel acquired a dark orange glow, a change in the discharge color from pink to turquoise became noticeable.



Then, in the center of the spot from the discharge glow, a bright green spot appeared in which a shadow was guessed from the boats embedded in the discharge channel, which partially blocked the channel’s light aperture. Generation has been received!



Following argon, helium was introduced into the AE, and argon was shut off, which immediately increased the generation power several times, and by selecting helium pressure, it reached a maximum power. The selection of the pump mode (voltage, ChSI) made it possible to keep the stationary thermal regime of the laser and a little look at the radiation. A bright, well-visible, poisonous-green laser beam of irregular shape poured from the exit window due to boats blocking the aperture.









At the same time, there is enough power to carbonize wood and cardboard when focusing. Due to the useless shape of the beam, it focused rather poorly. Considering that a poorly focused beam was able to carbonize plywood, I would estimate its power at a value of about 1 W, but no more.









After playing enough and taking a picture, I turned off the laser and let it cool. After this experiment, several conclusions could be drawn. The main conclusion - the placement of the working substance in boats - is a so-so idea. The light aperture suffers very much and up to 70% of the potential remains unrealized. Not for nothing in copper vapor lasers, the discharge channel is made of a variable cross section with sections of a larger diameter. It is precisely in them that the working fluid is located, and the channel aperture is limited to narrower sections of the tube. If possible, the diameter difference is made as small as possible. The creation of a well-functioning “working fluid vapor generator” has become a separate non-trivial task in the development of industrial AEs, the solution of which required the manufacture of a bunch of different prototypes. In my conditions it was easier to put up with the loss of part of the power,but you can leave the boats if you wish. The second unexpected surprise for me was that when disassembling the laser in the discharge channel, little was left of the boats. It turned out that molten manganese dissolves quartz. Thus, the more you need to get rid of the boats. But if you do not get rid, then at least make them from another material, for example leucosapphire. It seems that this is the only option for such aggressive conditions. Characteristically, it is easy to get a suitable workpiece - sodium lamp burners are made from leucosapphire. Well, the mounting of the electrode heads needs to be redone, as further disassembly-assembly operations have rendered the laser case unusable. As for the rest, I can call this laser justifiably exclusive in my practice, since not a single enterprise produced any manganese lasers in the former USSR,nor in the world. The manganese laser never left the walls of the laboratories. Why - I do not understand. Perhaps because its wavelength is very close to the standard 532 nm, which are obtained using solid-state lasers and frequency doubling, which do not need a complex high-voltage power supply and a fragile laser tube with a limited life. And where diffraction divergence of radiation with high power and also at visible wavelengths is needed - there are already copper vapor lasers, in which the efficiency is 5 times higher. Nevertheless, I would say that the manganese laser is undeservedly forgotten and could very well be commercially available before the DPSS era. Also, the thallium vapor laser, which generates radiation at a wavelength of 535 nm, did not leave the walls of the laboratories. But no, I won’t do it,since why work with an extremely highly hazardous substance and wildly complex pumping when there is much easier to use manganese with a wavelength of 534 nm? But I tried lead, but there will also be a separate note about this side experiment.

Oh yes. If this article (like all my others) deserve it, then you can send a completely voluntary donate dogecoin “amy to the following address: DLLNGqSeyXo786jiurzMAnj8USwaU5u9xY