Hi Xing, you're welcome. Have something to add after you have updated the information. I wouldn't take the responsibility of the cleanroom manager, but I would strongly recommend to stay well below 200C during baking. Normally, baking temperatures are set between 120-150C, and baking is used in UHV systems opened/vented on a regular basis, mostly in order to remove water film from the walls of the chamber. However, any kind of temperature treatment has its effect on the system, due to a different thermal expansion coefficient of the parts of your chamber (viton sealing rings, if any; copper gaskets; stainless steel; glass in the viewports/windows, screws, etc...) The most dangerous part, to my experience (which might be pretty obvious, though) is the glass viewports/windows. These MUST be properly covered with Al foil, and MUST NOT be under the heating tape. It is better to try to avoid having the heating tape over the screws, holding windows, or directly on the window flanges. A "hot spot" might provoke an expansion of the screw, or non-uniform expansion in the flange, which in turn would cause some stress on the window and is capable of braking it. I never experienced that myself, but I saw the examples. The other thing is that if you overheat the flanges, sealed with copper gaskets, the copper would expand, then contract - upon cooling, and might develop a leak. Not to say about Viton and other rubber-based sealings - Viton survives, I guess, smth below 200C, but not sure about your vacuum system. Also cover all the ceramic connections on the chamber with foil - again, just in case of having a hot spot close to it.

Returning to your question, there is no preference in wrapping your system with Al foil to using heat blankets. Just make sure, you have as little holes in your blankets assembly as possible. Do not forget to cover windows with foil, prior to baking. Same applies to the bellows' (if any) and ceramic feedthrough's. Do not plug your heating tape directly in the wall plug, use a voltage transformer to increase the voltage on the heating tapes slowly. It is always a good habit to ramp the voltage on the heating elements, hence you avoid the thermal shock of your system.

You may try to heat the entire chamber with thermoresistive element in order to improve the desorption of species from the walls.

If you have a Ti crucible you can also evaporate a little bit of it as it acts as a getter.

There is a good answer above. However, I'd try to add some words here.

First, what is the state of the roughing (preliminary) pump for your turbo? You might want to check the vacuum on the EXHAUST of turbo (remember that it has to be in the range of at least low E-1 - E-2 mbar). Your roughing pump (pre-pump) might be damaged as well. Second, if the pre-pump stopped as well (which I suggest was the case), and if it is an OIL pump, you might want to check oil/hydrocarbon contamination in your vacuum line. This is a very often issue for the vacuum/UHV systems during powercuts. In order to fight this, there are normally oil traps, filled with zeolite, put in between the exhaust of the turbo and inlet of the pre-pump. However, if you use oil-free roughing pump, this wouldn't be the case. Just check that the pre-pump delivers good vaccum (this could be done by a Pirani gauge with KF fitting, like this:

http://www.directindustry.com/prod/thyracont-vacuum-instruments-gmbh/pirani-vacuum-gauges-39352-556606.html)

If the pre-vacuum is not the case, you take a look at the power consumption of your turbo. I do not know neither your system, nor the leakage you have in it, but I'd suggest, from my own experience, that the power consumption of the normally operating turbo should be in the range of 7-17 Watts (you can see it on the control unit of the turbo). If it approaches 15-17 W, I would strongly recommend performing leak test, or, at least, checking all the vacuum-tight connections: flanges, T-pieces, etc. If this is not the case either, then you vacuum chamber is full of shit, which your turbo is not capable to remove. Hence you either wait a couple of months for the pressure to restore, or you do a proper bakeout (remove all plastic connections, electronics, and everything from the chamber that won't heating to 120-150C; don't forget to cover the windows and viewports with foil), and follow the Guillaume's advice on Ti sublimation. That's pretty much it, hope that helps.

This happened to me a few years ago; I mis-diagnosed the problem and did a bake-out. However, oil from the backing pump had gotten into the chamber so the bake-out simply baked the oil onto the surfaces!

In the case of oil you must dismantle everything and perform a thorough cleaning. For some items like tubing just replace it. You can clean stainless steel parts by mixing clean water with "Mr Clean" - the older formula, which has NaOH - wear rubber gloves! For small parts an ultrasonic cleaning for 30 minutes or so will do the job. Then rinse with plenty of water (Mr. Clean is water soluble) and then follow your usual vacuum cleaning procedures. For larger items you can simply scrub the surfaces with a lint-free rag, and then spot-check with an acetone dipped swab for any remaining debris. This took a couple of days to clean everything, then rinse with purified water, then scrub again with acetone, then rinse with methanol.

When reassembled, baked out, and pumped down everything worked fine. I did have to send the turbo pump out for cleaning (if it has oil you must tell them!), and also had the ion pump serviced. I had to junk my RGA head.

Note that copper and aluminum will be pitted by Mr Clean, so those items require a different procedure if exposed to oil.

PS: I got the "Mr Clean" method from an electron microscope customer service engineer.

Thank you. You guys are awesome. I talked to our cleanroom manager and it looks like we need to bake out the system to the extreme. Since heating tape is only 3 meters it is not effective to heat a giant chamber with half a meter diameter, would a heating blanket suits this job? The option that I can find can heat up to 270 C degrees. Will that be high enough temperature to bake out the chamber? (it is 12 inch by 36 inch large blanket)

Just cover it all with aluminum foil (the cheap kind!), then apply the heater belt(s), making sure that they don't cross each other - else they will burn out - then cover loosely with more aluminum foil.

You can buy extra heater belts, or borrow some from anybody that runs a vacuum chamber. They are not expensive at all.

Just another suggestion. What I used to find with my chamber (similar size to yours) was used for dc sputtering. I used to find that after bake out I could get 10^-7 mbar easily, but then when I started sputtering, say for 30 minutes with no aim of making a film, the sputtering process would remove any excess impurities in the chamber and my vacuum would improve to around 5 x 10^-9 mbar. I had the ability to load samples while still under vacuum, so if you have this option I would recommend using the e-beam just to remove impurities before you begin making your thin films.

Hi Xing, you're welcome. Have something to add after you have updated the information. I wouldn't take the responsibility of the cleanroom manager, but I would strongly recommend to stay well below 200C during baking. Normally, baking temperatures are set between 120-150C, and baking is used in UHV systems opened/vented on a regular basis, mostly in order to remove water film from the walls of the chamber. However, any kind of temperature treatment has its effect on the system, due to a different thermal expansion coefficient of the parts of your chamber (viton sealing rings, if any; copper gaskets; stainless steel; glass in the viewports/windows, screws, etc...) The most dangerous part, to my experience (which might be pretty obvious, though) is the glass viewports/windows. These MUST be properly covered with Al foil, and MUST NOT be under the heating tape. It is better to try to avoid having the heating tape over the screws, holding windows, or directly on the window flanges. A "hot spot" might provoke an expansion of the screw, or non-uniform expansion in the flange, which in turn would cause some stress on the window and is capable of braking it. I never experienced that myself, but I saw the examples. The other thing is that if you overheat the flanges, sealed with copper gaskets, the copper would expand, then contract - upon cooling, and might develop a leak. Not to say about Viton and other rubber-based sealings - Viton survives, I guess, smth below 200C, but not sure about your vacuum system. Also cover all the ceramic connections on the chamber with foil - again, just in case of having a hot spot close to it.

Returning to your question, there is no preference in wrapping your system with Al foil to using heat blankets. Just make sure, you have as little holes in your blankets assembly as possible. Do not forget to cover windows with foil, prior to baking. Same applies to the bellows' (if any) and ceramic feedthrough's. Do not plug your heating tape directly in the wall plug, use a voltage transformer to increase the voltage on the heating tapes slowly. It is always a good habit to ramp the voltage on the heating elements, hence you avoid the thermal shock of your system.

Have you leak-checked it? Do you have access to an RGA to determine what the residual pressure consists of?

Baking will only accelerate outgassing of adsorbed water - that's all it does. It would only be useful if your pressure has been slowly getting better over the days. If it hasn't been doing that, I don't think baking will achieve much.

By using a RGA (if you have one) you can decide the better strategy to improve your vacuum. Water? Bake out will make the job (though I would not exceed 120C, and if you have KF fittings with rubber/viton o-rings, I'd stay around 80C). Oxygen can be absorbed by evaporating some Ti (or Nb) from your ebeam. Water can also come from a leaking water cooling pipe inside the chamber... but if the ebeam was not damaged,nor recently used, it's unlikely.

Chromium does better than Titanium.. both are good..

I notice you say that the turbo had been returned for service. Do you have the service notes? Either through these, or contact with the company, it may be possible to assess whether pump oil has been drawn through to the chamber. In the case that it has, a thorough clean prior to baking is required, due to the reasons described in the answers above.

There are VERY GOOD advice(s) on using RGA/mass-spectrometry/performing direct leak-test with He for understanding the origin of poor vacuum... It was so self-evident to me, that I didn't even think of that :) Anyways, thumbs up for you, guys!

Thanks guys! I really learned a lot from here. I will talk to the manager and see what we can get from the He leak test.