Good UHV Pressure

There are several ways to get good pressure in your system. A lot of it depends on...

1. Quality/strength/size of your pumps
2. Size/surface area of the chamber
3. Access of your pumps to the chamber (i.e. size of cross section)
4. Pump time (hours vs days vs months)

The best thing to really ever do is a nice, long bakeout. However, with a small chamber, medium sized turbo pump, and a long pump time, I've been able to reach low 10^-9 torr without baking. Again, it depends on your system.

Good General Practice

"The problem with going down to UHV is the desorption of the water vapor that sticks to the chamber walls when you open it to the atmosphere. If you check the tables of vapor pressure, you will see that for water at room temperature it lies around 10-8 mbar. This is why you need to bake the system out, to force the quick desorption of that water.

So, if you need to speed up the pumping the only way is to prevent or reduce as much as possible the adsorption of water inside your chamber. There's a number of strategies to do that, although none will solve the problem completely" - Juan Jose de Miguel

To deal with this, we have a few steps of good general practice:

1. Vent your chamber to N2 instead of air. If possible, you can even keep a low pressure of N2 gas flowing through your exposed system to reduce adsorbants. Ideally, you should use low humidity N2, but this may be out of your control.
2. Heat a little bit as venting to reduce adsorption of air
3. Keep your chamber vented for as short a time as possible. Get most of your work done before venting and/or keep your chamber at least scroll/rough pumped.
4. Install a loadlock to have one subchamber with so-so pressure that accesses the chamber with really good pressure for a short amount of time. This is mostly used for sample transfer.
5. Coat your chamber's inside with something (i.e. gold film) to passivate the surface, thus reducing adsorption.
6. Use a carousel or garage inside your chamber to store several samples so that you don't have to vent.
7. Use a copper gasket (rather than a rubber one) to really get a good seal from ambient. This my require your system to use different flanges, but is worth investing it. Some people use a copper gasket 2 or 3 times before replacing it, but for best pressure practices, a new copper gasket should be used every time.

More good discussion of this topic can be seen here.

Getting to 10^-7 Torr

Some systems have temperature limits, and bakes can be unnecessarily difficult. In this case, there are still a few tricks you can try. You should be able to access 10^-7 torr pressures with a decent turbo pump on your chamber. If not, you might have a leak or need stronger pumps.

Getting to 10^-8 Torr

From 10^-7, I've found that even a light bake at 80C (cheap system here, or a cheap variac with ebay heat tapes)  and long pumping can move you into the 10^-8 range with just a turbo. Again, the caveat is.. it really depends on the 4 parameters mentioned at the very beginning of this post.

In addition, Dr. Faebian has an excellent blog, and in one post he explains some N2 magic for flushing your chamber and decreasing your pressure. In summary...

"Evaporation will transport moisture from the chamber moist inner surfaces into the dry N2 gas. Not all, but a significant fraction. The evaporation rate will depend on:

The concentration of water in the N2: which is zero at the beginning
Flow rate: initially high as the N2 penetrates the vacuum, but constatly dropping
Pressure: evaporation happens faster at lower pressure
Temperature: evaporation happens faster at higher temperature

Simply let the N2 in until ..500 mBar (remember the process happens faster at lower pressure). Wait a minute. Then pump the chamber down with the scroll pump again. Once it has reached 0.05 mBar, turn the pump off and flush it with dry N2 again. On the second pump down, open the As and P cracker’s needle valve manually a whole turn. The bulk should have been protected from atmosphere when you shut the valve during Essential Maintenance: System Venting. It should therefore be free from moisture, but there is always a chance the valve has a little leak. To be safe, we will pump down the bulk with the scroll pump now.  To increase the effectiveness of the N2 flush you may consider heating the N2 gas to ~50°C. This can be highly regulated with an inline gas heater or poorly regulated by baking the SS N2 pipework to 50-70°C with heat wraps during the flushing phase"

After repeating the vent and (rough) pump 5 times, you should be able to get to 10^-6, but by adding a turbo, this should get you to 10^-7. Now, if you leave this overnight pumping, you should see 10^-8 by morning.

Getting to 10^-9 Torr and beyond

Once you've hit the 10^-8 range, you can start using ion pumps and TSPs to do some dirty work for you. Turning an ion pump should do quite a lot. For the TSP, you should probably run a few cycles. Some controllers allow you to program the TSP to cycle every hour or so for a set period of time. Both pumps are powerful, and over time, can get you to pressures as good at -11.

• "An ion pump (also referred to as a sputter ion pump) is a type of vacuum pump capable of reaching pressures as low as 10−11 mbar under ideal conditions. An ion pump ionizes gas within the vessel it is attached to and employs a strong electrical potential, typically 3—7 kV, which allows the ions to accelerate into and be captured by a solid electrode and its residue." - Wikipedia
• "It consists of a titanium filament through which a high current (typically around 40 Amps) is passed periodically. This current causes the filament to reach the sublimation temperature of titanium, and hence the surrounding chamber walls become coated with a thin film of clean titanium. Since clean titanium is very reactive, components of the residual gas in the chamber which collide with the chamber wall are likely to react and to form a stable, solid product. Thus the gas pressure in the chamber is reduced. But after some time, the titanium film will no longer be clean and hence the effectiveness of the pump is reduced. Therefore, after a certain time, the titanium filament should be heated again, and a new film of titanium re-deposited on the chamber wall." -Wikipedia

Remember, even the best of systems can have a small leak. This can be extremely difficult to find, and typically you will need to use a mass spec to do a helium leak check. Without a completely sealed system, the rate of leak will counteract even the best of pumping. This post may make it seem like you never need to bake, but honestly to reach pressures of 10^-9 and better, you will typically need to bake. Sometimes, if you have an accident vent/leak, you can still recover good pressure, but a real vent for repair and such (without use of a load lock) will usually require a bake of a day or two at 120-150C.

Helium Leak Check

Your pressure needs to be in a good enough range to use your mass spec (~10^-6 torr).

1. Turn on your mass spec.
2. Enter leak check mode.
3. Start He gas flow. You want to use a small tube or even a syringe needle to leak a gas flow just above ambient (when you can barely hear gas hissing out).
4. Leak He gas into possible areas of leaking in your chamber (flanges, bellows, valves, ceramic feedthroughs, water cooling lines, etc...). If your chamber pressure was better before, you should concentrate your leak check to areas that have been added or changed. Watch for the Helium signal to increase. When the signal increases, this corrolates to a leak being at the location you are leaking you are leaking He at. Possibly, if your system is large, there may be a delay. Because of this, leave enough time leaking at a particular location to identify if there is a delayed He signal increase.
5. If this fails, but you are sure there is a leak, try using the balloon method: using plastic bags (grocery, trash, etc), section off all areas of your chamber into different bags using tape to complete the seal. Poke your He leak into the bag and balloon the gas into the bag. If there is no mass spec signal increase, move on to the next balloon. If you see a signal increase with a particular balloon, then you have narrowed down your area of search. You know the leak is contained somewhere in there.

More detailed explanation here.

Baking

-still in work-

Loadlock Chamber

-still in work-

Working with UHV and Turbo Pumps

One of the best general introductions I've ever found.

(Also here.)

Making a Sample Filament (STM)

The wire used for the filament is in the wire drawer downstairs (it's on a green spool labeled "STM wire") and should be .1mm thick.

1. Find a small rod. Typically, I've used one of the needles from upstairs to wind the wire around.
2. You want to coil the wire so that you can generate a lot of heat in a localized place. The final length should be a little more than an inch long, but use the old filament as a guide for length.
3. There should be a clip on the back side of the sample holder that you can slip your newly made filament through. This will pin down the filament close to the sample at half its length.
4. Spot weld each end of the filament to the cylinders on the sample holder (these are the ones that the fork slide through).

Optional: Use an electrical feedthrough flange to connect to the sample holder. Test the filament in UHV by using the pump cart and a thermocouple (place on the side of the copper for good contact and appropriate measurement). Then you can know how hot the sample gets at varying power.

Opening the STM Shields

Notes from Wenhao Lu.

Parts update from after machining the inner shield.

STM Electrode Test

The electrode test tells you if you electronic components are working correctly. The principle is that by stimulating the inner electrode of a piezo with a square wave signal from the oscilloscope, you can see a square-like charge/discharge response (~2V) on the outer electrodes of the same piezo.

• Because all of the outer piezos have their inner electrodes connected, you can stimulate their common input ABCZ and check the response of each outer piezo's outer electrodes (AX-, AX+, AY-, AY+; BX-, BX+, BY-, BY+; CX-, CX+, CY-, CY+). 
• Similarly, you can stimulate the inner piezo's inner electrode Z, and measure the inner piezo's outer electrodes. These outer electrodes are daisy chained together inside the STM, and so you can measure the X-X+Y-Y+ BNC plug.
• If you stimulate the inner piezo's inner electrode (Z) and measure the outer pizeos' outer electrodes (AX-, AX+, AY-, AY+; BX-, BX+, BY-, BY+; CX-, CX+, CY-, CY+), you should see no response (in reality a smaller response of ~300mV).
• Conversly, you should see the same response (or lack thereof) if you stimulate the outer piezos' inner electrode (ABCZ) and measure the inner piezo's outer electrodes (X-X+Y-Y+).
• (Optional) If you stimulate an outer electrode and measure the other outer electrodes on the same piezo, you'll see a square-ish charge/discharge response that's smaller than if I stimulate the piezo's inner electrode.

STM Wiring

Here are a few different diagrams that show how the STM is currently wired up. I'd suggest looking at the powerpoint so that you can see things in higher resolution and move object around or even un-crop pictures to get a good idea of how things are placed.

Wiring is made using a very thin insulated wire (actually pacemaker wire use in heart monitors). You will need to either scrape or burn away the insulation (using a lighter). Then you will need to crimp connectors on the end. I use a method that involves sandwiching the exposed wire between two connector pieces, but play around with what works for you.

(Left: Original connection with insulation. Right: New connector with a front piece that can insert into a socket and a back piece that pushes the wire in for electrical contact, clamping it in place.)

A note on the Z electrode (inner piezo's inner electrode): The original socket for this somehow fell out. I used a worked around this issues by creating a connector piece with some copper braiding sticking out of the end. I made an extra, and it's in the STM wire box in case you need it later. (But you will still need to crimp that piece to some wire.)

A note on the bias electrode: This broke off during repairs and had to be reattached with silver paste (must be UHV safe). Stuff like this happens, and you can figure out from instructions online how to use the silver paste, I just thought I'd include a snap of the setup I used for curing. I used a solder iron to locally cure the epoxy, propping up the iron with clamps and clips.

Etching an STM Tip

1. Set up a 1M solution of KOH (.5 L of DI water and 56 g of KOH pellets).
2. Find tungsten wire of a diameter that matches your tip holder.
3. You want to make a stand with some sort of fine adjustment for holding the tungsten wire.
4. Using an AC power supply, connect one voltage to the wire you will etch. Connect the other to a metal/wire loop you will dip in solution.
5. Barely just dip with tip into the solution by using the fine adjustment and back out of the solution a bit so that the tip is just barely submersed in the surface (pulled upwards via capillary action).
6. Turn the voltage on (4-14V, but you can play around with this to see what works) and leave for about 4 minutes.
7. When the wire breaks, turn off the voltage. Collect the bottom tip using tweezer to pick it up from the solution.  Snip the top tip off with cutters.

Setup (from http://goliath.emt.inrs.ca/surfsci/docs/sc04/Olfert.pdf)

More information than you could ever really want here.

Opening the STM Chamber

1. Turn off the STM controller box completely (software off, HV off, LV off, capacitors off).






2. Turn off other items such as mass spec or sputter gun controller.
3. Turn off both ion pumps (located on the bottom shelf of the cart).
   
4. Turn off both ion gauges (flip off back switch on controller (top of cart)).
   
   
   
   
5. Close the turbo/chamber valve and the turbo/rough valve (both are electrically controlled so just unplug the appropriate power cords).
6. Turn off the turbo (unplug) and allow to spin down.
7. Turn off the rough pump.
8. Give some time to allow filaments to cool.
9. Leak a low pressure of nitrogen gas into the chamber via the vent valve. The bellow below the STM will deflate appropriately when the chamber reaches ambient pressure.
10. Open the bottom flange to unhook the STM chains. Cover with foil when not in use.
    
11. Disconnect the electrical connectors atop the STM.
    
12. Open the STM flange on the table.
13. Use three wire cables and attach to the ceiling crank. Slowly lift up the STM and place to rest on the STM holder.

Extra Notes: Always store the bolts and nuts in a place so that you will not have to search later to replace them. When possible, cover opened flanges with foil so as to minimize chamber contamination.

Dosing Molybdenum for MoS2

1. Prepare the Evaporator
• Open the turbo corner valve and pump the evaporator with the pump cart. You will have two turbos in series.
• Overnight, bake the evaporator at 70-80 V to reach >100 C. Be careful with the baking tape and make sure to wrap the evaporator with fiber glass.
• The next morning, stop the bake, set up the water cooling lines, and degas the filament for the evaporator. You should go to 5.6 A and 5 V for 15 minutes. At the same time, make sure to watch the temperature and pressure. You do not want to exceed 150 C or overpower the turbos.
2. Dummy Dose
• Insert the thickness monitor so that it is in front of the evaporator. Twist the knob until sleeve reaches the label which marks the correct position. You can double check the location is correct by opening the evaporator valve and looking through the view port on the other side.
• Set the thickness monitor controller to have a density setting at 10.2 and a Z-factor setting at .25. Watch the monitor a bit to check that it is stable.
• Apply the 3kV high voltage power supply. This is an external unit which is home built and directly connected to the evaporator source rod.
• Increase the filament current to around 5 A such that the dosing rate is ~2 angstroms/min. The dosing rate is related to V, I, high voltage (HV), and the Mo rod position, so you can fiddle with all of these.
3. Real Dose
• Quickly take the sample out of the STM chamber and align it visually (using the view port) so that the sample is in front of the evaporator.
• Retracted the thickness monitor.
• Introduce BT at ~3*10^-9 torr (this will not be a very stable value) and at the same time, anneal the sample at 10 W.
• Open the evaporator valve to begin dosing. You will dose for about 1 hour, while keeping an eye on temperature and maintaining dosing with adjustments to rod position and current. Although flux and emission current are imperfect measurements, you would like for them to be somewhat steady values.
• Stop dosing the Mo by closing the evaporator valve. Close the BT leak valve. 
• Anneal the sample at ~20 W. While waiting for good pressure to restore, turn off the evaporator. 
• Let the sample cool before putting back in the STM.

Preparing Sulfur Terminated Copper

1. 2-3 * 10^-7 benzenethiol for 5 minutes.
2. Anneal with 10 W for 10 minutes.
3. Simultaneously anneal at 10 W and dose BT for 10 minutes.
4. You can either conclude here to dose Mo immediately, or you can wait until good pressure (3*10^-9 torr) is restored and anneal at 20 W for 20 min.

Aligning the Sputter Gun on the STM Chamber

1. Position the sample. Watch to make sure you do not hit the mass spec with the transfer arm. To avoid crashing, you will need to move the sample up and away from the mass spec. You should also angle the arm block 45 degrees from horizontal.  The final position should have the sample in from of the mass spec with the fine adjustments set such that the sample is as far down and out (towards the view port) as possible.
2. Close the mid valve (if you haven't already) and the turbo valve.
3. Make sure the ion pump is off by turning off HV1. The controller is on the bottom shelf of the cart and is leftmost. The ensures that you don't pump away the argon you are leaking in. This pump won't be turned back on until the sample is back in the STM chamber.
4. Leak in 1*10^-4 torr Argon (and then close). Switch off the ion gauge (using the back panel switch).
5. Turn on the sputter fun. Increase the filament current to 20 mA and the beam voltage to 3kV.
6. Alight the beam so that you can see a blue spot on the sample. Adjust x, y, and focus to get this to align with your sample as much as possible.
7. Open the valve to the turbo to pump away the Argon.
8. As good pressure is restored, turn off the sputter gun.
9. Let pump until good pressure is achieved and then you can start (probably) sputter/anneal cycles.

Clean the Gas Line

Quick Description: If you have a gas line that you have connected to your system (typically through a leak valve), you want to bake it and make sure it is clean before connecting with your UHV system. Often, you will be using a gas bottle (Ar, H, etc...) and this should be fairly clean but requires a few rounds of flushing to ensure that a pure material leaks into your chamber.

The Point: Pure gas.

Prerequisites: Set up the pump cart.

Notes: 

1. Bake the gas line
• Ensure that the two gas valves are open (one connects the gas bottle to ambient/pumping and the other to the chamber).
• Start the pump cart (see link above).
• Begin the bake. On the STM system, there are two bake plugs attached to the Argon line. On the old XPS, there is only one plug attached to the Argon line. The voltage is ~55V and the time span is about 2 hours (or overnight).
• Turn off baking and let cool for 30 min or so.
2. Flush cycles
• Close the gas valve to the pumping and open the gas bottle. This fills the line with gas.
• Then close the gas bottle and SLOWLY open the gas valve to the pumping. This flushed the gas line. Watch the pressure to make sure that you're not making the turbo struggle too much. It's generally a good idea to stay under 20 W, but this does vary for different pumps.
• Cycle through this about 5-6 times and you should be good.

Pump Cart

Quick Description: For small things that need to be pumped, our lab has a pump cart which includes a rough and turbo pump. Common applications include cleaning out a gas line, freeze/thaw cycles, pumping down a temporary chamber, and more.

The Point: A quick set of pumps to get the job done.

Notes: You may want to check that everything is plugged in and connected. If you are experienced, this should make intuitive sense to you. Otherwise, you can check with a more experienced labmate.

1. Open the turbo/chamber valve. Open the turbo/rough valve. This allows everything to stabilize to the same pressure.
2. Begin rough pump, wait a little bit so that again pressure can stabilize. It should be able to reach at least 10^-2 torr (and for a new, good pump 10^-4 torr), and ideally you can check this with a pirani gauge.
3. Turn on the turbo and wait for it to spin up to full speed and low power. If this does not happen, you may have a leak that needs to be addressed or a problem with your pumps (for example, if your turbo is back by an oil pump, this pump may have spewed a bit of oil).

Mass Spec Procedure

Inficon Mass Spec (from Cindy Merida):

1. To begin, open TWare32 program. Make sure power supply for actual mass spectrometer is connected, and the RS232 cable is connected as well.
2. When program is opened, Sensor1_P1TSP2 will appear green if everything is connected properly, if not, it will appear blue.
3. Click on the Sensor icon, and monitor, to monitor the full mass spectrum.
4. The defaults on the mass spectrum are set at MaxAMU of 70, which can be changed to 100.
5. Make sure both the Emission and Multiplier icons are clicked to have the mass spectrum running properly. If pressure is bad, turn off the multiplier.
6. Also, the range can be changed, to give bigger peaks, adjust to a smaller range and vice versa.
7. When finished running, click stop on the software, and close the window. Unplug or switch off the power supply