Starting requires you to have a good sample. Usually we use lithium Niobate. We need to use the Aluminum Etch in the acid bench to clean away all of the aluminum and resin and anything that could possibly be on our samples. Then we put them in nanostrip (nicknamed piranha) to make sure that nothing is left at all.

Using the Denton E-Beam Evaporator we can deposit a layer of aluminum on our samples. The machine can do many diverse materials but most often we choose to use aluminum.

Make sure that everything is clean, if not you can take it to the solvent bench and use acetone and IPA to get clean. Once they are clean, take them to the spinner. Put on a bonder and let it spin for a little while before putting on the resin and letting it finish the cycle. Then that resin must be baked for about 20 minutes or so to finalize the process.The resin for simple samples is EBR::S1805:1:1

Using matlab scripts which can be found on the group j-drive, design all sorts of devices. These scripts eventually will help to create dxf files which the hiedelberg can read and interpret.

Pick something with similar features and run a dose test. This is done by using a batch on the heidelberg. For example the thick sample dosage test I ran swept powers 20 to 38% in a line.
You are looking for something with a 50/50 duty cycle meaning that the pattern and the in-between the pattern look to be about the same width. Also you are looking for something that looks really clean and pretty.

Using a recipe found on the Wiki that is based on the resin that you used you will choose Hiedelberg settings to use. After that machine is done, take it over to the bench and put it in a certain chemical also listed on wiki. You need to make sure that you have it in the center of the machine, make sure it is focused so that you can see the grains (you might need to take down the brightness factor or gain). The you need to find the plate center, usually it is sufficient to use fast mode, then you can click expose and come back one hour later. You need to develop it in a solution of 300 MIF for 30 seconds if you used the 1:1::S1805. After the sample has been developed, you should look at it under the microscope to see if it worked and did what you expected and what you wanted. If you are satisfied then you can take it to Al etch.

Take over to the acid bench and put in the al etch. Usually it is good to use a dummy sample to find out an approximate time. A dummy sample is one that is approximately the same design but broke somewhere earlier in the process. If you do not have a dummy sample then you will just need to be careful. Go in small increments until you see that the aluminum has the design you are looking for. This step takes the design from the resest and puts it down one more layer to the aluminum. Lately we have been having problems with the Al etch cracking our samples, this problem is most likely caused by thermal shock. A possible solution is that we need to warm up the samples before we place them in the acid. The acid should be kept at 50deg Celsius. You can either put the samples on a hot plate with a couple cleanroom towels underneath or you can place a beaker of water on the hot plate so that it is closer to the 50deg. If it will not be developed again, when you take off the acetone does not matter as long as it is after you Al etch to get the pattern into it.

To prevent cracking and breaking in Proton Exchange, it is good to polish the sides before. Use the polisher in the grey room. Make sure that you protect the top of your sample, usually a small paper towel will suffice. Use the spinners that you need and make sure that the water is running to cool down from friction and clear away any particles left behind from the polishing. The wheel does not need to spin more than about half of its speed

Pure :
Take it into the laser room acid bench. It is preferred to use a lab coat, gloves and a hood. Place your sample in the heat, but not in the acid for about a minute so that the temperature shock does not crack your sample. Place in benzoic acid for about 12 minutes. Take out slowly, allowing it to sit in the heat above the acid for about a minute to cool down. Let the sample sit and cool for about 10 minutes for it to fully cool. Place test tube in Acetone to clean off any of the acid. Remove sample and handle with tweezers. Place in acetone and then quickly but carefully place in IPA. It is necessary to place in IPA before acetone dries and forms a white residue on the sample. If it dries, simply place again in acetone and repeat until sample is clean.Then you need to bake it in the oven for about 45 minutes.
Dilute :
Same place as the pure but use the acid on the right instead of on the left. This will take a lot longer but yields better results. You need to have it in the acid about 4.5 hours. Make sure you are careful to slowly control the heating up process by sticking above the acid for a minute or 2 so that it doesn't go into heat shock. Same when leaving the acid, do all you can to be careful and move slowly. Once it has been cooled down to room temperature, clean it like in pure and anneal it for the same amount of time, 45 minutes.

After completing proton exchange, the design has now altered the crystal lattice structure of your lithium niobate sample. It is probably safe and a good idea to take off the rest of the aluminum. This can be done at the Al etch station of the acid bench. To see if the proton exchange worked you can use the metricon machine. This machine measures the modes by using power density of a laser. Use a dummy sample to get everything aligned properly. There is a prism near the top and your sample should be placed directly the side of it and held close by a moveable part which applies physical contact. Be careful with it so that it does not smash your sample as it moves back and forth. Once in place you are looking for a wet spot. A wet spot looks like a fuzzy little area usually in the corner of the prism. You need to align the laser so that it is directly on the wet spot for maximum results. You can use manual mode and twist the dial back and forth to verify that you get two spikes of activity before using auto scan and seeing the results. Once you have verified and set up with the dummy sample, you should be able to put your sample on the prism and then run it directly. You should see modes pretty close to where the dummy sample gave them as well.

Now you should go to polisher and polish each of the edges which are important. Important probably means how much light will be shined through it. If you look at the right light, you should be able to see some small lines across the edge. Those all need to be gone for you to be finished. Start with the bigger sized plates and gradually move down to the smaller plates until the lines are gone. Smaller plates are more fine but take longer to do anything, bigger plates work faster but may not get quite as fine. It is a balance to know what to use and for how long.

Run whatever tests you deem necessary so that you can verify your device worked and/or figure out what didn't work so you can continue on in the glorious research process of discovery, failure, experimentation and eventual success in changing the world.

If you are using the red laser apparatus in the laser room, you want the laser to go in the back side and then hit the grating, go through the wave guides and come out. To do this, you will need to align the angle, x, y, and z knobs correctly and precisely. Eventually you should see some mode lines come out that are pretty close to vertical except for the spreading that naturally occurs. This can be difficult if you don't know where the wave guides are so it might be beneficial to start testing before the aluminum is totally etched off, that way you can find the waveguides and close measurements that should only need to be tweaked a tiny amount once the aluminum has been totally etched off.