FacebookTwitterDiggStumbleuponGoogle BookmarksRedditTechnoratiLinkedin

Learning CR-39: Pictures from the New Toy [UPDATE: Electrolysis Begun!]

on .

As promised in the last blog post, we picked up a new microscope to aid in our CR-39 trek. Here she is in all her glory!  It's nothing too fancy, but we hope it will help us do this job.  See the listing here: Microscope listing on Amazon

The big advantage with this microscope is that we can take pictures of the chip exactly as we see it through the eyepieces, so you don't have to take anything we say just on our word! Also, now I can show you all why counting tracks is turning out to be such a royal pain in the rear.

Images of the Chips Immediately After Removal

As I mentioned in the last post, the biggest obstacle standing in the way of getting accurate track counts of the chips is the copious physical damage present on all of the samples. I am unsure if this damage is a result of damage during manufacture and shipment or of something we did in the handling procedure, although the latter seems somewhat unlikely, as our procedure contains very little possibility for chip damage. About an hour before the etching process begins, the chips are punched from the sheet and a hole is drilled in the upper left corner, they are washed in ethanol, then they are immediately wrapped in a lint-less cotton cloth (to prevent scratching), then wrapped in Al foil (to prevent stray particle damage), then placed in a Ziploc© freezer bag. The chips are kept in this arrangement until just before etching. When they are removed, they are threaded with a Ni wire and hung from a ceramic rod into the NaOH solution. When they are removed from the etchant they are soaked in water for about 5 minutes then washed with ethanol and wrapped in cloth and Al foil exactly as before. All of this handling is done with BlueNitrile examination gloves to prevent possible damage from oils on the fingers. Despite this amount of care , after etching for 6 hours in 6.5M NaOH at ~70oC (varied from 68-73o), the damage almost overpowers the particle tracks. Below are a few pictures to show what I mean.

This is a picture from chip 005, at 40x magnification. The chip was removed from the packaging and immediately etched as described above. The small track is circled in red among physical damage.

And here's an elliptical track on the same chip at the same magnification. (the big arc on the bottom right is the hole drilled in the chip.

Here's the same track at higher magnification (100x). The strand in the picture is believed to be a fiber from the cloth it was wrapped in.

Here are two examples of the ambiguous pits that are found all over on the chips (both examples are from chip 005, the first is at 100x, and the second is 400x). They do not appear to be tracks from particles as they are too shallow, but their circularity leaves me wondering. Does anyone have an answer for what these are?

Here is an example from chip 006 (etched immediately after removal from their packaging, just as done with 005). This picture shows three pits that appear to all have originated from the same source (at 100x).

Chip 001

Chip 001 was part of the first etching procedure we undertook. The chip was left out in the lab overnight in the lab and subsequently etched in 6.5M NaOH. Unfortunately, in this etching we didn't carefully control the temperature and temp reached 100oC for about 3-4 hours before we found the issue. This resulted in a huge over-etch of the chip, which made the pits very large, as shown below.

This picture is from 001, shown at 40x magnification. One can see that in comparison with the tracks in 005 & 006, these tracks are enormous, however they are directionless.

This picture shows tracks next to a chunk of physical damage (the large non-circular pit in the bottom left and the scratches in the background). In this over-etched chip, the tracks are very clearly distinguishable from physical damage.

Etching #3

For the next etching, in order to try and find the happy medium between the 100oC over etching and the 70oC etching, we plan on etching for 6 hours in the same 6.5M NaOH solution, but this time regulating the temperature at 80oC. This increased temperature will increase the etch rate (the rate at which material is removed from the CR-39 by the NaOH) and the result will be larger tracks. Also, we hope that this new procedure will reduce the appearance of damage tracks on the chip. Pits and scratches that result from physical damage to the detector are generally much more shallow than tracks created by incident particles, so the thought is that the accelerated etch rate will remove enough material from the bulk that the shallow physical damage will disappear. This effect can be seen in the pictures shown from chip 001, where the tracks are very large and quite clearly distinguishable from any physical damage. Unfortunately, this faster etch rate will also make incident direction of particles more difficult to determine, but hopefully the trade-off will be worth it.

If this etching goes according to plan and we can get an accurate count on our chips, we will FINALLY be able to start the electrolysis cell. Stay tuned!


UPDATE: Electrolysis Begun!

Well we finally have a method for counting tracks on CR-39, so we decided to just move on with things and start the experiment today. The current has been set constant at 70mA and will remain that way for the entirety of the experiment, which will last at least until next Wednesday. From the pictures below one can see voltage is 4.32V, meaning the resistance can be calculated at R = (4.32V/0.007A) = 61.7 Ohms. In the first few minutes here it can be seen that voltage is slowly decreasing, implying that resistance is also dropping, we'll see how that trend continues.

Below is the joint used in each cell, going: glass, o-ring, 6 micron Mylar, CR-39, o-ring, glass. Chip 011 is used in the live cell and chip 012 is used in the control

Here one can see the live cell (cloudy from bubble production) along with the power supply, which is current limited to 70mA.

And here are the two cells side by side. The control cell is on the right, set up exactly the same as the experimental cell except that electrolysis is not run in it. At the end of experimentation, both chips will be etched and we will search for a difference in the track count between the control and experimental cells.

Update Within an Update (Updception)

Well... The damn intern (me) screwed up already. Turns out the crimp from the Pt wire to the Cu wire (made of Ni plated Cu) was slightly in the electrolyte when electrolysis started. As a result, some of the copper from the crimp was absorbed into the solution, creating the blue layer seen in the picture below. The crimp has now been moved so that it sits out of the solution. We thought about it and the worst thing that will potentially happen is that the copper will plate the nickel cathode, and since nickel's cheap and we have a bunch of it, we're not too concerned about that happening. So, we're just going to let the cell run at least for the weekend and re-assess on Monday. It must now be noted that this is no longer a direct Oriani replication, but since Oriani's results have been a bit contested as of late, we figure adding a new element, however accidental, to the mix can't do much harm.

Add comment


Security code
Refresh

Comments   

 
0 #3 treadmill machines 2018-02-18 01:32
You have made some good points there. I checked on the
web to find out more about the issue and found most individuals will go along with your views on this website.
Quote
 
 
0 #2 Kapytanhook 2013-08-16 22:24
Really nice toy indeed, Getting some sharp footage off it too.
Quote
 
 
0 #1 Robert Greenyer 2013-08-14 22:47
First of.... ooo - nice toy!
Quote
 

Here is your generous contributions so far towards our $500,000 target, thanks everyone! : $45,020   Please Donate
See the current state of our booked costs here