NSLS - X23B - 11-2000

2000-11-15
Got SAF form and enabled beamline.
Johnny is using a new pumping stand for the mono chamber (pair of Si crystals). The old pump caused to much vibrations. As the pumping stand is located so that it can be kicked out by X23A users, checking the mono tank vacuum would be pretty wise.

There are a few important changes in the handling of the lockin:
On starting XDAC turn on Digital Feedback in the Mono setup menu.
Changing the gain is not necessary any more. (That is what the feedback does.)
Adjust mirrors in the middle of the scan, then return to the beginning and set BIAS to zero there.

We enabled the beamline and let it warm up.
A beamdump occurs just before I wanted to start the first edge scan on a V foil. They had to access the ring tunnel and repair a cooling water supply.
After the injection (15:30) I start the edge scan:
 

Another beamdump at 16:30!

I checked which motor numbers are used:
The correct names are Sample-Horiz, Sample-Vert and Sample-Rot with the numbers 3,4,1

mounted sample 1156 K2 S...  with E vector parallel c axis
 

After the third beamdump for this afternoon, I went to pick up Mihail and Roman from the station.
The beam returned at 8.30 p.m. We let the mirrors warm up again and started with a readjustment of the sample:
 

2000-11-16
Johnny obvoiusly was here lat at night (11 p.m.?). He left me a note, telling that he changed the scale on the oszi and turned on Io plotting in the XDAC.
At the beginning of the third scan it lost lock and Johnny retuned the optics and started the scans over.
He reduced sensitivity from 500mV to 1V on the lockin amp. as is was in overload.

He also mentioned that Io looked very big to him and suggested that I should add som He to Io or to shut down the slits a bit . Otherwise Io might not be linear.
Gain on Io is quite high (1e9), so probably thats not necessary. I have to talk to Johnny about that today. Closing the slits should not hurt in any case with my sample size.
 
 

Emergency ring access seems to take longer than they thought. They're starting the injection at 8:45.

At some point I changed the Gain reading for Io in the XDAC to its correct value of 1e7
 

Noticed a hole in the aluminized mylar foil at the end of the beamline extension in the hutch. fixed it.
Reduced exit slits to 1 x 1 mm.
I_t is significantly reduced. Doing motor scans again:
 

No good result: linearity changes 5-10%, Johnny says tey got it to as good as .5%.
So I open up the slits again, increasing the I_t.
Using mixture of N2 and He for Io chamber now. Flowmeter shows 40 for N2 and 10 for He. Io reduced from 3.05 to 2.65. Testing linearity again:
 

A propos gas flows:
 

Yesterday we connected the wrong hoses and were blowing N2 into the hutch instead of into the ion chambers. That problem is fixed, now.
 

I decide to leave the gas flows constant now. They are:
 

Doing another linearity check. Hopefully it's the last one:
 

Final gas flow table:
 

During the overnight scans, the energy has shifted by MORE THAN 100 eV. I corrected that this morning, but rather do another edge scan:
 

Edge was off  50 eV
 

Switched sample to have E perpendicular c now. Doing motor scans
 

Mounted new sample 608 E8 Rest w/ E perpendicular c (??? Sample orientation has to be rechecked, Perpendicular orientation results from comparison with spectra of other samples). Using graphite for thermal contact and wiring for resistance measurement.

Starting with motor scans:
 

After dinner we decide to go on with Mike's and Roman's experiment.
We tilt the Displex out of the way and put up their Ti foil.
 

2000-11-17
Just looked up the transition temperature of charge 608: 141 K on cooling, 148 K on warming up.
We have set up the Displex. I took my chance and spilled some cooling water on Roman :-(
The forepump runs quite smoothly and does not seem likely to fall off the stool.
Resistance of sample is 76 Ohms @ 220 K

Looking for sample with motor scans:
 

Found the sample, calibrated the edge, retuned the optics.
Starting an EXAFS scan now
 

Johnny tried something on the mono motors which did not work yet. He restored the old settings and I take another scan after retuning the optics.
 

In each of the scans there are strange structures in the preedge region ( around 5300 eV) that change slightly from scan to scan. But that was not the reason why I aborted the scan. I noticed that the sample position has changed and the I_t signal was very low.
So doing motor scans again:
 

Measured offsets now, after changing gain on I_t to 1e10. Starting an EXAFS scan:
 

Reset I_t gain to 1e9 and measured offsets (really did it this time befor the first scan!).
Probably low gain is better for the scan, even if signal is low. But Johnny says that linearity is not guaranteed abov 8 Volts - and I was reaching 11 V!
 

Went to 149 K (73 Ohm), maybe the transition occurs at a temp. a bit lower than measured in Augsburg.
Beam dumped (2:45). I "stand by".

New injection, new photons, new alignment:
 

Just talked to Joe. He will have some beamtime in Feb. or March.
He looked at my k-space data of the 155 K spectra and was suspicious about the sharp peak at 12 A^-1. I will rotate the sample a few degrees after the next scan and see if that changes (and is a Bragg peak then) or if it's really in the data.
 

I believe that the sharp structure is just noisy. All three scans look quite the same.
Returning to 0 deg with the sample rotation and realigning:
 

Lowering temp by 5 K to 144 K
 

Went to 139 K, still no sign of the MIT
 

2000-11-18

Mono recalibrated from 5429 eV
Optics retuned
Sample aligned
 

Changed temp to 130 K (126.9 Ohms) - we're close to the transition now!
Changed the mono calibration by 10 eV to be on the safe side.
Taking a scan after retuning the optics, will realign the sample after this scan
 

Looks like there was some nonlinearity in the signal: Glitches at 6000eV (11.56 inverse Angstroem) seem to be related to glitches in Io.

Finally I found the gas mix calculation window in the XDAC (its a subfunction of the Detectors screen). Changing the gas mix from N2 to 1:1 N2:He should reduce absorption of the Io chamber from 14 % to 7 % at the V K edge. I set the respective flows to 20 units on the flow meters. The Io reading before the change was 2.76, so it should go to 1.4 when it is stable. I don't know if I can wait that long. I start another scan at the same temp.
 

Checking linearity w/ timescan:
 

Going down to 127 K and realigning the sample
 

Going to 124 K
 

Thought over another possibility why the peaks in Io could result in peaks in I_t: Possible pinholes in the sample! So part of Io is "leaking" through the sample without attenuation and reproducing the Io signal shape!

Gone to 120 K and 283 Ohms, aligning sample after injection while beamline is warming up:
 

Reduced Temp. to 116 K (395 Ohms)
 

Going down to 112 K (503 Ohms):
 

2000-11-19

Still there on a new day!
Going to 105 K
 

Skipping 95 K and 1265 Ohm and going to 90 K and 1665 Ohms right away.
 

I try to go down to 30 K. The temperature is reached for a few moments but going up again, then. The part of the cup that covers the 2nd stage cooling is icing. Probably the Al foil I used for insulating the 2nd stage touches the cup (got loose?). Shit! But I haven't found Johnny's super-insulation foil that worked so well the last times!

Spent a great hour from 4 a.m. to 5 a.m warming up the Displex, changing the Al foil that made thermal contact to the cup and cooling the Displex down again.
Sample alignment fortunately still ok, just doing the usual fine tuning:
 

Heating to 150 K and taking a last scan
 

Returned at lunchtime after 5 hours of sleep and really feeling better now.
Displex had warmed up. We rotated the sample to the second orientation which should be E parallel c.
Unfortunately one of the wires for the online resistance measurement went off together with a part of the glass the sample is mounte on. So no more resistance values for the new temperature series.

While I brought Vitaly to the train and picked up Scott, The Displex cooled down to 150 K and the beamline warmed up (hopefully). Aligning the sample and taking the first EXAFS scan:
 

going to 133 K, retuning optics (didn't change) and realigning sample
 

going to 130 K, retuning optics (didn't change) and realigning sample
 

going to 127 K and realigning sample
 

calibration: two subpeaks in calibration region; chose higher one. Recalibrated 5454.5 to 5475.
Aligned mirrors.
Set to 124K (dropped to 123.7 while setting)
 

2000-11-20

Still there on a new day!
Finally got two scans for 124 K that look alike and finishing that temperature that really gave me a hard time!

Going to 120 K
 

Strange things happen: Both scans are usable!
Changing temp to 116 K, doing the usual tuning and alignment:
 

Next morning:
Went to the beamline befor breakfast to warm up the mirrors and change the temperature of the sample to 112 K
 

Once it's working, it's really great: got two identical scans up to 14 A^-1 !
Going to 90 K now:
 

Going down to 30 K
 

Finally heating to 150 K again:
 

FINISHED WITH V2O3 FOR THIS RUN!!!

Checking energy calibration of monochromator now, using Ti, V and Cr foil:
 

Putting in the Mn foil.
 

2000-11-21
 

returned next morning:
 

Reducing temp to 200 K.
 

Reducing temp to 150 K.
 

Reducing temp to 100 K.
 

Reducing temp to 50 K.
 

Returned from dinner, put in Cr foil and Fe as a reference (16.7A^-1 above Cr edge).
Tried to tune optics. Flux was very low, so I tried to tune it for the V K edge where I knew the flux had to be around 1.0; was .01 what did not add to my pleasure. Noticed we had run out of He. Probably that already caused the flux to be so low during the Mn scans. Got a new bottle of compressed He for 34.91$ from the Stockroom.
Flux is slowly increasing.
Changing gain to 500mV

Starting 1st scan while gas flow still is settling
 

Lowering temperature to 250 K.
 

Starting 100 K scans:
 

Starting 50 K scans: