content(ment)
Sep. 27th, 2019 10:26 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kaberettThis morning, at around a quarter past nine, having been in work for an hour and a bit, I decreed myself Done For Now and left the basement and pootled down to the station in the sunshine and got myself an almond croissant.
And then ate it, still outside and in the sunshine, while sorting out MY DATA.
I showed up on Wednesday, when my instrument booking started, to discover that the samples I'd been intending to run were... much more viscous than they ought to be. This was particularly frustrating because the instrument itself was (mostly) running beautifully: very stable, ridiculously high sensitivity (I was getting ~1000 V/ppm on that machine for lead and thallium, which is about twice as high as normal and much higher than I've ever previously seen it).
There was also the minor issue that the Uptake Tube had got fairly badly bent earlier in the week, so it kept woobling off sideways if it wasn't watched even more attentively than normal, which was... certainly a thing.
Happily I had some leftover sample solution from my last run that it wouldn't hurt to measure again, so I started out -- having been delighted by how well the machine was behaving, then crestfallen about the new samples -- setting those up to run (and watching them some more). While that was going on I tried boiling up the new samples again, in the forlorn hope that it might help, and was then astonished when it... did seem to.
So: day 1, I got A Bunch Of Duplicate Data, and I rendered my New Shiny Samples usable (which was in doubt for a lot of the day), and then to my utter astonishment I did a quick concentration check (detailed explanation to follow a little way down the What Do I Even Do All Day series) and... I actually... had measurable amounts of thallium... in all of them. This was surprising because (1) I'd managed to spill a non-trivial amount of two of them (because Tired, essentially), but also (2) I was expecting at least four of them to be sufficiently low concentration that they were going to be a nightmare to analyse. That this turned out not to be the case was very exciting, at least to me.
However. By this point it was Really Quite Late on Wednesday night, and I wasn't happy running samples overnight (because of the issue with the tube). While I'm normally twitchy about turning the machine off once I've got it stable, in this instance I... had brought it up very rapidly with absolutely no problems in the morning, and the previous user (on Monday) had apparently had a similar experience, so I was willing to gamble on it still being in a good mood if I shut it down overnight and then restarted it on the Thursday morning. Which has the added benefit of not using up argon on analyses I didn't really care about, notable mainly because argon is kind of expensive.
So I came back in on Thursday, and I got the machine up and running, and it behaved itself (via an adventure with a blocked nebuliser), and after a great deal of back-and-forth I finally had my sample solutions, at the correct concentrations, with the correct amount of lead added to them, with similar Pb/Tl ratios, with actually matching standard solutions to measure them against, and... it was 6pm, and I had 11 samples each of which I wanted to analyse twice, plus another couple of standards, and the machine manages roughly two samples an hour, so I was looking at 12-14 hours of analysis time.
Which, given I wanted to be visitingsebastienne tonight, meant I really didn't want to be shutting the machine off again and then starting it up this (Friday) morning (because even when it's behaving startup procedures take a couplefew hours)...
... so I took another gamble and set an automated batch run going, with the intention of supervising it until about 9pm and then taking a call on whether I was willing to risk leaving it overnight in spite of the Adventure Tube, and while it was doing the first few samples I went and got the next batch of samples weighed out...
... and returned to the mass spec lab with some trepidation to find that not only had it behaved itself beautifully so far, the numbers coming off it were perfect.
Hmm, I thought to myself. Hmmmm.
And it got to a quarter past nine, and the mass spec had successfully got the probe into the fifth sample vial (which was where it had gone ADVENTURE twice over the course of the day), and I decided that actually sleep sounded big and clever and I could? go in a pile with Adam? and get cuddles? and have to get up at 6am but actually getting up at 6am is less bad than an all-nighter even if it doesn't feel that way at the precise moment of the 6am in question?
So I went home via a Bus That Broke, and mostly managed to blame TfL's subcontractors for failing to adequately maintain their fleet rather than myself (it got taken out of service, everyone got kicked off, it was a bit awks), and I was in bed by midnight...
... and then, as mentioned, I got into lab at 8am and looked with More Trepidation upon the autosampler and discovered that it had successfully eaten up all of the sample solutions, and therefore with Great Fear I inserted a USB stick into the computer and acquired The Data and copied it over to my laptop and fed it through my Python script that turns a million unhelpful not-even-CSV .txt files into something I can straightforwardly bung into a spreadsheet...
... and it was all usable and all good, in spite of spilling my samples everywhere and having them be distressingly sticky and having a higher lead blank than I Really Like To (like, ~60mV of lead, and the machine was happy, and it is about thirteen hours later and I've got all the data tidied up & plotted up & sent off to my supervisors and I'm still a little giddy with relief and amazement and gratitude, and now I'm in a nice flat on the coast and I... am going to go to sleep.
Probably while continuing to perserverate about the chemical partitioning of cerium between clinopyroxene and a silicate melt, but that is a story for another day, I think. Possibly tomorrow.
And then ate it, still outside and in the sunshine, while sorting out MY DATA.
I showed up on Wednesday, when my instrument booking started, to discover that the samples I'd been intending to run were... much more viscous than they ought to be. This was particularly frustrating because the instrument itself was (mostly) running beautifully: very stable, ridiculously high sensitivity (I was getting ~1000 V/ppm on that machine for lead and thallium, which is about twice as high as normal and much higher than I've ever previously seen it).
There was also the minor issue that the Uptake Tube had got fairly badly bent earlier in the week, so it kept woobling off sideways if it wasn't watched even more attentively than normal, which was... certainly a thing.
Happily I had some leftover sample solution from my last run that it wouldn't hurt to measure again, so I started out -- having been delighted by how well the machine was behaving, then crestfallen about the new samples -- setting those up to run (and watching them some more). While that was going on I tried boiling up the new samples again, in the forlorn hope that it might help, and was then astonished when it... did seem to.
So: day 1, I got A Bunch Of Duplicate Data, and I rendered my New Shiny Samples usable (which was in doubt for a lot of the day), and then to my utter astonishment I did a quick concentration check (detailed explanation to follow a little way down the What Do I Even Do All Day series) and... I actually... had measurable amounts of thallium... in all of them. This was surprising because (1) I'd managed to spill a non-trivial amount of two of them (because Tired, essentially), but also (2) I was expecting at least four of them to be sufficiently low concentration that they were going to be a nightmare to analyse. That this turned out not to be the case was very exciting, at least to me.
However. By this point it was Really Quite Late on Wednesday night, and I wasn't happy running samples overnight (because of the issue with the tube). While I'm normally twitchy about turning the machine off once I've got it stable, in this instance I... had brought it up very rapidly with absolutely no problems in the morning, and the previous user (on Monday) had apparently had a similar experience, so I was willing to gamble on it still being in a good mood if I shut it down overnight and then restarted it on the Thursday morning. Which has the added benefit of not using up argon on analyses I didn't really care about, notable mainly because argon is kind of expensive.
So I came back in on Thursday, and I got the machine up and running, and it behaved itself (via an adventure with a blocked nebuliser), and after a great deal of back-and-forth I finally had my sample solutions, at the correct concentrations, with the correct amount of lead added to them, with similar Pb/Tl ratios, with actually matching standard solutions to measure them against, and... it was 6pm, and I had 11 samples each of which I wanted to analyse twice, plus another couple of standards, and the machine manages roughly two samples an hour, so I was looking at 12-14 hours of analysis time.
Which, given I wanted to be visiting
sebastienne tonight, meant I really didn't want to be shutting the machine off again and then starting it up this (Friday) morning (because even when it's behaving startup procedures take a couplefew hours)...... so I took another gamble and set an automated batch run going, with the intention of supervising it until about 9pm and then taking a call on whether I was willing to risk leaving it overnight in spite of the Adventure Tube, and while it was doing the first few samples I went and got the next batch of samples weighed out...
... and returned to the mass spec lab with some trepidation to find that not only had it behaved itself beautifully so far, the numbers coming off it were perfect.
Hmm, I thought to myself. Hmmmm.
And it got to a quarter past nine, and the mass spec had successfully got the probe into the fifth sample vial (which was where it had gone ADVENTURE twice over the course of the day), and I decided that actually sleep sounded big and clever and I could? go in a pile with Adam? and get cuddles? and have to get up at 6am but actually getting up at 6am is less bad than an all-nighter even if it doesn't feel that way at the precise moment of the 6am in question?
So I went home via a Bus That Broke, and mostly managed to blame TfL's subcontractors for failing to adequately maintain their fleet rather than myself (it got taken out of service, everyone got kicked off, it was a bit awks), and I was in bed by midnight...
... and then, as mentioned, I got into lab at 8am and looked with More Trepidation upon the autosampler and discovered that it had successfully eaten up all of the sample solutions, and therefore with Great Fear I inserted a USB stick into the computer and acquired The Data and copied it over to my laptop and fed it through my Python script that turns a million unhelpful not-even-CSV .txt files into something I can straightforwardly bung into a spreadsheet...
... and it was all usable and all good, in spite of spilling my samples everywhere and having them be distressingly sticky and having a higher lead blank than I Really Like To (like, ~60mV of lead, and the machine was happy, and it is about thirteen hours later and I've got all the data tidied up & plotted up & sent off to my supervisors and I'm still a little giddy with relief and amazement and gratitude, and now I'm in a nice flat on the coast and I... am going to go to sleep.
Probably while continuing to perserverate about the chemical partitioning of cerium between clinopyroxene and a silicate melt, but that is a story for another day, I think. Possibly tomorrow.
(no subject)
Date: 2019-09-27 09:27 pm (UTC)(no subject)
Date: 2019-09-28 02:32 am (UTC)(no subject)
Date: 2019-09-28 06:26 am (UTC)(no subject)
Date: 2019-09-28 08:21 pm (UTC)(no subject)
Date: 2019-09-27 10:47 pm (UTC)(no subject)
Date: 2019-09-28 12:08 am (UTC)(no subject)
Date: 2019-09-28 01:56 am (UTC)(no subject)
Date: 2019-09-28 04:05 am (UTC)(no subject)
Date: 2019-09-28 12:23 pm (UTC)(no subject)
Date: 2019-09-28 12:31 pm (UTC)What a good! \o/
(no subject)
Date: 2019-09-30 08:27 am (UTC)(also hello sebastienne :) )
(no subject)
Date: 2019-09-30 10:18 am (UTC)(no subject)
Date: 2019-09-30 02:40 pm (UTC)(no subject)
Date: 2019-10-06 07:01 pm (UTC)