What Do I Even Do All Day: Part 1
Sep. 11th, 2019 11:21 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
kaberettI realised over dinner the other week that even A has very little idea what my day-to-day work life looks like; I think I may have inadvertently given the impression that I shut myself up in a lab for a few days and emerge at the other end, slightly dazed, clutching SCIENCE.
Therefore, in the interests of demystifying all of this a bit, I'm going to... talk about lab processes a bit? Talk about lab processes a bit.
The brief overview is that I'm measuring the precise composition of a trace element -- thallium -- in a variety of volcanic rocks. Every gram of rock will contain somewhere between 10 and 600 nanograms of thallium, which I want to separate out so that I can (eventually) measure the precise ratio of its two stable isotopes, 205Tl and 203Tl, for reasons. This means I need to work in ultra-clean conditions, because I'm working with such tiny quantities of material. So, for example, basically everything takes place in a suite of clean rooms: laboratories with positive air pressure, i.e. air pressure deliberately greater than local atmospheric pressure, where the entire air volume of the room gets replaced with completely new thoroughly filtered air multiple times an hour.
Skipping over all of the details of how I decide which rocks to analyse, and also all of the preparatory beaker-cleaning, let's start with Digestion.
The first step is weighing out my sample. If I'm lucky I'll have a vague idea of how much thallium (my element-of-interest) the rock contains, which lets me work out how much I need to digest in order to have enough thallium I can measure its isotopic abundances usefully; if I'm less lucky I have to guess, based on what kind of rock it is and its abundance of other relevant elements (rocks that are high in potassium tend to also be high in thallium, for example, because thallium can sit in the same crystallographic sites in silicate mineral structures that potassium hangs out in).
By the time I get my hands on them, my samples are all finely-ground powders, generally grey, in a series of oversized labelled vials. All I gotta do is suit up, grab some clean beakers, and settle myself down in the isolated weighing room with the microbalance (so called because it measures to microgram precision & accuracy).
(A digression: "suiting up" in this context means swapping my horrid outdoor shoes for a pair of entirely enclosed white Crocs that live in lab, putting on my low-residue anti-static hooded lab coat, putting on a Buff or other headscarf to keep my hair off my face because No, swapping my glasses out for my prescription safety specs, washing my hands in purified water, and putting on appropriate gloves. The purpose of this exercise is for the most part to protect my samples from me, rather than to protect me from them & the chemicals I'm processing them in. "Beakers" means screw-top PFA (generic Teflon) vials, which have been washed in a series of acids, rinsed in ultra-pure "MQ" water, and dried.)
Depending on how many people are coming in and out of lab (opening and closing doors affects the air pressures, and I'm working at high enough precision that that's enough to unsettle the balance) and how fussy the scales are feeling anyway, weighing out a batch twelve samples will take me between an hour and two and a half hours. Of those twelve samples, one is a total procedural blank (it's an empty vial, I don't put any rocks in it, I just open it up to the atmosphere in lab & put it through all the other steps of the process with all the other chemicals, so I can tell how much background thallium there is in the process), one is a well-characterised reference material (that's been measured a bunch of times, so I can check I'm getting the right ratio of thallium isotopes out - if I don't, that means I can't trust any of my sample measurements either), and ten unknown samples. For each of these I open up the balance, stick the capped vial on it, close it up again (because the scales live in a tiny box, because otherwise the fluctuations in air pressure caused by me moving around would prevent it from ever settling down), wait for the number to stabilise, write it down, dump my powder into the vial, close it up again, and weigh it again - repeatedly - until the balance stabilises to the point of giving me two numbers within (in theory) about a microgram of each other.
Sometimes I have to resort to using the antistatic gun, because wee powders you're weighing into plastic get very excited about static electricity and that... fucks with your weighing. I'm not entirely convinced it does anything, but the placebo effect is a marvellous thing.
Having weighed my powder out, the next -- immediate -- step is to drop some concentrated (14M) in-house-distilled nitric acid on it. Generally about half a millilitre to a millilitre, once I've confirmed the weight, in order to get it all taken up in liquid so it doesn't just... sproing out everywhere when I open the lid back up to add the concentrated (48M48%/28M) in-house-distilled HF, because (1) I just weighed that, and (2) it makes a mess and is a potential contaminant.
Once I've got everything weighed up it's back out of the balance room and into the main lab, where with any luck (1) I've booked the flow hood I need to work in, and (2) someone else will be around to babysit, while I do The Thing With The HF.
Hydrofluoric acid is Not Your Friend. It's very useful for dissolving silica, which means you can't store it in glass bottles because it just... eats... them. It gets used to etch computer chips. And in geology, because most rocks are mostly SiO2, and mostly you don't care about the silica, so you want to eat it all up and boil it off and generally get rid of it so that you can actually pay attention to anything else that's going on in your rock. It will kill you if you fuck up, you do need to know where absolutely every single drop goes, but it's basically "perfectly safe" in the sense that because it's so nasty the protocols for handling it are extremely thorough: I'd have to fuck up at least three different things to actually be at any risk of harm, so, I just don't do that, then. (This does mean I have to be very careful about resting on the days either side of handling HF, and about postponing if I'm not feeling steady enough, which is deeply frustrating when it happens, but there we go.)
I slosh (carefully! in extremely controlled fashion!) the HF onto the samples, I cap them very tightly, I stick them on a hotplate at 140°C, and I leave them to stew in their own juices for two days while I get on with the rest of my life. (You leave them at least 12 hours in order to actually get everything into solution; you don't leave them longer than 48, or at higher temperatures, because that increases the risk that you'll get your charming conc HF-conc HNO3 mixture condensing into the threads of the screw-top vial such that it's Hassle to undo safely.)
Sometimes, though, as is the case for my current set of samples, there is a Bonus Step. The Bonus Step is cold leaching: if the rocks I'm looking at are particularly susceptible to post-depositional alteration, e.g. they got picked up from underwater or they're a pumice i.e. full of holes, because thallium's fairly fluid-mobile and fairly easy to shift around it's important to make sure that the thallium I'm measuring is actually from the rock itself rather than whatever crud got stuck to the outside between its being erupted and being collected. For samples where this is necessary, instead of adding conc nitric after weighing, I add weak hydrochloric, and then when I take it back to lab I dump some stronger (6M, in-house-distilled, you get the idea) HCl on it (whereupon it goes yellow, because iron chloride is yellow) and... stick it in an ultrasonic bath for thirty minutes. And then I leave it to settle for at least 15 minutes, and then I carefully draw off the leachate with a precision pipette and dump it in another clean beaker and set it evaporating down at around 120°C, and then I put some more HCl on the powder residue and ultrasonicate it again (for twenty minutes, this time); rinse and repeat until the leachate runs clear. Then I get to do it all again with two rounds of MQ, to clear the HCl out, and by the way this all generates a ridiculous volume of leachate to dry down, and then when I have finally got all that done & the leachate's also dried I... apply HNO3 and HF, and follow the steps above.
I have been learning Several Things about sensible ways to organise my life around Letting The Bloody Leachate Evaporate Overnight. There is so much of it. So much. Cold-leaching is adding a solid week to sample prep time, and that's almost all because of how long it takes the leachate to dry.
NEXT TIME: getting ready to separate out my thallium fraction. BEAKER CLEANING IS VERY GLAMOROUS.
Therefore, in the interests of demystifying all of this a bit, I'm going to... talk about lab processes a bit? Talk about lab processes a bit.
The brief overview is that I'm measuring the precise composition of a trace element -- thallium -- in a variety of volcanic rocks. Every gram of rock will contain somewhere between 10 and 600 nanograms of thallium, which I want to separate out so that I can (eventually) measure the precise ratio of its two stable isotopes, 205Tl and 203Tl, for reasons. This means I need to work in ultra-clean conditions, because I'm working with such tiny quantities of material. So, for example, basically everything takes place in a suite of clean rooms: laboratories with positive air pressure, i.e. air pressure deliberately greater than local atmospheric pressure, where the entire air volume of the room gets replaced with completely new thoroughly filtered air multiple times an hour.
Skipping over all of the details of how I decide which rocks to analyse, and also all of the preparatory beaker-cleaning, let's start with Digestion.
The first step is weighing out my sample. If I'm lucky I'll have a vague idea of how much thallium (my element-of-interest) the rock contains, which lets me work out how much I need to digest in order to have enough thallium I can measure its isotopic abundances usefully; if I'm less lucky I have to guess, based on what kind of rock it is and its abundance of other relevant elements (rocks that are high in potassium tend to also be high in thallium, for example, because thallium can sit in the same crystallographic sites in silicate mineral structures that potassium hangs out in).
By the time I get my hands on them, my samples are all finely-ground powders, generally grey, in a series of oversized labelled vials. All I gotta do is suit up, grab some clean beakers, and settle myself down in the isolated weighing room with the microbalance (so called because it measures to microgram precision & accuracy).
(A digression: "suiting up" in this context means swapping my horrid outdoor shoes for a pair of entirely enclosed white Crocs that live in lab, putting on my low-residue anti-static hooded lab coat, putting on a Buff or other headscarf to keep my hair off my face because No, swapping my glasses out for my prescription safety specs, washing my hands in purified water, and putting on appropriate gloves. The purpose of this exercise is for the most part to protect my samples from me, rather than to protect me from them & the chemicals I'm processing them in. "Beakers" means screw-top PFA (generic Teflon) vials, which have been washed in a series of acids, rinsed in ultra-pure "MQ" water, and dried.)
Depending on how many people are coming in and out of lab (opening and closing doors affects the air pressures, and I'm working at high enough precision that that's enough to unsettle the balance) and how fussy the scales are feeling anyway, weighing out a batch twelve samples will take me between an hour and two and a half hours. Of those twelve samples, one is a total procedural blank (it's an empty vial, I don't put any rocks in it, I just open it up to the atmosphere in lab & put it through all the other steps of the process with all the other chemicals, so I can tell how much background thallium there is in the process), one is a well-characterised reference material (that's been measured a bunch of times, so I can check I'm getting the right ratio of thallium isotopes out - if I don't, that means I can't trust any of my sample measurements either), and ten unknown samples. For each of these I open up the balance, stick the capped vial on it, close it up again (because the scales live in a tiny box, because otherwise the fluctuations in air pressure caused by me moving around would prevent it from ever settling down), wait for the number to stabilise, write it down, dump my powder into the vial, close it up again, and weigh it again - repeatedly - until the balance stabilises to the point of giving me two numbers within (in theory) about a microgram of each other.
Sometimes I have to resort to using the antistatic gun, because wee powders you're weighing into plastic get very excited about static electricity and that... fucks with your weighing. I'm not entirely convinced it does anything, but the placebo effect is a marvellous thing.
Having weighed my powder out, the next -- immediate -- step is to drop some concentrated (14M) in-house-distilled nitric acid on it. Generally about half a millilitre to a millilitre, once I've confirmed the weight, in order to get it all taken up in liquid so it doesn't just... sproing out everywhere when I open the lid back up to add the concentrated (
Once I've got everything weighed up it's back out of the balance room and into the main lab, where with any luck (1) I've booked the flow hood I need to work in, and (2) someone else will be around to babysit, while I do The Thing With The HF.
Hydrofluoric acid is Not Your Friend. It's very useful for dissolving silica, which means you can't store it in glass bottles because it just... eats... them. It gets used to etch computer chips. And in geology, because most rocks are mostly SiO2, and mostly you don't care about the silica, so you want to eat it all up and boil it off and generally get rid of it so that you can actually pay attention to anything else that's going on in your rock. It will kill you if you fuck up, you do need to know where absolutely every single drop goes, but it's basically "perfectly safe" in the sense that because it's so nasty the protocols for handling it are extremely thorough: I'd have to fuck up at least three different things to actually be at any risk of harm, so, I just don't do that, then. (This does mean I have to be very careful about resting on the days either side of handling HF, and about postponing if I'm not feeling steady enough, which is deeply frustrating when it happens, but there we go.)
I slosh (carefully! in extremely controlled fashion!) the HF onto the samples, I cap them very tightly, I stick them on a hotplate at 140°C, and I leave them to stew in their own juices for two days while I get on with the rest of my life. (You leave them at least 12 hours in order to actually get everything into solution; you don't leave them longer than 48, or at higher temperatures, because that increases the risk that you'll get your charming conc HF-conc HNO3 mixture condensing into the threads of the screw-top vial such that it's Hassle to undo safely.)
Sometimes, though, as is the case for my current set of samples, there is a Bonus Step. The Bonus Step is cold leaching: if the rocks I'm looking at are particularly susceptible to post-depositional alteration, e.g. they got picked up from underwater or they're a pumice i.e. full of holes, because thallium's fairly fluid-mobile and fairly easy to shift around it's important to make sure that the thallium I'm measuring is actually from the rock itself rather than whatever crud got stuck to the outside between its being erupted and being collected. For samples where this is necessary, instead of adding conc nitric after weighing, I add weak hydrochloric, and then when I take it back to lab I dump some stronger (6M, in-house-distilled, you get the idea) HCl on it (whereupon it goes yellow, because iron chloride is yellow) and... stick it in an ultrasonic bath for thirty minutes. And then I leave it to settle for at least 15 minutes, and then I carefully draw off the leachate with a precision pipette and dump it in another clean beaker and set it evaporating down at around 120°C, and then I put some more HCl on the powder residue and ultrasonicate it again (for twenty minutes, this time); rinse and repeat until the leachate runs clear. Then I get to do it all again with two rounds of MQ, to clear the HCl out, and by the way this all generates a ridiculous volume of leachate to dry down, and then when I have finally got all that done & the leachate's also dried I... apply HNO3 and HF, and follow the steps above.
I have been learning Several Things about sensible ways to organise my life around Letting The Bloody Leachate Evaporate Overnight. There is so much of it. So much. Cold-leaching is adding a solid week to sample prep time, and that's almost all because of how long it takes the leachate to dry.
NEXT TIME: getting ready to separate out my thallium fraction. BEAKER CLEANING IS VERY GLAMOROUS.
(no subject)
Date: 2019-09-12 09:09 pm (UTC)On reflection, I'm glad to be doing back office library work these days!
And thank you for reccomending Ruby Violet - I had the blackberry sorbet and it was amazing!
(no subject)
Date: 2019-09-13 04:47 pm (UTC)*snort* re "just don't eat any": yes indeed.
... but! my organic chemistry! never included ferrocene! WHICH I THINK I'M RELIEVED ABOUT.
(I remain Wry about the part where I consistently did Very Well at my organic chem practicals during undergrad - good yield, right compound, etc - and consistently did terribly at inorganic chem, for reasons that remain completely opaque to me, and yet somehow ended up... doing a PhD... involving a whole bunch of inorganic chemistry...)
I am DELIGHTED that you enjoyed Ruby Violet, thank you for letting me know, and ALSO thank you for flagging up that they have blackberry sorbet on at the moment, because I am now contemplating A Detour on my way home...
(no subject)
Date: 2019-09-14 04:55 am (UTC)