 disturbed Rwy'n ddweud anghyrch yn mynd i'r fregyn ddweud ar gyfer a seradwch i ddweud sonf也可以in i'r dweudio'r sgwer dros y bwysig Felly yw weld yw ondau werthwch ar gyfer yrhaf A gweld yn fannydd yna. Felly mae'r ddannig ystafell a'r mewn lle yma i fyw mwy o'r cyfnodau Ieithiwch i bellach I measuring gravity and then finally finishing up with atomic force microscopes. So okay, where does this come from? I currently work in IT but when I was young and needed the money I did a PhD in genetics. I did a lot of techniques for this. I don't know if any of you have seen the XKCD comic, The Best Thesis Defences, a good thesis offence. So at the end of your PhD you have in the UK anyway, you have two experts who they try and pick holes in all of the work you've done. And I knew that one of the guys who was going to be assessing me, he liked to trip people up on how techniques worked. So I spent the whole PhD whenever I came across a technique just learning exactly how it worked because I knew it would help me out at the end. So that's where this comes from effectively, just a fascination with all these little weird techniques. Disclaimer, it's been a while since I've done scienceing professionally. So there's a lot of these techniques which are maybe a bit old and the cool kids are doing funky new things. So yeah, disclaimer there. These are the ones I worked with, not necessarily what state of the art. OK, so first off, NMR. You may know NMR from such hospitals as MRI machines. So these machines that produce the kind of the very cool live scans of bodies where you can see the brain and all the organs and things. These are based on NMR, nuclear magnetic resonance. And we're going to talk a bit about how it works because it's freaking incredible. Basically it works with aliens, it's magic. So this was the machine for those who've got the actual monitors. So this was the machine which introduced me to NMR. At the time it was the most powerful one in the UK and it was pretty cool. The health and safety briefing for this machine was simply epic. So I was told I had to take all the metal off because it contains a giant magnet and I went through like wandering with a metal detector to check out nothing on me. Because they'd had a case with another machine where a guy had got shoes with traditional iron nails in and he hadn't realised this so he'd taken the rest of his metal off. Then when he went near the machine it ripped the iron nails through his foot, through the shoe and onto the machine. The magnets these things are dealing with are truly incredible. Beth health and safety demonstration ever. So yeah, back to this. NMR, MRI, it was invented in the 40s and I'd say the common way you'll know it is hospitals. But how does it work? It is literally the hardest parts of maths, physics, chemistry and biology in one topic. It's probably the most interdisciplinary topic I've ever come across. It's incredible at how it was invented. I've no idea, a complete massive respect to the people involved. So first, quantum physics. Okay, quantum. Everything is spinning. So atoms have spin and there are two types of spin that are important. The first is nuclear spin which is the spin of the protons and neutrons and then you've got electronic spin which is the spin of the electrons. Now in some atoms they don't cancel out so they're called unbalanced and those are the ones that are useful for NMR. So in the presence of a magnet these spinning atoms act like tiny magnets and they line up to face with a magnetic field. So whilst they're spinning in all these weird directions if you put a very strong magnetic field in place then they align to work with the field. So here is my fantastic PowerPoint animation. So you've got some atoms spinning in different directions and then you introduce a very strong magnet and then this causes the atoms to all line up and go with the field. Then what you do is you introduce a radio. So you flood it with electromagnetic energy with radio energy and when you hit the right amount of energy and I'll get to that in a second the atoms are able to flip their spin back to their original state rather than the magnetised state. If you then take the radio away they flip back and they emit the energy and they just absorb to get them there. So what you do is you detect that energy. You detect that radio wave that's coming off and you work out where it was in your sample and how much energy came off and that tells you where it is but the important thing it tells you is the amount of energy required because the amount of energy required to flip a spin state is different depending on what's nearby. So you've got something called electronic shielding which is the effect the electron cloud has on the nucleus. Now different atoms will have different shielding and so they'll require different energy to be able to flip them but most importantly is atoms will be affected by nearby atoms. So a hydrogen atom connected to a carbon atom will require a different amount of energy to a hydrogen atom connected to an oxygen atom. So by knowing how much energy it takes to flip the state you can work out what was actually near that atom at the time when it flipped. So what you get out of an NMR trace is this. You get a signal wave and this is what's called in the time domain and I don't understand that bit. I'm not a mathematician. What you want to do is put it into what's called the frequency domain so instead of seeing the signal over time you want to see the frequency and the power that came out of it and how do you do this? Mathemagics. So as far as I can tell it is mostly magical but if you ask a mathematician it's actually Fourier transforms which again I can't explain but someone here will be able to. And what you get out of the end is you get this trace which shows you the frequency and the amount of energy at the amount that came out at that frequency and at this point you can start to do analytics so you might be able to say that okay I know that this peak corresponds to water and then you can start, you know, picking apart the different things in NMR. So how is this useful? How does it get us pictures like that from MRI machines? So an MRI machine is effectively a fairly low resolution NMR machine and it measures the water across the body so if you think different tissues will have different water contents you've got bone which will probably be fairly low, blood fairly high and by measuring the water throughout the cell live you get a fairly good image of what's going on in the body and if there's autism or radio waves there's no risk, it's not ionizing, it's not like x-rays. So use NMRs sort of to get a picture of what's going on live in the cell. Something called functional MRI which is fascinating which what it does is it measures the difference between oxyhemoglobin and just hemoglobin so it can tell when a cell is using oxygen. So what you do is you show someone a picture and then you look at their brain and where their brain is using oxygen that part of the brain is currently active so you can start to work out which parts of the brain and that gives you a whole bunch of interesting experiments you can do. It's also used for samples so I'll talk more about this with the mass-spec section but you can take a sample and very accurately work out what's inside there, even down to kind of atom level. It's very useful for protein structure so proteins that you can't get the structure of by other ways you can use NMR to work out what atoms are nearby and start building up the structure of the protein. What I used it for was NMR where you take a sample do an NMR of your sample then add something and do it again and see what changed so I was looking at proteins and trying to find out where things were binding on that protein so you take a sample of the protein, you get the atoms then you add the thing you think will change and some of those atoms shift because there are new electrons in their electron clouds and you know that's where the thing you're looking at is binding to your protein so that was how it helped me. Onto mass spectroscopy you may know mass spectroscopy from airports so after they do the kind of swabs of your equipment and things and put them in a machine that machine is a mass spectrometer what it basically tells you is what's in a sample it breaks down a sample and tells you what's actually in there How does it work? So basically you ionize a sample so you give it a charge, you break it down and give it a charge and then you use a spectrum of power hence mass spectroscopy to select the ions at different energies and masses so you know how heavy it is and how charged it is and then you measure it so there are different types of mass spectroscopy but basically they all break down into these three stages creating ions splitting ions and then measuring them and there are various different ways to do each stages the one you came across with quite a lot is called electrospray so what you do is you get a charged nozzle and you squirt your sample through there and as it goes through the nozzle it ionizes because the nozzle is charged but it also disperses and then you funnel it into your detection thing so that's quite a crude way of doing it it works but it's cheap my favourite is using a laser obviously and in this you get your sample you hit it with a laser beam and that causes the ions to break off and then it goes into your machine and then you have to separate them so ions have mass and charge so what you're trying to do is apply different electric fields and the ions will behave differently depending on how heavy and how charged they are so you can select for them that way by putting them in a field some of them will go one way, some will go another way and depending on where you put your detector you can attach different ions there's another one called quadrapoles where you have four electrodes and you create overlapping electric fields and then by varying the voltage on the fields you can select for different ions going through because say if the electric field is too strong for that ion it will just crash into one of the poles but at some point it will be exactly right and it will pass all the way through and you detect it so by sweeping through you can see all the different ions however the current state of the art or at least it was when I stopped scienceing was the Fourier transform internal cyclotron resonance which is basically a particle accelerator so you blitch a sample stick it into the particle accelerator it whizzes through and then you can select particles coming out and that's incredibly accurate lets you really determine very small differences so what can you do with it like NMR you can get a sample so it doesn't have the same resolution as NMR but you can find out what's in a sample this can be useful for things like say you've got a new protein where you don't know what it does you can break it down and see what the individual pieces were something that is useful in medicine is you can use it diagnostically so for instance say you take a urine sample from a normal patient and then you take a urine sample from someone who say prostate cancer even though you might not know what the individual peaks are if you can see a difference in profile then maybe you can start to use that diagnostically so you can check for prostate cancer from a urine sample and yeah and I think probably the most common usage in terms of what you guys will see is testing for samples so for instance testing for the presence of explosives presence of drugs etc is kind of the go-to technology for that because it's pretty cheap right, gene activity so this you may have heard of in terms of statements in press releases like gene X was more active under these conditions so I wanted to quickly dive into how on earth do you work that out a quick dive into transcription science so I don't know how much you guys have come across this but you've got DNA and then DNA is read in a process called transcription so proteins read it and then they create something called mRNA which is effectively a copy it's a single-stranded copy of the DNA the mRNA is then read by another protein and that's turned into a protein the DNA is read to make mRNA and mRNA is read and that makes protein quick aside this is one of my favourite pieces of art which is called Dance of the Polypeptides it's in the States and it's a completely structurally accurate depiction of a protein being made so you have the mRNA depicted as the line the ribosomes making the growing protein a touch I quite like is that the structure of the protein is accurate for the thing that's coded on it so they really went to great effort to make it accurate so in your DNA you've got a few things I think the important things to focus on here is you've got what's called the coding region and that's the bit that the protein actually comes from so at some point that bit is read and they make protein but then upstream of that you've got this kind of promoter area which is where other proteins can attach and that's where mRNA is made so some proteins will attach and cause more mRNA to be made and some will attach and cause less and that's where you get this whole idea of gene activity with them being active in different times and in different ways so how do you measure it well the old way is you take your DNA and you replace your gene with a gene that you can measure so in my case I used a protein that catalyzed a chemical reaction so by running the chemical reaction you know how much of your protein was in there and you can also use fluorescence and things like that however there are some arguments against this technique such as obviously you're interfering with the DNA downstream so that might affect the results and it's also considered a bit old world when I was wrapping up the new world was this way so I say DNA is turned into RNA so you get this single stranded version so what you do is you make a little bit that codes for the thing you're interested in the RNA you're interested in they will naturally they will naturally associate and when they do you can detect a fluorescence so you attack the chemical marker that when they have bound it fluoresces and so you can start to measure the amount of mRNA in the cell and that's how you can measure the gene activity now in terms of usefulness this is mostly useful for scientific research that aren't currently as far as I'm aware commercial or application of this because it's quite a lengthy process so it's not something you wouldn't go to a hospital and they just quick a jab and tell you how active your gene are but it's something they could take a sample and then maybe tell you later down the line but it's something I think we might see more being more common quickly gravimeters so these ones are pretty interesting so they measure local gravity now gravity is obviously gravity but it's based on mass so the more mass the more gravity and on the earth that's the major mass so that's where most of the gravity is coming from but all of us have gravity very subtle differences in gravity depending on how big something is, how dense etc so measuring it very very precisely can be interesting and here's how you do it the old way is what's called a spring perimeter so what you do is you have a mass which you know and then you have a spring which is what's called a zero length spring so it increases proportionally to the mass so to the force on it and then you measure how much the spring is stretching so the weight is going to be the mass times the gravity so any change in the spring is likely due to a gravity variation now the problem with this is it's relative so if you have two machines you can't compare them because there's going to be very subtle differences in the springs but if you have one machine you can take multiple readings with that one machine and move that machine around and get valuable information the current state of the art for relative is what's called a superconducting gravimeter and what you do here is you get a neobium core and you suspend it in an electro magnet so the core is trying to fall under gravity but you're keeping it up with an electromagnetic field now the problem with this is it's relative so you can't compare two machines because of subtle differences in the set up but you can get relative values to give you an idea of how sensitive these are I was reading a thesis by one of a guy in a Finnish lab who was working on this and they found they were able to detect the gravity difference and they found they were able to detect the amount of energy you need to supply to keep the field going is going to be a function of how much gravity there is so the more it's trying to fall the more energy you're having to put in so you can measure gravity incredibly precisely this way and they found they were able to detect the gravity difference from when one of the workmen cleared the snow off the building that they were in so just the weight of snow they were able to detect the difference in mass from that not being there anymore which is staggering if you think about it there are some absolute gravimeters so the most common absolute gravimeter is you get a vacuum tube and a mirror and you drop the mirror while timing it with a laser so you have an atomic clock and a very accurate laser measure and you time how long it took the mirror to fall in the vacuum because you're a very precise measurement for gravity that is comparable so once you've corrected for noise you can compare different machines at different places these are pretty accurate current state of the art for absolute is what's called an atomic gravimeter which is where you have a cluster of atoms held in a laser trap in a vacuum you then turn it off to let them fall you then hit them with a laser to cause them to break apart so you've now got different masses happening they'll fall under gravity and at the bottom with another laser to recombine them and then look at the differences and so you know what happened that gravity was the same but the masses were changing so you can work out what gravity was at the time and these again are super accurate a story I read was about when Stanford uni installed one of these they were trying to calibrate it in they found they were able to detect the mass of the moon passing over which again I find staggering the other thing is they found there was a slight offset every Saturday and they couldn't work out why until they realised they were detecting the mass of the crowd at the local store sports stadium which was just one building over incredible stuff so what can you use it for a lot of earth science stuff so again you can detect differences in density like if you've got nothing but granite beneath you that's going to be very different if you've got granite and limestone beneath you so you can get interesting readings about the earth and it's used very extensively for that a major commercial application is that that lets you detect oil to rock the most interesting application I've come across though is actually submarine navigation so the Americans for a fair while now have been using gravimeters to detect where the sub is under water like how close it is to the rocks because you don't want to send that stone arc as other people might detect that so a gravimeter is a completely passive measurement an interesting aside here is that to film The Hunter of Red October the Americans let them film it in a real stuff and there's one scene in which you overhear one of the crew members shout out a reading in milligals and milligals is a measurement of gravity and at that point the Soviets realised the Americans must be navigating with gravimetry because at that point that was classified as top secret right final one final one AFM, atomic force microscopy and we're going to go into magnetic force microscopy so you may have seen this in news articles where they have incredibly detailed pictures of very tiny things so here there is a this is DNA and you see here a 200 nanometre scale so to move incredibly accurate microscopic images how do you get them way of taking pictures of really tiny things how it works is actually quite surprising so I want to quickly go into normal microscopy first normal microscopy is you get light you shine it through a specimen the specimen interferes with the light and then you look at the light on a very small scale light is quite big so today we came up with electron microscopes where you get a stream of electrons ordered in as far as you can because electrons are smaller than light you can get a better resolution you can see smaller things because there's more disturbance that can happen so how do you do AFM? what can you do that's better than electron microscopy? introducing the science stick so atomic force microscope is effectively a very very sharp stick on the order of atoms so at the end it's incredibly sharp to the point of atoms and what you do is you very literally drag it over your sample and then use a laser so have it on a cancer lever bounce the laser off the tip and see how much it went up when it hit stuff that's honestly how they work so what can you use it for? you get incredibly high resolution images so here are two pictures of atoms so you've got here carbon rings and you can actually see the carbon rings the fact that this crude technology can produce images like that blows my mind however, this is reaching into my current idea and one of the more interesting application of this is purezix so to dive very quickly into purezix traditional hard drives are spinning platters you may have heard them called and they're basically magnetic storage so data is stored as 1s and 0s on the magnetic drive and what that means is a 1 is magnetic and a 0 isn't so if you get an atomic force microscope and you magnetise the tip you can actually start to measure the magnetism of something so you can actually produce a picture of a hard drive like this now in this the peaks are the ones and the troughs are the 0s so you can start to measure very very accurately the magnetism of specific areas this is interesting is because it doesn't what it does is it takes the index of where that file is stored and it raises it in the index so if you just read into the hard disk you can often find the file that was there because it's not been overwritten if you want to securely erase something so that that can't be done what you do is you have to write that data over with more data so that you can't just do this it's now gone however, magnetic force microscopes are so accurate that you can determine the difference between a 1 that was overwritten with a 0 and a 0 that was overwritten with a 0 so even if you've overwritten the data previously before it was overwritten now this is quite obviously a very costly and time consuming processor it's very rarely done but I think it's a really interesting application of the science so conclusions however tiny things are people seem to be able to measure them because they try hard enough and sometimes these harebrained things actually end up quite useful in the everyday world I realise that's been quite a whirlwind tour so I hope that was interesting I'm around if people want to chat about any of this more thanks for listening