National Automated Fingerprint Identification System

Transcription

Claude Roux: So my name is Claude Roux I am a Professor of Forensic Science and the Director of the Centre for Forensic Science here at UTS. It is with great pleasure that I thank you for coming tonight to attend this fantastic event. I hope it will be fantastic and listen to what Dr Xanthe Spindler has to say. Xanthe is a Chancellor's Post Doctoral Research Fellow at UTS within the Centre for Forensic Science. She is also lecturing in the School of Chemistry and Forensic Science. She comes with a very strong background in chemistry and forensic science. She has an undergrad degree, past degree, an honours degree from the University of Newcastle. Then she moved on to do a PhD at the University of Canberra in the area of fingermark detection. But that was done in very strong collaboration with UTS. So we've known Xanthe for many years now. She's got a real passion for fingerprints. So I am sure she will share this passion with you. You will see really what I mean. I mean first another thing Xanthe has published many, many peer reviewed papers. I don't have the number because we can't really keep track of all these papers. She's been presenting at a lot of different international conferences. Actually next week she'll be at the European Academy of Forensic Science conference in the Netherlands presenting a few papers including a keynote about fingermark detection. Personally it's been a great pleasure to work with Xanthe over the last five years. I've learned a lot of things. First that my organic chemistry is pretty crap and two I could learn what the term coastie, actually means. So please welcome Dr Xanthe Spindler. Xanthe Spindler: Thank you, Claude, so welcome tonight everyone and my talk tonight is titled The Fingerprint Detection Revolution. I hope to take you through a little bit of a history of fingerprints. How the fingerprinting community actually evolved and then take you some through some of the research and the really exciting stuff that we're doing here at UTS at the moment. So the first question I get asked whenever I tell anyone what I do is, wow, is it really like CSI and it's always in that kind of tone of voice. To answer that question well I like to think that all forensic scientists are kind of sexy, really stylish people. Unfortunately we don't go to crime scenes wearing Armani and we don't carry guns around the place. But a lot of the science that you see on TV is based in reality. Particularly a lot of the fingerprint techniques that they use on these shows do have some basis. They may not always get the science right and they may jazz it up a bit to look really good, give it some production values, but there is a strong basis there in reality. The other question I get asked is why fingerprints? Hasn't DNA taken over as the gold standard of identification? Well that's why we're here tonight and this is why I still have a job is that DNA hasn't taken over fingerprints and why? Well first off fingerprints have a very rapid throughput. We can process fingerprints quite quickly. The development techniques can take anywhere between, if you're looking at a powder, say 30 seconds. All you do is get your brush put the powder on and take a photograph. If you're looking at something like superglue fuming it can take up to a couple of hours but even then the identification process is still pretty quick. DNA profiling you actually have to find the stain, cut it out and then do all of your extraction, your concentration and then do you actual profiling as well. So we can get information very rapidly using fingerprints. We can also maximise our chances of identification so if we have both DNA and fingerprint evidence we're twice as likely to try and get an identification than if we just have one or the other. We also have the potential for intelligence gathering and I'll talk a bit more about this later on. But basically some of the new fingerprinting techniques you can actually tell whether a person is a smoker. You can tell whether they have been handling explosives or certain types of illicit drugs or whether they've taken certain types of illicit drugs as well. Last but not least both of the techniques are complementary. So in not all instances someone might not leave a DNA profile but they might leave a fingerprint and vice versa. So why fingerprints why are they such a good technique for identifying people? Well when you're looking - we're told from an early age that our fingerprints are all unique and that is true. Every single person has different fingerprint patterns. Now this is only part of what makes them such an amazing tool for fingerprint identification. It wouldn't be very useful if we had something that was extremely unique, we were the only person in the world who had that particular pattern but it could be easily changed or manipulated in some way to hide our identity. Now fingerprints we can actually tick that box and like DNA as well. We can't change our DNA profile the same as we can't change our fingerprints. Well we can at least try there's usually a lot of pain or a really horrific accident involved. Even then those scars that are formed in that process then become a tool for identification in themselves. They are still a unique identifying mark. They're unchanging across the course of our lifetime. The fingerprints you are born with they form in the second trimester of gestation. So by the time you're about 13 to 24 weeks through pregnancy, the foetus has fingerprints already or it's starting to form them. Those fingerprints that we're born with we eventually die with. They might wear down a bit especially if you've done a lot of hard labour or you're a tradie. Your fingerprint ridges tend to wear down over time, but they're still there and they're still the same pattern. They're also very easy to classify. There are four main types of fingerprint pattern, which we've got up here. We've got arches which are one of the simplest patterns but they're also the rarest. They are only around in about five per cent of the population. So if you have arch fingerprints, well done, you are quite a rare specimen, don't go committing any crimes you'll be easy to catch. We also have whorls, we've got loops and as they sound the names are all classified based on how they look. So a whorl actually folds, spirals in on itself like a snail shell. A loop loops around on itself in the core section. So the core is just the centre of the fingerprint where all those ridges come in together. Then we've got our fourth classification which is our accidentals. That's basically the catch-all term for anything that doesn't fit in the other three categories. They're imprinted on everything we touch so how do we leave fingerprints behind when we touch something. We don't leave an entire layer of skin behind when we pick up a drinking glass. So all of you would have left fingerprints behind on your wine glasses or your juice glasses when we were having pre-drinks before the lecture. We don't leave an entire layer of fingerprints behind. What we leave behind are sweat and sebum. So sweat is that really perspiration that really sort of wet sweat that I've got a little bit of now, I'm a little bit clammy, a little bit nervous. You can also have sebum which is that really greasy sweat. So for people who have oily skin, that's the sebum. Now that rubs off on the surface of our hands and whenever we touch something we transfer that in the shape of our fingerprint ridges just like a stamp. This is what we exploit in the case of when we develop latent fingerprints. Now it wasn't always the case that we used fingerprint development. Basically a lot of early cultures such as the Chinese and Mesopotamians used fingerprints as a way of identifying each other. They tended to stamp their pottery with their fingerprints in order to identify who made a particular item. Now as with a lot of knowledge from the Ancient World this was lost in the Dark Ages. French authorities used to brand their prisoners in order to identify them in the Middle Ages. Now you can imagine that fell out of favour quite quickly and as policing and forensic investigation started to evolve in the Victorian era we started to look at other methods of identifying prisoners within the prison system itself. Now these weren't methods to identify offenders. They couldn't go out to a crime scene as such and identify someone in particular. But they could use it to say oh this person has been incarcerated before this is their second offence or this is their fifth offence, we've got a record of them already. So in 1880, 1881 Alphonse Bertillon determined this method called anthropometry. Basically shown here they used to use measurements and photographs of a person's basic body measurements. So the length of their arm to the tip of their finger, pretty much any sort of standard height measurement, also measurements around the face. You can kind of think of this a very crude precursor to a lot of the biometric systems like facial recognition that we use today. Now this was standardised in 1881 by Bertillon and a lot of places used to use this or a lot of jurisdictions used to use this to identify their prisoners. Now basically around this time a few other people decided that fingerprinting may not actually be the best method, sorry the anthropometry system may not be the best method of identifying people. So we had five researchers here across the globe situated in Japan, India and in England and Argentina, looking at whether you could use those fingerprint ridges, those fingerprint patterns in order to identify prisoners. It was Dr Henry Faulds and Sir William Herschel that really did a lot of the legwork at the start to really identify whether fingerprints are unique and whether they stay the same across the population. You can imagine the amount of data they had to collect to do that. It was a monumental task. Now it was around this time that Sir Henry Galton got involved with the help of Charles Darwin. Basically they started to look at identifying, writing this book here, Finger Prints, which is a text book that we still revere today. This is a photo of us revering said text book at a conference last year, the International Fingerprint Research Group Conference in Sweden. Lisa, could you just go check what that was about? Now around this time an Argentinean researcher Juan Vucetich started to look at whether they could use fingerprints to identify the prisoners already in the Argentinean system. Around the same time Galton was tirelessly trying to get the English authorities and Scotland Yard to sort of come on board and to talk about fingerprints. He managed to get them to at least think of it as a complementary technique to anthropometry. So you've now got all of your body measurements your photographs plus the fingerprints of the prisoners in the system. Now in 1984 there was a rather brutal murder in a small county in Argentina. A man had attacked a mother and her two small children. The two children died. Now he left many, many blood soaked fingerprints at the crime scene. Back in these days fingerprint powders which we kind of think as the fall back for fingerprint dusting and fingerprint enhancement weren't actually all that common. They were around but they weren't really known about and weren't used. So they were really reliant on having some sort of colour contamination on the skin in order to photograph those fingerprints and to see them. Now the suspect had denied and denied and denied that he was involved. They finally took his fingerprints and when they compared them to the marks in blood that they found at the crime scene they eventually convicted him. It was at that point that fingerprints were starting to be used in Argentina as form of forensic evidence in their own right. They had now moved as a way of identifying prisoners to identifying offenders at crime scenes. Meanwhile back in England, Galton was still fighting pretty hard to get fingerprints used as a piece of forensic evidence. The 1900s dealt two major blows to the anthropometry system and these were the two death blows. The first was a robbery of a set of billiard balls. Yes billiard balls on Denmark Hill in England. The crime scene investigators found many dirty marks, fingerprint impressions on a window sill that the thief had used to enter and exit from the building. They identified these to be from an offender known as Harry Jackson who was known to police for other offences and he now has the dubious honour of being the first ever person in the Commonwealth to be convicted on the basis of fingerprint evidence. The other death blow was an unfortunate incidence at Fort Leavenworth prison in the US. A young man by the name of Will West was taken to Fort Leavenworth and convicted, or was charged with several crimes. Now he had again denied that he was involved in anything else. When they took his measurements took his photographs they found that he was - that he'd already been in the system and started to question him about prior offences. Now again he denied he'd been involved in anything, this was his first arrest. How could have he done any of these other crimes that he was purported to do. The prison officers didn't believe him, they thought it was just a case of another convict being coy about previous offences to try and get a lighter sentence. He also matched the name of William West who was already in the system. The problem here William West was still incarcerated in the exact same prison. At this point countries started to abandon the Bertillon system. We were the first to set up a fingerprint bureau I 1903 and New South Wales had already taken the measures the year before to start introducing fingerprinting in the prison system. Victoria, Queensland, South Australia, Tasmania - oops, what have I done there? Hello, ah there we go, technology it's a wonderful thing. So New South Wales and Victoria, Queensland and South Australia, Tasmania, Western Australia, the Northern Territory and then the ACT, on the bandwagon right at the end followed suit and set up their own individual fingerprint bureaus. The ACT to be fair to them were probably under the umbrella of New South Wales Police for quite a long time. Okay, so who stole my wine bottle? Did anyone actually see the thief? Yes, no? Well Lisa has kindly chased down two potential suspects here. So girls, they're both students. I understand I've been an undergrad before I know what it's like to be desperate for a bottle of red. So who here thinks it was Claire here? Unidentified Speaker: Yes of course it was Claire. Xanthe Spindler: Who thinks it was Leanna? So it's very hard to see from up here. But okay so we've got a few people. Who doesn't have a clue at all? The majority of the audience, nicely done. Okay so we can't separate these girls based on their measurements. They're both very similar height, they're very similar builds. They're both wearing the same clothes. If we were to basically use just the Bertillon system we wouldn't be able to identify who stole that wine bottle. Also using eye witness accounts, again we can't identify who is who because it happened so quickly, it was dark. You can't really tell them apart. Now there are two ways we can identify a person, the first is DNA profiling. Unfortunately you can see here I don't have any DNA profiling equipment with me. So I'll have to go with a much better method fingerprinting, sorry, biologists, sorry, Tamara, I'm biased. Can you please thank our lovely suspects here? [Applause] Xanthe Spindler: Thank you, girls. Okay so safe work practices and to protect the evidence I shall glove up. Now powders have been around since 1891 so they've been around for quite a long time and in fact before fingerprints were used as a forensic tool. They were very much concoctions of your own devising. There were no standard recipes. You couldn't go to a forensic supplier like we do these days and buy a specific powder that you like. Most of them consisted of pretty much anything that would adhere to fingerprint ridges. That was the only criterion for making a good fingerprint powder. Soot, which is the base of a lot of early fingerprint powders is still used today and the black powders that are commercialised are still made primarily of soot. So what I have here is a fingerprint powder known as Blitz-Green, as the name suggests it's a lovely green colour. This is a magnetic powder so we've also got magnetic iron particles in here which makes it much easier to apply. Most of the powders that you may be used to seeing are applied with just a standard brush. They are good but you can damage the print if you are a little bit heavy-handed. In our labs we tend to find girls are quite good at applying the fingerprint powders as a general rule because they're used to applying foundation and using a feather touch. The boys on the other hand, it's really a 50/50 bet as to whether they're going to wipe their evidence off the surface the first time they do it or whether they are going to get a usable print. So I've got the bottle here, I'm just going to take some of my powder trying not to spill it everywhere and just brush that over the surface and try and find some fingerprints. Ah, I've got lots and lots of fingerprints here. Okay ooh, this is actually live, folks, you can tell because it's a not so delicate process. Okay so I've got a few fingerprints here that might be quite useful. But it's quite hard to see because of the colour of the bottle itself is green. We've got a green powder and obviously from a distance it's quite difficult for you guys to see. Now if we want to improve contrasts there are a couple of different methods we can use to really bring up those ridges. The first is to just use different angles of light and different coloured filters to try and block out the background so we get really dark ridges on a bright background or really bright ridges on a dark background. The other thing we can use is something called luminescence. Now who here has seen one of these before? Yes, it's a typical glow stick. Now you can see here it's got some colour just as a visible colour itself. If we want to make that easier to see we snap it, mix the two chemicals in here together, give it a shake. A handy hint for those of you still doing High School chemistry, shaking things makes reactions go faster. So now you can see that's much brighter than what it was before I'd mixed those two together. Now this is a specific form of luminescence called chemi luminescence. So there's two chemicals mixing to form a reaction, which then gives off light as a by-product. The method we're using today is just - it's a slightly different form of luminescence. So what you do is you shine a light source onto the powder and what the powder does is it absorbs that light and then emits light a longer wavelength. So if I'm shining green on there it should emit around the yellow region of the spectrum. Again you need the right tools in order to see this, science is sexy, right? If you do that you can see the marks luminesce. I am not just going to let you take my word for it, but if we take one of the fingerprints from the bottle, you can see it's quite a nice fingerprint. We've got good ridge detail there. We should be able to get an identification. There is a bit of a problem with curved surfaces like bottles. You can get some distortion in the ridges but they're still identifiable. So here we've got a whorl pattern, you can see the core of the fingerprint spirals in on itself like that snail shell pattern. We've got these two triangular regions here where all the ridges meet and then diverge. These are known as deltas. It is these two points which are the first basis for a fingerprint identification. Now if we have a look at Leanna our first suspect's fingerprints she has predominantly central pocket loops. So this or double loops so this pattern where you've got two loops curling in on each other and also arches. So like I said, Leanna shouldn't really go committing any crime. She's actually a very small sub set of the population. We can immediately exclude her as the suspect because she doesn't have any whorl patterns in her fingerprints. Now if we take Claire for the other hand we've now got three whorl patterns that could potentially be hers. So have a look at these two in more detail. We've got the same basic core structure and the two deltas are in the same place. We've also got many different bifurcation points so where two ridges will meet into one or one ridge will split into two, depending on which way you want to look at it. You can also see from the general flow of the ridges, now again taking into account that there is a little bit of distortion because we are looking at a curved surface. So we're not looking at a flat fingerprint put down like in this exemplar here. There is a little bit of distortion because of the gripping of the bottle. But pretty much from the flow of the ridges, work it out. Look at the minutiae which are those individual points and they're what's randomly formed and that's what we use for identification. So from that we can pretty much work out that it was Claire that stole the bottle. We'll forgive her this time round, okay. Now for decades powders remained the only way for enhancing fingerprints. So we're talking from 1903 through to about 1954, pretty much the only thing you had to use were powders. Now in the early 1950s two scientists, Odén and von Hofsten came up with an ingenious idea, like all scientific advances, which was equal parts brilliant thinking and serendipity. They were studying a chemical known as ninhydrin which is used for detecting and measuring how much amino acid or how many amino acids are in a particular biological sample. For about 40 years before they'd started their research biologists had been writing in all their notes, please wear gloves when using ninhydrin. Any paper equipment that you work with such as filter papers or chromatography plates will develop your fingerprints if you touch them before adding ninhydrin to the experiment. Now it wasn't until the 1950s that Odén and von Hofsten thought, ha, I wonder what's actually causing these mysterious fingerprints to start popping up whenever we touch these filter papers without gloves. They started testing this. This example here is of a ninhydrin developed fingerprint and pretty much any paper surface that you've touched whether it's a fraudulent cheque, allusions to, I did say I wasn't going to mention the Mickleburgh case. But since it has been on TV recently the Perth Mint swindle, part of their central evidence was a fingerprint on a fraudulent cheque enhanced with ninhydrin. It wasn't faked, don't believe the movie. So from then on, 1954 was really the catalyst point for fingermark detection or fingerprint detection to take off as a discipline in its own right. From there we started seeing the development of techniques such as super glue fuming. So basically any of you who have used superglue or ethyl cyanoacrylate to give it its proper chemical name will know that when you allow it to set in air it forms this really hard white polymer. Now if you vaporise that so you heat up to about 80 degrees and you place it in enclosed chamber with something such as this smiley face ball here or a bottle of water or the bottle of wine. You allow those cyanoacrylate vapours to deposit, to flow through, they will actually deposit in the fingerprint itself. It prefers the sweaty components in the fingerprint and then it will start to harden into that white polymer. Now we can also, if there is not enough contrast, say for example you've got a clear bottle and a little bit of a white polymer forming, it's not very visible to the naked eye, you can't get a good photo of it. We can then add stains. This is an example of a stained fingerprint developed with cyanoacrylate up here. Now the Australian National University have also left an indelible mark on the forensic community. Has anyone here heard of a little invention called the poly light? There's a few people. It's basically a box - it weighs about 12 kilos, it's about that big. It's a forensic light source. It's been made famous a few times by shows like CSI. Basically we can see anything, form any light or use any light from the UV, the ultraviolet all the way through the coloured spectrum up to the infrared. Now this was actually the result of research done at ANU in the Eighties. The other thing that came out of it was an adaptation to the ninhydrin method. This is one of the researchers from the ANU Group back in 1984. Basically what they looked at was adding zinc salts and cadmium salts to the ninhydrin reaction. This gave them a better result. It allowed them to see fingerprints that may have been quite faint using ninhydrin and basically these now suddenly appeared. They were now suddenly usable evidence. Although it looks like from this that we've got a fairly comprehensive suite of fingerprinting techniques. There are fingerprints that still aren't captured by these techniques. We're still missing evidence and frankly that's not good enough. Some fingerprints may be too weak, so someone may have very dry skin, they don't have enough sweat there in order to leave a good fingerprint that we can detect. Maybe they haven't touched the surface thoroughly enough or hard enough in order to get transfer of the fingerprint ridges or the fingerprint detail onto the object. Maybe the object is just too old, it's been sitting in the cold case storage for 40 odd years. Maybe we just can't get a fingerprint off it anymore because it's been and gone. Now as a result we're constantly having to search for better techniques, more sensitive techniques in order to get usable fingerprints off old, damaged or very weak evidence. This is where it gets complicated. Our sweat and by extension the latent fingerprints that we leave on objects are made up of thousands upon thousands of different chemical compounds. Now you can imagine this makes it quite difficult in order to come up with a technique which will work that is robust. That will work in a lot of cases or the majority of cases. Now when a crime scene investigator goes into a particular scene the fingerprints are latent. They are invisible they don't know where they are. Much less they don't actually know what the chemical composition is at the time they go in there. They don't know how old the fingerprints are. Basically our mission statement is to come up with new techniques that allow us to see these fingerprints that work in the majority of instances or hopefully in all instances. It sounds simple, right? The challenge is to work with something that is so minute down to a billionth of a gram of material. We've got to try and target that in place so that we're not pre-concentrating the sample or taking it off the surface. Usually when chemists do analysis of very low concentrations or detection of very low concentrations, they'll extract it from whatever matrix or whatever surface they're looking at. They'll then pre-concentrate it to the smallest possible volume and then analyse that so that they get as much material as they possibly can to analyse. If we did that to a fingerprint we'd actually be basically a lost cause. We would have to destroy the fingerprint in order to make it a stronger signal. So not only have we got to work with a very low quantity, we've got to do it without destroying the ridge detail. How many of you are fans of shows such as CSI? What about NCIS? Not as popular, what about Bones, fairly popular, okay. Now the fingerprint detection methods that they use in these shows aren't actually as robust and don't give us quite the same results in reality as what they do on these TV shows. Like I said they need good production values in order to make it watchable, in order to make it entertaining. Even we're not immune to the sexiness of these techniques that they use on TV. We want to be able to do what they do. We want to be able to get the results they get. This is the real VMD, this is the real vacuum metal deposition apparatus. This takes up its own room in a standard police lab. It's so loud that even with the door shut you can hear it at the other end of the building. Now basically the description they gave was fairly accurate. It does use gold, it does use zinc and they are vaporised using an electric current. The gold deposits first and then the zinc deposits on top of that. Unfortunately we can't quite get the same results that they go there. Now vacuum metal deposition is a really handy technique. It is expensive and it is cumbersome to use but it gives us some really good results. Say for example for polymer bank notes. When the Reserve Bank changed the Australian currency over to the polymer bank notes from the paper currency that we used to have, forensic scientists everywhere started banging their head against their benches. Basically the polymer bank notes are really hard to get good fingerprints off. They're a really bad surface, they are what we call semi-porous. They are not quite porous so we can't use techniques like ninhydrin on them. They're not quite non-porous. So cyanoacrylate fuming which we would usually use and powders don't work quite that well either. It's kind of that in between and it gives some really weird results. The other problem we've got is you can see the printing is still visible. So this has been developed with cyanoacrylate and then a stain called rhodamine added to the top of it. You can still see that the $10 insignia and some of the printing, intaglio printing which is actually raised. If you feel the surface of the bank note all of that black printing is slightly raised from the surface. If you use vacuum metal deposition, slightly different lighting conditions you can get quite good results. You get rid of all of that or the majority of that background to get a usable and identifiable fingerprint. Now this was work that was done by Naomi Jones, now Naomi [Spiers] during her PhD thesis here at UTS. This work has continued recently, an honours' student of mine from last year Tristan Merten was looking at the use of single metals rather than the gold zinc, looking at using just silver or just copper. Basically what he found is that these single metals give comparable results in some instances. Silver is quite good but then just as a failsafe he can then apply gold, zinc, the standard VMD technique afterwards. So basically we like sequences, we like to be able to say okay so if this technique doesn't work, it doesn't give us a usable fingerprint we can move onto a better technique and not destroy the evidence and just keep going until we get a result. You can see here results on fabrics as well. So part of Tristan's project was to look at whether we can get identifiable fingerprints off fabrics. In the case of nylon it's actually quite feasible. Nylon and polyester they don't absorb water very easily so they don't absorb fingerprint residues or sweat very easily either. If you're looking at something like cotton you're going to get all of that moisture absorbing in. After about an hour all you end up with is a hand shaped or a finger shaped blob. Even then after quite a while with these nylon and polyester you're still going to lose that ridge detail after a couple of weeks. So unlike what they can do on Bones, we can't actually identify just a random fingerprint or a random hand print on fabric just yet. Some of the other research that we've been looking at is down to the scale of nano-powders. Now most of the powders that we use at the moment, the particles are a few micrometres wide so basically a millionth of a metre or a thousandth of a millimetre across in diameter. What we are now looking at are nano-particles. So things that are a degree or order of magnitude smaller than that. The whole idea behind this is that the smaller you go ridge, fingerprint ridges themselves are only quite small. So if you have a small powder a very fine powder deposited on that fingerprint adhering to that fingerprint the smaller the particles the better definition that you get. Again this is an example of some work that was done by Mi Jung Choi again another former PhD student of UTS. This particular example here was a zinc oxide powder that was doped with lithium. So taking ideas from the materials' world so building things such as better semiconductors and basically applying that in a fingerprinting context. This is a lovely little micrograph of the powders on the fingerprint ridges themselves. She also looked at using other types of nano-particles, so titanium dioxide, which allows you to get fluorescent ridges. So we're looking at the development of fluorescent nano-powders or luminescent nano-powders as opposed to just coloured powders. We've also got magnetic powders so zinc oxide with our iron magnetic particles in them that can be applied with this style of brush here. Again instead of using the standard squirrel hairbrush. That was performed by a German research intern that we had a couple of years ago, Ron [Yerog]. This is something very close to my heart, this is actually part of my PhD work. So not only are we trying to develop new techniques for fingerprint analysis, or what I like to refer to as conquering the world. We're looking at trying to understand the world. Trying to understand what's actually going on at a molecular level. Reaction by reaction, atom by atom, trying to work out why in some instances we get really, really good fingerprints and in others you can barely see the ridges at all. One of the methods that's quite regularly used these days is known as indanedione-zinc. This is the little sister of ninhydrin. It reacts in a very similar way it's used in the same instances but it's much more sensitive to the climate. So we were finding that even though we were getting really good results with this here in Australia the US were getting really good results and so were Israel. Places like the United Kingdom, Canada and even New Zealand were getting quite poor results with indanedione-zinc. For years and years and years we couldn't understand why. This saga has been going on for 15 years and it's only just starting to be resolved now. So basically trying to understand how to go from this, which is a very poor result, to a nice, bright, good yield of - or good fingerprint development. Basically what we're finding is that once we understand these reactions we can say okay so you've got a really dry climate, there's not a lot of or it's below 20 per cent relative humidity, really, really cold, you need to this in order for this reaction to work. Oh you have this type of paper, oh you should use this technique instead. It's not really feasible to use this one. Now one of the other things and this has been in the media recently that's come out of UTS is looking at something known as the Thermal Fingerprint Device. If you - there were two research students again honours' students, Adam Brown and Daniel Sommerville back in about 2006 who determined that if you heated fingerprints in a certain way. So you just put fingerprints on paper in a heating press or under an iron for a particularly - anywhere between about I think it was two minutes through to about five minutes, you actually got this charring of the fingerprints. So you get brown fingerprints on a white background. Now when they did it for shorter periods of time at very high temperatures, so for example 30 seconds at about 220 degrees they were getting these fluorescent ridges here. No chemicals or no actual reagents or processing involved, just heat. So we thought, okay we're onto something here and so did Foster + Freeman. They've now marketed, this is the second generation. Basically what this relies on is that the sweat in your fingerprints causes this - degrades rapidly at high temperatures compared t the paper itself. So you end up with this difference in the thermal degradation or the charring and burning of the fingerprint ridges compared to the paper. This is really, really useful in instances of covert operations. So for example if you intercept something during an intelligence operation, you want to be able to identify fingerprints on that evidence but you don't want it to be visible when you put it back into circulation. Now if you use a standard technique which gives you a colour reaction so ninhydrin and indanedione both give you very brightly coloured fingerprint ridges once you use them. You want something that's covert, you can just stick it back into the system no one is any the wiser that you've actually done anything to it. The TFD gives you these luminescent prints. You can't actually see any difference if you look at it with the naked eye. As I mentioned it was the new release of the second generation has been in the paper recently. Sometimes we just need to look to the other end of the rainbow. Most of the techniques we use look at around the violet to the green region of the spectrum. So we're looking at the high energy part of the spectrum. A lot of other substances actually luminesce at the same wavelengths. Say for example the barcode on a Fanta can, you can see here using something that has - that gives you a light emission around the yellow region, you can see that the barcode is actually interrupting that fingerprint detail. Now if you have a look at something like a Coke can or any sort of soft drink label under something like the poly light, it doesn't matter what wavelength you choose that thing is going to light up like a Christmas tree. This makes it particularly difficult to get fingerprints off this kind of evidence. So what we need to do is actually look to a different part of the spectrum towards the red and the near infrared in order to get a good fingerprint development. So this particular example here is something called STaR 11 and this is from a finishing PhD student this year by the name of Scott Chadwick. Basically what he's done is combined two different dyes, styryl 11 which is a near infrared dye, it emits just on that cusp of the red and the invisible part of the spectrum and rhodamine which is a common stain. It's that stain on your left that emits in the yellow region of the spectrum. What you can do by combining two dyes is that where the rhodamine emits light so it's absorbed in the green region of the spectrum. We've shone the green light on it, it then emits in the yellow. The stryrl 11 actually absorbs the light from the rhodamine and then emits in the near infrared. So what we've done is doubled that gap between the light that we're shining on the sample and the light that we're actually viewing. This gets rid of the a lot of the background interferences. So we can actually view more fingerprints on really difficult surfaces. The first method he tried was just using it as a stain for cyanoacrylate fuming and since then there's been a fair bit of work done on this for powdering on difficult surfaces. So this example here is on a laminate bench top. The good quality fingerprint is the STaR 11 combined with titanium dioxide. The really poor quality print is actually Blitz-Green. It's the same thing we used on the glass. Now it performs really well on smooth surfaces and a lot of the traditional powders work really, really well on those smooth surfaces like glass. As soon as you introduce some texture things such as a laminate bench top you start to get really bad results. That's why we need to start looking at new powdering methods. It's also looked as a method for getting fingerprints off the sticky side of tapes. Now you're probably thinking why would we want to do that? Cases where someone has been bound with tape, a package has been bound with tape you can sort of see where this is going. The sorts of things that you could package up with tape that you don't actually want to be found by police or customs. These things do need to be analysed and the better the method you've got for getting those fingerprints from that evidence the more likely you are to get a conviction. Now taking that one step further, now we're not just looking at the opposite end of the light spectrum, we're actually looking at things slightly backwards. This is work done, you'll have notice the insignia there for those of you who are quite astute is actually part of a bank note. You can see the Australia writing down the side. From memory this was a $5 note. When you - this came out of some work by Elicia Bullock. What she found was when she used a particular type of powder called an up-converter that she could get really good ridge detail from fresh fingerprints off these polymer bank notes. Now how does an up-converter work? I said we're doing things slightly backwards. So instead of shining our light source at the short end of the spectrum and moving to a longer wavelength, we're actually doing it the opposite way. An up-converter takes a long, low energy wavelength of light and then emits a shorter high energy wavelength of light. So in this there's very few things that do this naturally. Basically you have to make up-converters in the lab. You can't find an up-converter in the commercial environment. You can't find it in the natural environment. So what we've got here is a method that can eventually get rid of background staining or background interference with no real hassles. So this has been continued throughout Rongliang's thesis and you can see with standard techniques again cyanoacrylate fuming, very poor ridge detail on the left. As soon as you use the up-converter you can start to see the ridges and you're getting very minimal disturbance from that background. So you've now got usable evidence where you couldn't get usable evidence before. Again you might recognise the name on this slide. Taking the next step and this what relates directly into my current work here at the university. Now what we need to do is really focus on a non-chemical technique in order to get the next generation of fingerprint reagents. Most of what we look at involves chemical reactions. You're taking one component reacting it with another to get the same amount of another component. What we want to do is take one component react it with something, or detect it with something and then get a signal that's 10, 20, 30, 40-fold greater than that original amount of material. So you can see where this might come in handy, really weak fingerprints. If you've got very little sweat there to begin with you want something that's going to give you the best possible visualisation. You want something that's going to give you a very strong signal for not much starting material. Now this all came out of some talks with a colleague in the US by the name of Oliver Hofstetter. Basically we contacted him and said, hey we're interested in some of your research. We've just found out that you've made antibodies. So the same proteins that are responsible for keeping you healthy, part of your immune system, they're what attack germs, viruses, any sort of illness to keep you nice and healthy. He'd actually raised some to target amino acids which are a natural component of your sweat. Now he basically said, here, have a couple of grams, see what happens. We're still in contact with him, we still work quite closely with him. What we found is when we combine these with nano-particles and then put again a luminescent tag on them we're actually getting fingerprint ridges. It's pretty exciting especially considering this was the first time we'd been able to get amino acid rich fingerprints on glass. Usually you use or you target amino acids on porous surfaces because they tend to be more stable. We've managed to do it on a non porous surface which was pretty exciting. I think from memory there were cartwheels down the hall when that happened. Last year the media finally got hold of it. For those of you who listen to BBC Media, you may remember me stumbling across my words on radio, which was kind of exciting and kind of frightening at the same time. Since then we've actually had a spin off project start. Michael Wood, again another finishing PhD student, who's been looking at the use of aptamers. Now aptamers can kind of be thought of as synthetic antibodies. They're made up of DNA rather than proteins but they do the same basic job. He's got some really, really good results. These are hot off the press that he's presenting next week at the European Academy of Forensic Science Conference. I'm slightly miffed and slightly proud at the same time that his fingerprints look better than mine. So what does the future hold for fingerprint detection? Where do we go from here? This was just a snippet of some of the research that's out there. There's been work on new ways to detect blood marks, so not just looking at photographing them. What do you do with blood marks that might not be quite visible? They're really weak, there's only a little bit of blood on them. Looking at techniques which give you really, really fluorescent fingerprints or really luminescent blood fingerprints. A lot of the current techniques you can get some ridge detail, but they're not fantastic, they could be better. There's also research that's been going on recently at looking at techniques that not only detect blood but can also detect the latent components. So the non-blood component of the fingerprint at the same time. When we use these you can actually tell the difference, what's blood, what's latent. So not only can you photograph them you can then do a targeted DNA recovery from the blood portion of the fingerprint. I mentioned right at the start of the talk for those of you with good memories about intelligent fingerprinting. So now not only are we looking at an identification of a person, we're looking at a method that can actually tell us what they've been in contact with. Have they been taking drugs? What drugs, are they a smoker? Have they handled explosives recently? There is one particular research group out there that's really driving this particular field forward. Looking at chemical imaging, so you'll notice this picture here at the top of all of the slides. This is what's called a Fourier transform infrared chemical image. So what, a bit of a mouthful. Basically what this image is composed of is it's a normal fingerprint. It's been stained, or it's been fumed with super glue and then they've taken the chemical spectrum, which is basically just measuring the different vibrations of each of the atoms and each of the bonds between the atoms in that fingerprint. From that they can actually build and image. Now at the moment it's quite a cumbersome method to use. It takes quite a while to build up this image, several hours to several days depending on the type of equipment you're using. But in 10 years time, who knows. Technology may have caught up and we may be seeing a chemical imaging method that is actually not only field portable but really rapid. That's a pretty exciting space to be in from both a technology perspective and a research perspective. We've also got what's noted as one of the Holy Grail's of fingerprinting, is being able to determine how old a fingerprint is. Now you can imagine someone has been accused of a crime. The prosecution is questioning them and they say, oh no, I left that fingerprint there weeks ago. I wasn't there when she was murdered. I'd visited her about four weeks beforehand and she was fine when I left. How do you dispute that? At the moment there is no specific way of being able to determine the age of a fingerprint left at a crime scene. Researchers at the University of Lausanne in Switzerland are now working on this to try and really build up a profile and a usable method to say, no you were actually there at the time. That particular chemical composition of y our fingerprint indicates that you were there when she was murdered. That would be again an absolute breakthrough in fingerprint detection. Now I'll leave you on one last note, scientists are essential for the advancement of society. Now, and I'm not just saying that because I'm a scientist. Basically we get into science because of the people around us. Science isn't seen as an exciting career path to a lot of kids because it's seen as dorky or seen as really geeky. Talk to your children about science. Get them interested. Get them involved, get them into things such as Ultimo Science Festival, which is happening right now until the end of next week and the National Science Week which is wrapping up in a couple of day's time. Apart from that, thank you all for being a very attentive and enthusiastic audience. Thank you to our hosts here at UTS and to my lovely sprouts and Lisa for organising everything. [Applause]