Trace Evidence 2011 : Fiber : Claude Roux


[ Music ]>>Our next presenter
is Claude Roux whom many of you probably know already. He’s been on the
program several times. Claude is a professor of
forensic science at the Centre of Forensic Science
at the University of Technology half way around
the world in Sydney, Australia. And he’s going to give us the
first of a couple of talks today on time-of-flight
mass spectrometry. Claude?>>Thank you Keith. Good afternoon everyone. Yeah, I must say that
my simple talk won’t be about mass spectrometry
so one is enough for me this afternoon
about that topic. I’m not a mass spectrometrist. I’m only a forensic scientist. I’d like to first acknowledge my
coauthors especially the first two Jason Lee and
Kelsey Woodhouse. Jason basically did most
of the work here as part of his onus project
which is a short term, a research project
at the University. And Kelsey is currently doing a
followup work on the same topic. So it’s pretty much an update about where this research
is going and I have no doubt that it will continue
for some time. Very briefly because Cheryl
did a great job there. We all know fibers are an
important type trace evidence. A lot of research has been
done on a lot of things and especially when we talk about dye analysis
we are focused on increasing discrimination
of any comparison. I must say that here that fiber
dye analysis still remains a challenge for various reasons. Case work examples, I guess,
so Harrold Deadman is here, Wayne Williams, you know,
he’s a big case in the US. But it looks like every
country had its big case, seminal case about
fiber evidence. In terms of the analysis
I just mentioned before that dye is an issue. I mean when we deal with color and dyes are quite often
we all use microspectometry and we saw an example before. When we want to move
a bit further and get the actual dye analysis, the first problem is
to extract the dye. And quite often what people do in most forensic laboratories
are thin layer chromatography. But as you know, there are a lot of different techniques
that are available. If you attended the workshop
on Monday given by David Hinks and Keith there were
a lot of discussion about various techniques. They talked about LCNS and even
including using a time-of-flight mass spectrometer with the LC. We did something very
similar but instead of the liquid chromatography, we
use capillary electrophoresis, which is another type of
separation technique based on the charge of the analyte
instead of other things. The aims or objectives of the
project, you’ve got them here, so basically the first
thing especially the big job for Jason was to configure the
interface, to interface the CE with mass spectrometer
and then develop a method. See how it would work
on extracted dyes and see how it would
work especially with a QTOF quadropole
critical time-of-flight mass spectrometry. So why CE? We got some reasons here so
it can be highly automated. It’s much more sensitive than
TLC, small sample requirement. It’s pretty efficient. It’s fast. The runtimes are pretty short. It’s highly adaptable to a lot of different types
of separation modes. And here you’ve got an
example with, you know, UV detection of a dye mixture. And here you have a 50 ppm
mixture and it’s not a big deal, you know, something like
less than 12 minutes, you’ve got a pretty
good separation. Now, this is the common problem. It works when you
have a relatively concentrated solution. Now when you extract the
tiny little piece, you know, dye from a tiny little piece
of fiber, we got into this sort of issues with normalcy
UV detection where basically we are
right in the noise. So it’s why we try to move
towards CE-MS where basically after the C separation
we can analyze, you know, peaks and get mass spectra from
the various peaks, same ways, very similar principle as you
have with GC-MS which is used around the world in
forensic laboratories. Now I’ll show some sum-up of
the advantages of adding the MS. So we increase the sensitivity
and obviously we hope that we can increase– you know, at first we can get an
easier identification because we’ve got the mass
spectra we can work with. And hopefully we hope that we
can get a better discrimination. So this is a table, it’s
probably too small for people at the back of the
room but hopefully if you download the presentation
you’ve got a few information about recent works,
recent trends in CE-MS. What people have done
in the last few years. And once again this is what
we’ve tried to do by, you know, adding the MS. So with just–
you know, instead of stopping at that stage what you
would have the detection of the analytes here, we
try to push them somehow to the mass spectrometer
and there are a lot of different ways
to do that a lot of different ways obviously to– depending on the type
of mass spectrometer. So at QTS we are very
lucky because my colleague, professor of analytical
chemistry Philip Doble managed to get a very good relationship
with Agilent Technologies so we have access to a
lot of their new toys. And we had access to this
machine, so, you know, QTOF-MS. So if you remember
Monday’s workshop so with– you know, the TOF tube here. And basically we used this kind of ionization adapter
kit for the CE. Now, how it works– so I
don’t have the little videos like Keith had on Monday. I wasn’t going to risk that. But, you know, keeping
very simply. So when you’ve got the
ions coming through here– so you have a first– you know, some kind of mass
filtering going on here with the quadropole. So you can, you know, be interested pretty much filter
the specific ion you want. Then the ion analytes are
just going through there. They are pulsed upwards
and they go up to a certain point depending
on the amounts obviously and then they will go down on
the detector and, you know, they will– it will take longer
time or faster time depending, you know, primary on the MS. So we can a very
accurate mass detection and it’s very sensitive. So in terms of the dyes we use
for this project, so we use– we worked primarily
with polyester. So we use a lot of
disperse dyes. And we had a few acid dyes. So we bought these dyes from
Sigban and Sigma-Aldrich. And for the experimental
itself, so we started first with these sort of parameters. So here you’ve got the
parameters for the CE and the parameters
for the QTOF-MS so we had start somewhere. So most of the first part of the experiments were
done using these parameters. Then it’s pretty much
the job of Kelsey. She’s been working on trying
to improve the buffers. So she changed a lot of things
with the buffers so I won’t go– I won’t read all that. You’ve got that on the slide. But so she used different
types of buffers there, so known aqueous
buffers as well. So she’d been very busy playing
with the method development. Then she focused on
ammonium acetate buffer and she changed the
concentration. She added some acetonitrile,
changed the pH and ultimately she found that
type of buffer was the optimum for the kind of experiments
we were doing. So ultimately, these are
the optimum parameters that have been used for this
research or at least most of it. And these are the parameters
for the mass spectrometer. Okay, a few results. So here you’ve got an extracted
compound electropherogram of metanil yellow and it’s a
10 ppm extended and, you know, no big problem to get a
mass spectrum from there. A similar thing with an
acid green 25, again, 10 ppm and no big deal
to get a mass spectrum.>>When we go to a dye mixture
at 10 ppm, so here again, we’ve got a few things
here going on. But ultimately, it’s not– you
know, again, we’ve got a pretty, relatively good result
here in terms of detecting this dye mixture. Now, if you remember the 500
ppb I showed you before the UV detection which was
pretty much the noise. So this is what we have
now this time with the MS. So you can see immediately
visually the sort of big improvement we have by
moving towards MS after CE. Now if we push to 20 ppb so we
still retain one of the dyes, terasil yellow which
is still detected. Now there were some
experiments to, you know, deal with linearity, investigate
linearity, a limit of detection, and all that was
done in triplicate. And you can see, you know,
pretty much the big advantage of CE-MS when you compare
to CE-UV both in terms of, you know, their linearity
and the detection limit. I think that’s not
a big surprise and here are some graphs
showing again the linearity with pretty good
correlation coefficients so very good technique
for that purpose. We did some experiments
about precision. So there were six rounds here. And here again, here you have
the overlay of the six rounds with 10 ppm and if you
prefer that in terms of a [inaudible] table. You– Again, you
can see obviously– you now, there is a bit of
variation depending on the dye but still you can
see a big improvement or some improvement
between, you know, when you go from E-CUV
detection to CE-MS. Now the big– I don’t
want to play with the word and say an acid test because
we did some acid dyes. But, you know, that was really
trying to see what it would be when we move towards a
single fiber extraction. So we started not necessarily
with a single fiber first with thread which is the photo
you’ve got at the bottom. But ultimately we did
experiments with single fibers so you’ve got here the
operating procedure. There is no– nothing
really special here as Keith mentioned on Monday. It requires a bit of
practice and sometimes some– you know, an art to do that. And, you know, I was very lucky. So students can do
that instead of me. So here you’ve got
some results based on the 5 millimeter
of a yellow fiber. And here you– it was
a yellow fiber dyed with a terasil yellow dye. And you can see– you know
we can have a relatively good result. So we can detect that dye. We have the mass spectrum. The same things with a red fiber
with terasil red, and again, extracted from a 5 millimeter
fiber– of polyester fiber. So at this stage, this
research is going on. What we want to do is maybe
to pay bit more attention to the mass spec
source and try to– so we’ve got some plans with
Agilent to try an APPI source, possibly further optimize
the CE parameters. There were some effects that
we have to investigate due to the MS source
and CE positioning. And even possibly move
towards another type of MS that again we have access
to– thanks to Agilent, triple quad mass spectrometer to further increase
the sensitivity. I must say at this stage one
thing that we really need to do as well and it’s
always the problem with this sort of
fiber projects. We’ve got access
to a large variety of techniques in
fiber examination. So each time we try to try
a new type of technique. We always have the question does
it bring any value to the scheme of examination we already have? So I’m very keen to do further
experiments and have access to some samples that maybe
people have already analyzed for a full sequence of
analysis so we don’t have to reinvent the wheel
and then we could take that further using
this technique. So if you have access to
such samples, if you think that you could assist and
contribute please talk to me. I’m sure we can find
an arrangement but I would be really
keen to see from a forensic view
point really how much more discrimination we bring. Another thing as well from a
forensic view point is to– because if we add the
technique which is so sensitive then we have the
question whether we bring, you know, some contaminations,
maybe it’s not the best word, but at least some contribution
from environmental conditions such as washing powders,
you know, sun exposure, tumble drying, fabric
softeners, et cetera. There was a whole list on
Monday in the workshop. So that can be positive
or negative. I mean it can be positive if it’s something we can
actually identify and use that as a further
point of comparison, but it can be negative because
it can lead to, you know, false negatives in a case because we’ve got new
peaks, all things like that. And the other things,
again, that we try to– yeah, to do more research
to try to see the effect of an increased extraction
times or increased temperatures, I guess we could go
never ending in terms of the method development. And I mentioned the
assistance required before. So please if you have any access
to such samples and you would like contribute,
you can talk to me. So in conclusion at this
stage about this work, so we think that
CE-MS, you know, has big advantages
over normal CE-UV. So it’s more sensitive. It showed that, you know, a lot
of parameters such as linearity, limit of detection, et
cetera are much better. It looks like it’s a
reliable system to maybe for the identification
discrimination of the dyes. It’s reproducible and, you know,
at least theoretically feasible and efficient alternative to
current method that we have. But obviously, we need to
do more work and hopefully at some stage has the
potential to be useful in forensic case
work and really try to crack this issue
of dye analysis. So a few acknowledgments here. So Agilent Technologies
especially Chris Fouracre who’s a technician that
I’ve been working with at Agilent Technology. Dr. Susan Bennett from
New Southwest Police Force who assisted with some samples. Professor James Robinson
at University of Canberra and the Australian Federal
Police staff who helped as well with the project. And again, if you don’t know
what to do next year or even if you want to contribute
to the project with samples. Deliver the samples yourself! We’ve got this fantastic
Australian-New Zealand Forensic Science Society Symposium
in September 2012 in Hobart which is a very nice
little place in Australia. It’s very green, very different
from the rest of the country. And thanks for listening. [ Applause ]>>The floor we– The schedule
for us to have the floor open for question for
about 15 minutes now so I’ll open the floor. I think we’re supposed to have
a microphone, but I can’t– too many photons in my eyes. I can’t see anything out there. So are there any
questions from the floor? Yes.>>I have a question for Claude. Is the — you say the value
of the techniques more for identifying dyes or for
quantifying the amount of that.>>The first advantage to is
really increase sensitivity because when we– you know, when
we really extracted a tiny piece of fiber, you know, we quite
often have this problem of being below the
limit of detection or in the noise or
very close to it. So that’s the first thing. And then obviously, I
mean, you are right, the ability to identify
the dye would another big– a big advantage. Now, I can see the
next question, the really big holy grail
there in terms of dye with fibers would be the
analysis of the dyes in situ without having to
extract the fiber. Because whatever technique
we use if we have to start by extracting the dye, we already have a big limitation
there whatever technique we use. So, you know, thinking about
your experience with Raman and all the work that has
been done with Raman, I mean, you know, it’s another
way to look at and I guess that will be another
avenue for research to look at the analysis of dye in situ.>>Other questions for
any of the speakers? Okay, you can’t get out
’cause I have questions. The moderator is always
supposed to have questions. I always do this
with my students and they probably feel
the same way, why? Dale, I noticed in your green
fiber the largest differences were at long wavelength. The differences were smaller
at the shorter wavelength, is that a structural feature
that we get the big differences out at the long wavelength?>>That’s a good question. I don’t know the answer to because I have not
characterized these polymer films yet. That’s actually a film so I don’t really know the
orientation of that film. I don’t know enough information
yet about how it’s produced to know the answer to that, so.>>Okay.>>But with the conjugation in the visible range it’s not
abnormal to see a larger shift in the red a bathachromic
shift in the red.>>Okay, might be a good
question for your final exam. Cheryl, red– why? Being from NC State I wanted
to know why you picked red when Dr. Morgan might
have picked another color.>>That’s the exemplars
we were given.>>Okay. And there was
a second part to that. Any ideas why there
was a problem in classifying Reactive Red 2?>>We did notice when we looked
at the structures that it was like an isomer of each other. So we we’re wondering if maybe
that’s why it was really hard to classify one or the other.>>I guess one of the things
that I think about when– most red dyes have two
absorption peaks in them. So by picking red you
had two peaks already. You find a lot of
blues and yellows. They only have single peak. So you started off with a little
more difficult classification scheme there. Claud it appears as if you were
running a negative ion mode, is that correct?>>Yes.>>There– The difference
between acid and dispersed dyes because the acid dyes are
already charged negatively and the dispersed dyes
obviously there’s no problem in charging them. And we’ve seen the same thing
there’s no problem charging them to get them negatively. So I guess that’s
just a comment. The next question is do you know and you may not have done this
yet, can you run cationic dyes, basic dyes in your capillary
electrophoresis instrument? I’m guessing you probably have
to change the mobile phase?>>Yeah we would
have to change that. I don’t see there would
be a real problem there.>>Okay. Other questions
from the audience? [ Inaudible Remark ]>>Yes, yes. Steve.>>I’m Steve Morgan from the
University of South Carolina. In response to your
question about the red dye, [inaudible] provide other
samples to [inaudible]. It turns out that the–
I think the impetus for that project came from
the Indiana State police who were interested in red
dyes because there are gangs that wear read scarfs and
they were finding these fibers in possibly related crimes and they’re trying
to make sense of it.>>Okay. Since I went to Purdue
does that mean they go to IU? [ Laughter ]>>No comment. [ Laughter ]>>Yes.>>I just have a sort of a
comment for Claude actually. The– I’m glad you
brought up the ability to add discrimination. Unfortunately, I don’t have
fibers to provide for you. But my understanding and based
on research that has been done that the current scheme, the
discrimination ability is very, very high like 99 percent
or something like that. So given the potential
for added discrimination, how does that balance with
the cost to the laboratory for such instrumentation.>>That is an excellent
question. I think it’s really– I guess that will be pretty much
case specific really. If you have a specific case which is really high profile
you really want to go to the end of it, you know, for completeness
purposes as they say. Or you have a specific– I
don’t know, specific problem because it’s a specific
type of fiber. There is some kind of
industry information you have and you think that would
really bring, you know, something to the case. But I agree. I mean, on a daily basis
this pool of techniques of [inaudible] already
so powerful and I guess it goes back to
what I mentioned to Chris before that I still really–
the big thing with dye and fibers is really
the analysis on in situ. But whether we’ve completely
cracked that– I don’t know.>>Other question? Yes. Bill? [ Inaudible Remarks ]>>That’s an excellent question. That’s one thing that
I’m looking into to see if the match point of the
birefringents to the liquid or to the fiber is–
we already know that there’s gonna
be scattering effects when the refractive
index don’t match. But that is something
I’m going to look into. I did not change the oils. I try to keep them
all in the same oils. I’ve done this for several
different types of fibers, nylons that are multicolored
blues, wool, and with polyester being
such a high birefringent and it’s gonna cause
us the most trouble, that’s the first one
I’m gonna look at. But my guess is that we are
going to reduce some scattering. But the dichroism
will still be there. To what extent, I’m not– you know, I don’t have
an answer for you yet. But I think that we’ll still
see the linear dichroism there. [ Inaudible Remarks ]>>It’ll be interesting to see
Dale and Cheryl get together and see if dichroism could help
distinguish some of the flavors or some of the dyes that couldn’t be
distinguished in your work. [ Inaudible Remark ] Let me just comment. We have looked at
some acrylic fibers with a very low birefringence
and we don’t see much change and there certainly are effects. Ideally, you’d want to set
up your fiber in one index where they are matching
the index to the fiber. [Inaudible] The way to
cheat is just do acrylics. [Laughter]>>If I can add to that. The acrylics that we were
looking at, they’re very, very faintly yellow and they’re
typically very, very difficult to get a spectrum of
and we were still able to get good useful data out it,
then comparing those fibers side by side they look the
same, but we’re able to differentiate them.>>Other questions? Everybody is looking
for a break. Okay, we’re going to
break for now 20 minutes. We will convene– reconvene
at 4:30 for the next portion. [ Inaudible Discussion ]