00:12
It's Asami here and today we're gonna be talking about Part 2 of the 論文紹介 episode
that I started. So if you haven't listened to Part 1, go back to one episode before
where I give you some background information, historical context about the questions and
the problems that we want to look at in this paper. And today we're just gonna dive right
into the paper's detail and what the researchers did. So once again, this paper is entitled
Time-Dependent ATR-FTIR Spectroscopic Studies on Fatty Acid Diffusion and Formation of Metal Soaps
in Oil Paint Model System. And this is a team of collaboration between University of Amsterdam
and Rijksmuseum. They're both in Amsterdam and they are interested in the problems associated
with oil paint. So just to quick recap, we have talked about in a previous episode how the metal
soap formation that disturbs the surface of the artwork and creating this foggy appearance,
creating a bunch of these crater-like stuff on the surface of the artwork and how it interferes
with the visual experience of the viewer and how the conservation community as a whole got
together to figure out what they are and how to deal with the problems. And we also talked about
the different types of fatty acids that are present in a drying oil paint mixture, which are
unsaturated fatty acids, so those with double bonds, and saturated fatty acids, those without
the double bonds, and inorganic pigments with metal center. So those are three key components
that are in the oil paint mixture and the question is what happens during the drying process and how
does the metal soap get formed during this process. So we also learned that during the drying
process the metal ions migrate into the polymerized oil network of the binding medium
and they form this thing called ionomers. So it's ion plus polymers, right, ionomers.
And these ionomers are reactive towards saturated fatty acids, so the fatty acids
with just the single bonds, and they are also present in the mixture of the oil paint as I just
talked about, and this is going to cause the metal soap formation. They know this much.
03:04
And they have looked at the model system under four different conditions, which I'm gonna
go over in a bit, and the team has looked into the diffusion process of saturated fatty acids,
so that's what they're tracking, that's their observable, through the technique called ATR-FTIR.
And just to give you a spoiler already, they found out that solvent flow, presence of water,
and presence of pigments, all of it enhanced the metal soap crystallization in the model system.
So just taking a step back just once, I wanted to share quickly about what the technique
is that they're using to study this. So it's called ATR-FTIR, it stands for
attenuated total reflectance Fourier transform infrared spectroscopy. So if you have done
organic chemistry, maybe in high school or college, you probably have encountered IR
spectroscopy, FTIR spectroscopy, just very similar version of that, that's how people
mostly use IR spectroscopy nowadays, just a better signal processing method, FT, the Fourier
transform part. What makes this special though is the ATR part, the attenuated total reflectance
part. So what does it do? Essentially, what ATR enables you to do is that it enables direct
sampling of solid or liquid sample without extra preparation. So if it was a normal FTIR,
like a transmission FTIR, then you're going to need some kind of solvent to usually dissolve
your substance, and make sure that those solvents are spectroscopically transparent.
i.e. what you're interested in, what you are expecting to see, has different peak from what
the solvent is that it's dissolved in, so that you can clearly see that you have what you have
in your sample. You don't have to do this in attenuated total reflectance FTIR, because
you don't have to prepare the sample in any way. And also in the transmission sampling,
the path length, so how much light travels through the sample, is the direct measurement of how good
your signal is. Whereas in ATR, it is an internal reflection-based method where an evanescence
waveform essentially bounces off from the ATR crystal into the sample, and they get the reflection
of the wave as they interact with the sample, and that's the signal you get. So the path length
doesn't really matter, and this makes it really good for looking at things that are optically
06:04
dense or opaque sample, which as you can imagine, an oil paint can be opaque, so this is a good
technique to look at oil paint model. And what you, the information you get in ATR FTIR is still
the same as the regular FTIR. So on your x-axis, you will get a wave number, so that's how many
wave cycles happen per unit time. In other words, you can say that it's the sort of vibrational modes
of the specific chemical bonds that are present in the mixture. And you can see the different
position on the x-axis tells you the different species of chemical bond, you know, whether it's
a C-O stretch or C-C stretch or C-H stretch, you know, what sort of things are there. You can see
those characteristic peaks depending on the position on the x-axis. And y-axis is absorbance,
and in this case, it's essentially the same as sort of signal intensity. So you can see how much
of particular vibration modes are happening in the sample volume, and that should tell you
quantitatively how much of your, you know, of whatever you're looking at, let's say how much
of the metal soap is in the mixture based on the characteristic peak that you know where to expect.
And to top it off, they're doing this in a time-dependent manner. So whenever you see a
paper that says time-dependent, that just means they did this over some extended period. So they
may have taken the measurements every 10 minutes, 20 minutes, 30 minutes, who knows. They did it
all the way up to 1200 minutes, so that's quite a long experiment. But then again, you know,
these oil paints are very old, you know, some of them are centuries old, so I guess it makes sense
that some of these are expected to take a long time. So they're looking at the changes in signal
over time, and that's also going to give you a new information that you don't get if you just
keep taking the snapshots at random moments, because then you can see a growth in intensity
of some peak, or maybe decrease in intensity of some other peak, and you can ascribe that
to different activities that are happening in the oil mixture. Okay, so I mentioned that they're
looking at four conditions. The four conditions that they're looking at are one, they have the
metal polymer, acetone, the solvent, and the saturated free acids in the mixture. So that's
condition one. Condition two is the same set of items, just with less water. So they have dried
09:07
out, they have sucked out all the water from the environment to look at the effect of humidity
for the oil paint drying process. And number three is metal polymer, acetone, solvent,
but then they let the system equilibriate. So they let them sit in the solvent with just the
metal polymer, and then they add the saturated free acid and see how they diffuse throughout
the ionomer network. Lastly, they looked at metal polymer, acetone, saturated free acid,
and then pigments. So these are sort of additional source of metal ions, you can say.
So they're not just metal polymers that are already embedded in the polymer, but they are
looking at addition of pigments. And I think the last condition is sort of the most realistic
of what an oil paint is in the real life. So they looked at all four conditions,
they looked at, they took the ATR-FTIR spectrum for 1200 minutes and plot the curve,
the time series curve of what is the signal under particular peak and how they developed over time.
And here are things that they found out. So first off, they found out that the presence of free
saturated acids and metal ionomers lead to rapid metal soap crystallization reaction.
So that's, they already knew this was going to happen, but it's a good confirmation that that
does happen. And they found out that removing water as much as possible, so that was condition
two. And when you do that, when you remove the water from the system as much as possible,
they were able to slow down the metal soap formation. So they found out that less water,
the better. At least you can slow down this metal soap formation. So that was number two.
Finding number three. So in the condition three, where they let the metal polymer system
and the acetone equilibrate, so they just let it sit in there, and then you add the
saturated free acid after some time. They found out that by doing so, also slows down
the diffusion of saturated fatty acid. I guess because there are solvents already present and
that sort of interferes or competes with the diffusion process of the saturated acid.
And saturated fatty acid. Whereas when they were introduced together, so when you add the solvent
and the saturated acid together, like you did in a condition one, you see a lot quicker
12:03
metal soap formation from there. Okay. So number four, that was the last condition. So that was
the most realistic one where you have the metal polymer system, the solvent, the saturated free
acid, and also the pigments, the extra source of metal ions. And they found out that having
pigments, surprise, surprise, increased the rate of metal soap formation. And this is particularly
not surprising because they use zinc oxide and lead oxide as the source of pigments,
which are sort of two most common ones that they've already observed empirically. So they
know that these metal centers, zinc and lead, tend to be pretty reactive in this oil paint system.
So to conclude, they found out that the solvent flow, which is how much acetone is there or when,
you know, how much acetone is present in the system and the presence of water. So how much
water is in the system or around the system and the presence of pigments. So do you have extra
source of metal ions? They all enhance the metal soap crystallization in the model system.
So that was their finding, pretty straightforward to understand this paper. But I think it's a good
carefully done study where they had a very specific question and they were able to get
a quantitative answer out of this, which is not always easy, especially in conservation science
field to get a good quantitative data. Because if you have a sample from real artwork, you cannot
really replicate that situation over and over again to get good statistics. Whereas because
they are using model system here, you know, they created a recipe of what an oil paint
system would be like. They were able to get good statistics and study and get the time
constants, et cetera, from these experimental, highly controlled experimental conditions.
Now the criticism is always that it's a model system and it's not always reflective of the
real situation. That's completely true. But I think it's still a very good factors that they
have identified. And these are factors that are immediately actionable, like the conservators can
do things to remove water, for instance, make sure that the oil paints are not sitting in a humid
environment. These are things that they can immediately do to improve the condition of the
artwork or at least prevent the artwork from going worse than what it is now. So that's it for this
論文紹介. These episodes are not easy. So I would appreciate if you give me some feedback.
15:04
Anything is helpful. You know, whether it's Asami, this is too much information.
Can you make it more fun? Or can you pick something more interesting to talk about?
Or, you know, or maybe that you just genuinely enjoyed. I would appreciate any feedback.
Thank you for listening. And if you haven't already, make sure you listen to the first
episode. I think this episode will make a lot more sense after listening to part one of this episode.
All right. Bye. That's it for the show today. Thanks for listening and find us on X at
Eigo de Science. That is E-I-G-O-D-E-S-C-I-E-N-C-E. See you next time.
Transcribed by https://otter.ai
