The home stretch

A lot has happened since my last post. Julien and I went to the C2RMF again last Thursday. After some unsuccessful attempts to get the microtopography apparatus to function, we performed optical microscopy and SEM on KETH kurai, the Roman coin, and AN43U. Bad news first: we did absolutely nothing to the Japanese coin. I suppose the fluence we were using (~150 mJ/cm^2) wasn't quite enough to break through the corrosion layer. The ablation marks on the Roman coin were visible with naked eye.
The correct description for this coin is:
a.) Toasted
b.) Well done
c.) It resembles R2D2 after the battle of the first Death Star
d.) All of the above

If you picked d., congratulations! Yeah, so we hit this coin with way too much energy. From right to left (i.e. most to least visible) the craters are from 1000, 500, 100, 50, 10, 5, and (barely visible) 1-2 pulses. I say 1-2 because the shutter isn't good enough to reliably pick off one shot. The second smaller spot is probably due to a ghost reflection off of the dielectric mirror, but evidently it carried enough energy to ablate the coin. Unfortunately, the day I shot at the Roman coin the energy meter was missing, and the following day Rod and Xiaowei tinkered with the laser, so I don't know the total energy in the pulse train. Using the equation that avg energy = average power/ rep rate, and using the spot size measured with the CCD camera, I estimate that the large spot is approximately 11.5 J/cm^2 while the small spot is 2.5 J/cm^2. That would explain the char marks.

I was pleasantly surprised with the nickel. Nothing was visible, even under the optical microscope at 20x visibility, but the SEM showed otherwise.
The lines are, from right to left: 500, 100, 50 (higher rate), 10, and 5 pulses incident. The ablation spots are really quite nice, and the EDX results show a successful removal of corrosion products.

On Friday, we had a group meeting with Dr. Mourou. I showed him the SEM results and he seemed happy with them. I really wish I had more time to do research now that the setup is functioning and none of the pieces are missing.

I spent Friday and today making my presentation for Wednesday. The students are meeting with Steve Yalisove tomorrow to discuss the projects/summer overall, and everybody is giving presentations on their work Wednesday.

I've ablated something! ...I think

As predicted, Bianca and I spent pretty much all day in the lab Tuesday. After realigning some misaligned components, and relocating the CCD camera--of all the components in our system that one is particularly problematic as it tends to wander off on its own a lot--we got to shooting. We used KETH kurai and went around the rim, varying the number of bursts. We had the shutter set for the lowest (quickest) setting, so we were getting 1-2 pulses per burst. No damage was visible to the naked eye.

Wednesday was Bastille day. I slept in, planned for my trip to Ireland, and went to the Eiffel tower to see the fireworks. The fireworks, although pretty cool, were not overly impressive. I wound up walking back to Cité U from there because there were so many people and the metro was packed. Many of the buses were not running, probably because of all the people in the streets.

I came in on Thursday and experimented with the shutter. After tracking down the manual online, I figured out how to change the duration of the open and close cycle. I wanted to see how precisely the shutter's timer was, but something electrical was wrong with the laser and I didn't get the chance. I spend the afternoon making a presentation for Bianca of the SEM results.

My flight for Dublin left at 0930 Friday morning, which necessitated my waking up at 0500 to be at Port Maillot by 0615. I spent a few hours in Dublin before taking a train to Waterford. A friend of mine lives in Tramore, a small seaside city on the southern coast, where I spent the weekend. The trip was a lot of fun. I got back to Cité around 0130 on Monday morning because my flight back had been delayed. I was worried because the metro shuts down at 0030 and we did not get to Port Maillot until then, but I happened to meet 3 French girls who happened to live near Cité and we shared a cab.

On Monday I spent almost the entire day in the lab. I scanned across the surface of KETH Kurai with 1 mm spacing, hitting the coin with a different number of pulses/burst at different heights. I used bursts of 1-2, 5, 10, 50, and 100 pulses. We measured the spot size with a CCD camera and found the best focus to be ~70 microns in diameter. Unfortunately, the energy meter walked off, so I couldn't get a good measure of pulse energy, but if the numbers from Tuesday were still valid, it was on the order of 5 micro-Joules, giving a fluence of ~130 mJ/cm^2. This should, one would imagine, be enough for results to be visible after 100 pulses, but this was not the case, so I'm a little worried about whether the energy is significantly less. I performed the same experiment except with 1-2, 5, 10, 50, 100, and 500 pulse bursts on AN43U and again saw nothing visibly different about the coin.

I will try more pulses today. We're already using the maximum energy from the beam Rod's giving us, so if that doesn't work, the only other alternative is to use a tighter focal spot.

It's a brand new day... sort of

This weekend was a good time to write post cards. It was miserably hot on Saturday and only a little better Sunday. weather.com kept promising rain, but it didn't come until this morning (and it felt fantastic when it did). The most productive thing I did this weekend was to go to Port Maillot because I have to go there Friday morning to catch a bus to Beauvais for my flight to Dublin and I wanted to see it once before I go half-awake at 0600 on Friday.

Julien took Johanna and I to the C2RMF at the Louvre today to have a tour and do SEM on 4 of the coins: AN43U, KETH kurai, ASD46 and the Roman coin. The Roman coin didn't do so well, or at least, we had a hard time seeing the coin's features with the SEM. The dime seemed to be in particularly good shape, and silver completely dominated the composition readings (as it should, it's supposed to be 90% copper) with some carbon, oxygen, and aluminum (Julien suggested contamination, or rubbing). Bianca thought we might have time to do the front and back of all 4 coins, but we only had time to do the fronts. I'll discuss the results in more detail tomorrow with Bianca. We have the whole day in the lab tomorrow, so I'm really hoping I get to ablate something before Xiaowei cuts off the laser at the end of tomorrow.

In the mean time, here's a close up of the "w" from the "in God we trust" on the dime from the SEM.
We took a more zoomed in image of the same spot; I just happen to like this one more. I'll probably post more tomorrow. We're going back to the C2RMF next week to look at some ablation craters. They WILL be there by then.

Tues 6 July - Thurs 8 July

We've been tiding things up in the lab for the last few days, tinkering with the system. Things seem to keep moving (vertically) and our HeNe is especially unstable, so we now have a series of pinholes in different places in the system. The lens we found that we thought was 30 cm focus was actually either an extraordinarily tight focus or a negative focus lens, so we had an adventure finding a suitable lens; we wound up going to another lab group and borrowing one of theirs. An optics lab without lenses... Anyway, once that got taken care of, we found that the CCD camera we were going to use for beam profiling disappeared. Antonen took it because he needs it for his experiments. He's going to be shooting today and tomorrow, and because of the high energy X-rays involved in his work nobody else can be in the lab when he's shooting. In other words, I can't do lab work until Monday.

On the plus side, Julien is taking Johanna and me to the C2RMF at the Louvre on Monday to do SEM and EDX on the coins I'm going to ablate. We're going to (hopefully, if time allows) do SEM of both sides of 4 coins: one nickel, the roman coin, one dime, and one KETH coin as well as element analysis. In preparation for that, Bianca helped me pick another nickel to use.


We picked this one because it exhibits the best corrosion product uniformity over the face of the coin. For that reason, I'll call it AN43U for uniform.

I've read two papers recently, both of which say something of the same thing. I'll summarize the general concepts.

Ultra-short Laser Ablation of Metals and Semiconductors: Evidence of Ultra-fast Coulomb Explosion
H. Dachraoui et al, Appl. Phys. A 83, 2006

Fast Electronic and Thermal Processes in Femtosecond Laser Ablation of Au
H. Dachraoui and Wolfgang Husinsky, Appl. Phys. Letters 89, 2006

"The natures of processes which govern the interaction of ultrashort laser radiation with materials can be a complex interplay between ultrafast processes and processes on timescales which one traditionally calls the thermal regime" ~ from the 2nd listed paper

The papers discuss Coulomb explosion as a means for ablation. Despite having such a spiffy name, the concept is quite simple. When atoms in a material are exposed to light, the electrons in said atoms are excited. If the light comes in the form of an intense pulse, the electrons may become photoexcited to very high energy levels. Depending on the duration of the pulse, one of two things may happen:
~If the pulse is longer or on the order of the electron-phonon thermal relaxation time, the electrons will transfer energy (heat) to the phonons via collision and will then fall back into their original state, or a similar state
~If the pulse is much shorter (Dachraoui states <100fs), the electrons will be knocked into the continuum band and will escape from their parent atoms. The atoms are thus ionized, and the coulomb repulsion between ions results in ejection of the ions from the lattice. This is referred to as "Coulomb Explosion". Interestingly, this can result in not only the expulsion of ions (electrons and ionized parent atoms) but also neutral particles. The neutral particles are most likely the result of electron capture between ionized atoms and electrons with similar enough velocities. The author used time of flight measurements assuming that time of flight was only dependent on the kinetic energy of the ejected particles.A typical measurement of theirs is found below. This particular measurement is for Iron.

The distribution is bimodal for both 25fs pulse energies, although the graph is not so fantastic. For the 150 mJ/cm^2 there are peaks around 1 and 3 microseconds, corresponding to energies of 3 and .2 eV, respectively. For the 400 fs pulse, only the peak around 3 microseconds is evident. This behavior suggests Coulomb Explosion as a mechanism for generating the high energy electrons and thermal effects for the lower energy electrons, as the high energy electrons are not present after the 400fs pulse.

This weekend was a little light in comparison to some of the previous ones. It was still obnoxiously hot on Saturday, so I wasn't motivated to do much. Sean and I walked around Rue d'Alesia doing some (unsuccessful) souvenir shopping, and then stopped at the creperie on the corner of Blvd Jourdan by the hotel to eat dinner and watch Germany destroy beat Argentina. I went to place d'Italie by myself afterwards to try and find stuff for people at home. Apparently this is massive sale week/month, as a lot of stores have stuff up to 50% off. I wound up finding a gift for a friend and buying an athletic shirt for myself, but I still have no clue what to get for my parents.

Sunday, Sean and I went to Fontainbleu with some of the other Cité residents. It was kind of a last minute invitation, so only the two of us went from the REU group. It was a 40 minute ride on the RER D followed by a 15 minute bus ride to get there, but admission was free since it was the first sunday of the month. The place is very pretty, but I'm not sure it measures up to Versailles. It's not quite as opulent, but some of the touches (stone walls painted to look marble) were cool.

Also, in a move demonstrating vast intelligence and foresight, the Cité people gave us card access to the buildings for 1 month, despite our being residents for two. We got this straightened out on Thursday night or so, but it also happened that they gave us wifi for one month instead of two as well, which we didn't notice until Friday night after "business hours". Luckily, one of the other residents let me copy here wifi code and we all have access now (and will try to get ours back tonight), but the oversight reeks of stupidity.

At work today, Bianca told me that the thunderstorm that forcibly retired out laboratory pursuits Friday also knocked out the climate control in the lab, which renders the laser kaput until the control is fixed. We're hoping to be back online by Wednesday, but it looks like my goal of ablating something before the date Michael did last summer (7/9) is not going to be met.

So instead of doing laboratory work, I read 3 articles today.

Short-Pulse Laser Heating on Metals

TQ Qiu and CL Tien, Int. J. Heat Mass Transfer., Vol 35, No 3, 1992

The authors discuss a two-step process for the laser heating of metals, namely:

(1) The incident laser pulse excites free electrons to high energies and

(2) said electrons collide with the phonons, and the phonons experience heat transfer through collisions

In the limit of long pulses (longer than the electron-phonon thermal relaxation time), the model treats the light as heating up the metal directly.

I have mentioned this scheme before. This is the second time I've tried to read this paper; the authors hand-wave a little in places, which is unfortunate, but I believe I've gotten the gist out of this article.

The author starts with two heat transfer equations:

where the first equation corresponds to the electron gas and the second to the lattice phonons (thus the 'e' and 'l' subscripts). The 'G' term represents the electron-phonon coupling, while the Q term is a source term designed to represent the laser. The authors make several simplifying assumptions:

~G depends weakly on Te (meaning that Te is neither terribly small nor terribly large)

~the number of free electrons is 1/atom

~The source term is of modest intensity (.1-100 MW/cm^2)

~The laser pulse duration is longer than the electron-phonon collision time (.02ps)

~The optical properties of the material depend weakly on temperature.

The authors then proceed to solve the differential equations numerically after switching to dimensionless equations, and find that their simulations (surprisingly) agree with experimental data. 

Simulation Study and Experiment on Laser-Ablation Surface Cleaning
Zhou et al, Optics & Laser Technology Vol 33 Issue 3 April 2001

This paper wasn't so useful from an understanding point of view but as a reference point of view. The authors basically amalgamate the results of a few different papers into their own model without much explanation, but they do provide the references to papers which hopefully will go into the derivation in more detail. I won't summarize the paper because I didn't get very much out of it conceptually, and the context in which laser cleaning is utilized is quite different (nuclear decontamination) but I intend to follow up on the sources cited.

Ultrafast Dynamics of Femtosecond Laser-Induced Periodic Surface Pattern Formation on Metals
Jincheng Wang and Chunlei Guo (yeah, U of R!), Applied Physics Letters 87 (2005)

The authors revisit the formation of periodic structures on metals after ablation (discussed by Fauchet among others in several papers in the early 80s. I had started reading one of them a few weeks ago and found that it drew heavily on a paper pretty far removed from my project. I later started reading the latter paper in my free time, but I'm not finished yet). The general theory is this: incident pulsed light strikes surface defects or nonuniformities on the material to be ablated. The light is then scattered along the surface, where it interferes with the incident light. Periodic structure formation is evident on the material from this interference.

The authors investigate structure formation on three different metals: Cu, Ag, and Au, in order of decreasing threshold fluence. By the old theory, one would expect to see stronger surface formation on the gold material since the abation threshold is lower and structure would form more easily. However, what the authors found was that Au had the least defined periodic structure formation and that copper had the strongest. The authors then conclude that there are two competing processes involved in the formation of the periodic structures, or more precisely, electrons lose energy by two processes: electron-lattice coupling, the efficiency of which is described by a coupling constant, g, and diffusion to a lower temperature region. Structures are formed via the first mechanism. Thus, materials with higher g values would be more likely to exhibit periodic structures. Copper's g constant is more than twice that of silver and almost five times that of gold, (10 vs 3.6 and 2.1 x10^16 respectively), which supports the authors' theory. The authors do not, however, go on to predict more explicitly the relationship between g and structure formation.

My next move is to read some of the sources for the second paper. Perhaps I should also set traps for small animals to sacrifice to the laser gods so that the system can become operational again.

As a total aside (now that the US is out of the world cup after playing hideously against Ghana, I seem to have less of these) I've been watching a bunch of badminton videos on youtube at Cité. I want to be able to do this when I grow up...

End week 5

 This week was very lab-filled, but looking back, it feels like we accomplished very little. This is probably untrue, but the setup still isn't experiment ready. We couldn't get lenses of the focal length we wanted, so we have to settle for a slightly longer focal length. Also a bit of bad news; the timescale on the oscilloscope when we determined that a single pulse was coming through was too short. There are multiple pulses coming through per shot, and the number appears to vary between 1 and 3 almost randomly. Lei wants to use an energy meter to count the number of pulses, which is a good idea, but given low inbound energy, I'm worried that the meter might not be sensitive enough. Bianca and I found that we were getting ~4.5 microjoules into the system. The beam size is maybe 1 cm in diameter. We were about to measure that more precisely when the power went out, and given that the laser takes a while to reboot, we'll finish that on Monday.

The weather here makes me feel like I'm melting. I'm having a hard time thinking about work and not the chocolate/pistachio ice cream in my freezer.