Mwchalmers

Gold

Hey. I didnt link it to nanoparticle gold because it is not colloidal gold. It is a thin film produced in an evaporator. It is for thickness tests in electronics — Preceding unsigned comment added by Mwchalmers (talk • contribs) 00:56, 2 June 2012 (UTC)

Sorry. All the links I made WERE to colloidial gold, so they should be fine. Jasper Deng undid your edit with the ostensible reason that we had an article on nanoparticle gold. If your use is different, then put it back in, as we apparently have no subarticle on it. However, the links I made should still be good. (Remember to keep it short, though, since this article is short of space. If you have a lot of mateiral, consider starting a stub or writing an article, and link from the gold article, with one line of description there.) SBHarris 01:01, 2 June 2012 (UTC)

Hey.

I explained this to Jasper. Actually what you are looking at is NOT colloidal gold. It is a 30nm thin film produced in an evaporator in high vacuum. Gold nano-particles (colloidal gold), are produced chemically acutally.

Also, in the case of a thin film, you get the same transmission spectra regardless of thickness (unless it becomes so thick that the film is opaque e.g. full attenuation). The transmission spectrum is essentially the INVERSE of the reflection specturm you get when you see bulk gold. It is a driect way of observing the color properties of bulk gold.

Thats why I took out the edits to colloidal gold. It is misleading because what is shown is absolutely not collodial gold (I made the film myself). The thin film has none of the cool properties that colloidal gold has.

A new article idea!--Jasper Deng (talk) 01:59, 2 June 2012 (UTC)

Your recent edits

Hello. In case you didn't know, when you add content to talk pages and Wikipedia pages that have open discussion, you should sign your posts by typing four tildes ( ~~~~ ) at the end of your comment. You could also click on the signature button or located above the edit window. This will automatically insert a signature with your username or IP address and the time you posted the comment. This information is useful because other editors will be able to tell who said what, and when they said it. Thank you. --SineBot (talk) 02:16, 2 June 2012 (UTC)

Some advice

The spirit of science is that things are repeatable; independent scientists should be able to reproduce your result and posted some works that would constitute reliable sources. This is why Materialscientist (yes, he does live up to that username) is reverting your edits.--Jasper Deng (talk) 02:28, 2 June 2012 (UTC)

Gold

What was the base vacuum during deposition? How was deposition done (target, deposition method, deposition rate and time)? How did you deposit SiO2 capping layer? I can explain possible artifacts. Materialscientist (talk) 02:30, 2 June 2012 (UTC)

Repost from talk with materials

Ok .. the details.

0) Clean slide. Acetone IPA

1) Evaporate metal. Electron beam - 100mA current soak 150mA for deposition.

 - deposition rate 1 angstrom/sec for first 100 angstroms
 - deposition rate 4 angstroms/sec for 200 more angstroms
 - vacuum pressure < 1e-7 mbar (turbopump 1500 rpm)

2) SiO2 deposition

 - RF sputtering on Si target 30, 50, 100W cycles of about 5 min each 
 - rate is somewhat machine depentant (I use 30ccm Ar with 5ccm O2) at 7mTorr base pressure
 - with this I get about 50 angstroms/sec at the 100W
 - You can really lay down the oxide if you want. Up to 1um is no problem for this experiment

If you dont do the SiO2 the slide wont make it out of the lab with a clean gold surface left.

3) Picture

 - Put the slide on a clear wafer container 
 - Get a nice light source and use flash + white balance
 - Take a few shots untill one looks nice

/M

Thanks for details. Looks Ok, but the color is off - sorry, don't mean to be unfriendly, but I've never see such in my entire practice. It should be much much weaker and have much more blue and gray. The color is unbalanced, as obvious from the green corners and reddish center, but this might be too weak to explain the image. Sure, you can neglect my experience, but your experiment is basically WP:OR. Look for independent evidence if you believe you are right. (you can reply here - I see that via my watchlist, a wikimedia software feature) Materialscientist (talk) 02:56, 2 June 2012 (UTC)

Alright. Ill give you another image. Ill repost it on the page and maybe you will like it. I am not neglecting your experience. I think the best explanation for an image is the camera that took it. I dont know how much independent evidence you need. I just told you the full details of the experiment and you seem to agree that it is ok.

Physics Discussion

Wikipedia is not the place to publish your original research. Okay? Even if it's Nobel quality. Publish it somewhere else, THEN come here to put it in with a citation. [1]

Don't think we're picking on you. If you had a photograph of a robin in your back yard, nobody would care. But if you had a photo of a UFO that looked like a flying saucer, we'd be doing the same thing we're doing here. Extraordinary claims need extraordinary evidence. Publish someplace else. WP is not for hobbyists to trade photos. SBHarris 03:21, 2 June 2012 (UTC)

I dont understnad. I told you the process Sbharris. This is really NOT an original research project at all. It is a knitting together of very standard procedures which are used in all silicon processing labs to show something cool and interesting. If it was original, I would publish it in something real like IEEE, APL, PRL etc. Believe me - research science is what I do.

In fact what I am actually demonstrating is "Spectroscopy". I am not supposing anything super special here - I just made a really crappy spectroscope in a really "expensive" way to show something instructive and cool about gold. There is nothing really "extrodinary" about it at all. Its just demonstrating something interesting. Putting an old theory in a new light. It is not by any stretch an extrodinary claim.

The only thing I claim - is that gold has a transmission spectrum which you can demonstrate in the visible spectrum. Other metals do the same thing, but up in the UV. This is all well known. You can find all of it in Griffits: Introduction to Electrodynamics or any upper division undergrad physics book (if I site girffiths can i put my image up :-) ??? )

Mwchalmers (talk) 03:37, 2 June 2012 (UTC)

Citing sources can be technically tough; I can show you how to do so, but first you must name the source.--Jasper Deng (talk) 03:43, 2 June 2012 (UTC)

Mwchalmers, you could have hit some plasmon resonance vs. thickness, and "amplified" its green color by an artifact, especially by your illumination+camera+flash (color balance still looks odd). Gold-coated glass is a commercial product, look for examples on the web and on Talk:Gold#Green_gold_image_removed; this one [2] is familiar to my eye - it has some green-blue, but very different from yours, very appropriate for a thin film. The main problem is we have to present characteristic color, not an oddity unseen by others. Materialscientist (talk) 03:48, 2 June 2012 (UTC)

Materials: Possibly - I really think it is a color balance issue (though there may be plasmon resonance ?) Im not into plasmon science specifically (I am into microwave electronics ... different frequency). But the motivation behind the experiment was a plot of the skin depth versus frequency for gold. This is about 20nm for ~600nm and about 50nm for ~300nm. I designed the film to have greater than the skin depth for 600mn and less than the skin depth for 300nm. So reds are reflected and greens are transmitted etc. This is why I get the spectrum I get (I have reason to believe). Of course the PHYSICS of the skin depth in Au is the plasmon resonance as you suggest. If you have say a 40nm film you will pass more blue but you will reflect more green because of the greater skin depth (e.g. as the Tip Enhancement guys have published on)

Check out the plot on page 290 of this book: Tip Enhancement: Satoshi Kawata,Vladimir M. Shalaev. It is available in google books.

Here is the link to the page with the plot: http://books.google.se/books?id=IzS6x19JEcQC&pg=PA289&lpg=PA289&dq=gold+visible+skin+depth&source=bl&ots=J7DZn0fpoM&sig=8wMp1xZSp4-MJF1AzInupl8kqH4&hl=sv&sa=X&ei=EJDJT42xKq-B4AS9kYQK&ved=0CFcQ6AEwAg#v=onepage&q=gold%20visible%20skin%20depth&f=false

I know that this is also treated in the undergrad texts (Jackson, Griffith, Wangsness) but not in as much clear detail as here. I suppose that this book deals mostly with the enhancement of plasmonic tips for nanoscale imaging at optical frequencies.

We can't explain green color by bulk gold properties (gold has flat absorption in the green-blue, i.e. should be blue, or green-blue). The fundamental absorption/reflection spectra might change with thickness, but here we should watch for numerous artifacts, especially chemistry and continuity - the plasmon frequency depends on the particle shape, i.e. the situation may be different for a continuous film, or a film made of nanoislands of different shapes. Another unavoidable factor is angular dependence of interference between gold and other layers in the sample. Materialscientist (talk) 04:11, 2 June 2012 (UTC)

Watch out with p. 290: its Fig. 9 shows red gold (see Im(e)), which is typical for gold nanoparticles (see colloidal gold) rather than films. Many authors actually deposit films composed of nanoparticles. Materialscientist (talk) 04:15, 2 June 2012 (UTC)

Well, I do think that chemically, the film should be very flat and very pure because it was evaporated. Also the glass has a random atomic structure, so I sould assume that this would prevent nucleation sites for the growth of nanocrystals. Because the gold is a gas and it is rapidly cooled when hitting the SiO2, it should not order into a crystalline structure (I suppose). The growth of nano-particles is usually very temperature dependent. Here I take Gold vapor at 2000C and slam it into a substrate at 20C - crystallization seems very unlikely to me.

I miss referenced here - what I want you to see is p.289 - the skin depth varies as a function of frequency (e.g. the evenescant wave penetrates further into the gold for blue than for green. Over 100nm of film none of this matters - absorbtion is flat. But at 30 nm the evanescent wave will extend beyond the film and become a propagating wave on the other side. My experiment references fig.8. on page 289 actually.

Also I notice the dates here on the plots, being 1950 etc. I think that this was well before the conjecture of plasmon quantization. I suppose that there is nothing too fancy going on other than the fact that I have engineered the film to have a specific thickness. This thickness just happens to be multiple skin depths for some optical frequencies, and a fraction of a skin depth at other optical frequencies. Hence the gold film is reddish on reflection and bluish on transmission. This is i suppose my explanation on what is happening with the film. In sunlight, the film is reddish when you look at light reflected from the slide and blue when you look through the slide.

Also, I recall that even in bulk material you have plasmon modes. In a nano-crystal, they are actually resonant e.g. like cavity modes (from what I recall - only had one course on this). It is these different plasmon resonances which give rise to the myriad interesting optical effects that you see in nano-crystals. In films, you have just the random interference of plasmon waves (e.g. they are all out of phase because you add a degree of freedom (... the infiniteness of the glass slide) and because the light which excites them is unpolarized sunlight or camera flash)

There are two effects: geometry and energy levels. Geometry appears in two ways: (i) there is a bulk plasmon, which can be simplified as a multielectron excitation of nd electrons to (n+1)s,p levels. This gives yellow color to gold. The corresponding optical threshold is not sharp, and thus actual absorption shifts with thickness (seen as color change from orange to blue here [3], I believe they've deposited a wedge). (ii) Resonator-like modulations, i.e. boundary effects appearing when the infinite bulk is reduced in size.

Energy levels: they shift at the nanoscale, affecting the fundamental plasmon frequency.

There is also the surface plasmon, which is sort of both together, energy and geometry effects. It is always present in bulk samples (they have surfaces), but is weak, and is only seen when selectively excited (by energy, angle, some AFM tip, etc.), or when the material is thinned so that the bulk contribution is integrally small.

Geometry shifts are always to the low energies, see File:Gold255.jpg, and are different for particles and films. Blue (or blue-green) gold film is simply thin yellow gold. To get green gold, you need some mechanism to increase the blue absorption. This could be organics, non-stoichiometric SiOx from the overlayer, or some specific (e.g. plasmonic) resonance due to particular interfaces, film nanostructure, etc. Materialscientist (talk) 05:36, 2 June 2012 (UTC)

I want to hash this physics discussion out in a different light if I may.

One thing that is striking to me about this is that what we are observing in the Au film image is electromagetic radiation (rather than surface plasmons). The surface plasmonic interaction though is responsible for the electromagnetic radiation we see (e.g. the reflection and transmission spectra). Classically, we describe this phenomenon by invoking an skin depth - a result of solving maxwell's equations for a boundary between vacuum and a material with finite conductivity.

However, if you use the conductivity of gold - for example, and calculate the skin depth of a 500THz (optical) electromagnetic wave, it is (from what I remember) less than a nanometer. The OBSERVED skin depth is different by a factor of 40 in this case, and the accepted explanation of this is surface plasmons as Materials has described. So the deviation from the classically calculated skin depth is due to a quantum phenomenon. At microwave frequencies (for example), one is still in the classical regime, and classical electrodynamics still gives the correct skin depth. In other words: the long skin depth of optical frequencies in gold cannot be explained classically.

Notwithstanding, we can still (certainly) define a transmission and a reflection spectrum for the 30 and 40nm slides. This transmission spectrum demonstrates something very fundamental abut gold (hence why I posted it). In this case the physics is very interesting and not so simple. The same phenomena occurs in ALL metals, but the transition is not in the optical.

Also, I did examine the slide before and after the SiO2 cap layer. There was a little change in the spectrum that I saw, but it was very slight. One would need a very nice spectrometer to really measure it as my eye was not sensitive enough to tell the difference. Note that the SiO2 in this case is a dielectric and acts essentially in the same manner as "air with a different dielectric constant". It would be interesting to see if a CVD oxide made a difference. I do not think that the SiO2 is responsible for the color we see - though certianly, it might have a slight effect. Without the SiO2 layer, the Au film is still attached to highly non-stoichiometric SiOx (glass). Still we see the effect.

Also materials: I deposited both 1nm and 5nm today as well via the same method. The slides were fully transparent which agrees with all of this. Recall that plot on p289 of Tip enhancement. There is no optical frequency at which the skin depth is less than 20nm - so the 1 and 5nm slides appear transparent (all of this also demonstrates nicely the finite conductivity of gold - for if gold were infinitely conductive, all the slides would be opaque and the skin depth would be 0).

Lastly, if you look at the slide in sunlight, it looks very similar to those that the company you directed me to makes. They are doing 50nm films which is close to what I have (30nm). The camera artifically supresses the reflection spectrum and highlights the transmission spectrum due to its sensitivity.

As always it is hard to get red to show up on a digital camera correctly. Red (on cheap cameras like mine) is always oversaturated or undersaturated. Thus this explains the greenness of the slide. It is a real green. In sunlight the slide appears more blue (because you have a more flat spectal input than the camera flash).

(Also, yes, my background is in experimental physics)

Mwchalmers (talk) 18:49, 2 June 2012 (UTC)

(Same here). As expected, the green color you see is altered by the camera (and illumination conditions, I guess - looks dark overall), and the actual color is green-blue, which is normal. Thus the capping layer hardly matters. Skin depth is just an exponent - there is no sharp threshold for whatever it describes. It does depend on the frequency, which is why it may be off by a large factor if calculated from the static conductivity. Microwave conductivity can be measured, but the measurement is different from that of electromagnetic transmittance, thus discrepancies are well possible. Optical transmittance measurements are easy and can be very accurate, thus I prefer using them for estimating skin depth. I would just measure transmittance spectra for the slide, slide+capping layer, and slide+gold+capping layer - you've got large homogeneous samples; any commercial spectrometer (200-800 nm) will do. Materialscientist (talk) 01:57, 3 June 2012 (UTC)

Tornado Article

Hi, I just wanted to say thanks for adding an image to the photo gallery that I created in the Tornado article! Spindocter123 (talk) 01:49, 20 May 2013 (UTC)