I now use Ar for sputter coating and never tried air but I used to coat sample in a very cheap method. We had a W filament, we put a very thin Cu wire (0.25 mm) coat the substrate with sample thermal evaporation under DP vacuum. Believe me though we always got graded thickness on the sample but we never had any problem. Later we got a Qurem sputter coat and since then we are doing it in "right" way. But I think there is nothing harm in using air unless you face any problem.

Your argument seems logical to me.

I would say that it depends on which property you'd like to measure in SEM: if it is just a topography contrast that you're looking for, I wouldn't worry except if you seek nm resolution. If it is a chemical contrast, then your surfaces will probably be a bit covered with oxygen and nitrogen, which might possibly change the result, but I wouldn't expect an effect that is above the uncertainty of e g an EDX measurement.

However when sputtering heavy metals like gold I do know that if the sputtering gas consists in light atoms as compared to the target (which is the case here: N and O vs Au and Pd), ion implantation due to sputter gas atoms getting bounced back by the target towards the sample can occur.

Argon, having a relatively high atomic weight, provides a suitable source of ions for effective bombardment of the target material. I am sure that if air had the same characteristics no one would use so expensive inert gas as consumable.

Oxygen from air can sometimes oxidize the surface; when you work with ceramics or metals it is not a problem, but when you want o image a surface of nanoparticles i.e. silver it can modify the morfology of the sample. Using air for biology and organic samples can make a huge damage and the imaging may be poor.

This question sounds tricky :) Usually manufacturers would say that we have to supply instrument with pure Argon and we need turbo-molecular pump to obtain fine grain coating. Actually I newer obtained good quality coating of Au/Pd for observations at higher magnifications. In the range above x50k I prefer Osmium. So, can You put some images or link to Your paper where You got nice images at high mag with the coating obtained by simple rotary pump and air ?

I think that with air you have lack of control over the deposited film. I haven't heard anyone using air in the sputter tools: 1) you will get oxide-nitride film (or poor quality "dirty" film if you're sputtering gold) and 2) your sputtering chamber will get "dirty".

so I would avoid using air if possible.

You didn't mention the kind of samples you are coating. For plant seeds, pollen, leaves, etc. I found air to be a fair substitute for argon. However, the use of argon is supposed to lead to a higher deposition rate and lower pressure at which the plasma can be sustained, therefore a superior coating. Argon is also inexpensive, so why not use it anyway.

For nomal observation, Air is good.

We sprayed samples in Ar as firm advised. When Ar was over, we noticed that the target and samples often became blue. They were polluted with carbon. Once again we have moved on Ar, target and samples were re-gilded. . We did not notice the changes in the sculpture of the samples.

For generation imaging air is ok, otherwise Ar

Argon atoms are much heavier than N or O atoms and therefore when Ar ion strikes the surface of gold ring it has more kinetic energy and it can easily separate Au atom. therefore if you use air instead of Ar, in addition to probable unwanted reactions, you have to use higher voltage and you need more time to rich the same gold layer thickness. furthermore the gold ring of the instrument can become polluted and these pollution can become source of error in future sample coatings. In addition to these you have to notice that oxygene and nitrogene are in their molecular forms while Ar is in atomic form. oxygene molecule is a biradical and it can result in many unwanted reactions

it is a wise thinking on the part of the author. May be experimented further with different samples, higher/lower magnification, identify issues and problems with air/ar and may lead development of new methodology, may be better tools?

For stable materials, u can use air for coating. oxygen in air might cause surface oxidation for the sensitive materials. It's always better to make sure with EDX

When you use air for sputtering, you should consider the following things:

1. Oxidation. (since you are imaging an organic specimen, oxidation will be low. Hence you wont see much of a contrast difference. If your sample is metallic this wont be the case. You will clearly see the contrast difference)

2. Nitridation (probability to happen is very low because nitrogen is less reactive than oxgyen) but, nitrogen is the main sputter gas when you are using air. 78.09% of air is nitrogen. (You can go for pure nitrogen which is a lot more cheaper than argon for sputtering. )

3. Other than nitrogen and oxygen, air contains a lot of other stuff. SO IT WILL BE DIFFICULT FOR YOU TO FIND OUT THE EXACT CHEMICAL REACTION HAPPENING ON YOU SPECIMEN. Especially, a lot more carbon contamination will be introduced to your sample from air. Other contaminants are sodium, pottassium, etc.

4. Since the sputter gas (in this case air) contains a lot of different gases (O2, N2, CO2, H2, He, etc.) the sputtered film will be highly uneven with respect to the thickness as well as the particle size because the momentum transfer will vary from one type of molecule to the other. So, eventually, you wont be getting a proper image of your sample.

Hence, in my opinion, a pure gas (atleast 4N - 99.99% pure) of a single element (either argon or nitrogen) would be the best choice as a sputter gas for coating biological specimen before SEM imaging.

I would always prefer an ESEM (environmental SEM) for imaging biological samples because I dont have to do any sample preparation and hence the images will be of the sample alone.

For our work with LTSEM we use always Argon. Almost 20 years have LTSEM technique in our laboratory, we have hundreds of samples and always with a extraordinary result..

Thanks for all the answers on my question. As I see, there were no observations that air is not a suitable substitute for argon, except for observed by Svetlana Polevova bluish colors on specimens and target (but specimens were good anyway). I have observed the same bluish colors with both Ar and air deposition, but as I found out it was because vacuum problems. So, for now I strongly believe air is a good substitute for argon (and why I would bother with Ar if I do not really need it). I am working with a wide variety of specimens: sell cultures, soft tissue, hard tissue, crystals, polymers, ceramics, nanoparticles, and metals of course, but metals do not need coatings. Working for years with air instead of argon I have never observed any specimen damage/alteration due to presence of oxygen, water and carbon dioxide.

To Christian Monachon: What is a chemical contrast? And about ion implantation – it need higher voltage than one used in sputter coater; anyway, Ar is as good as other elements as a material for ion implantation.

To Gregor Filipič: cold plasma (in sputter coater) is very different from hot plasma used for plasma treatment.

To Igor Dzięcielewski: no, there is no damage to biological specimens.

To Tomasz Płociński: In my original question I indicated that I am discussing coatings with rotary pumping for magnifications not higher than x50-100k.

To Alexander Kozhanov: how I can contaminate my camera with air?

To Jacques van Rooy: I use the same pressure for Ar and air.

To Gholam-Reza Nejabat: I do not need higher voltage, a bit longer time makes all the difference. Target made of noble metals is not contaminated by air.

To K. R. Ganesh: When using air coating is as even as with Ar coating: high number of ions provides excellent statistical averaging. I strongly advise you not to use expensive high purity gas for sputter coater with rotary pump since oil back streaming will provide enough contamination anyway. Also I rarely use ESEM for biological specimens – high vacuum pictures are usually of superior quality. ESEM is for specific tasks, not for general observation.

Thanks to everybody for interesting discussion.

Actually, there are plasma cleaners for SEM specimens, which use oxygen as one of working gases (http://www.gatan.com/products/specimen_prep/products/950_Solarus.php) Removal of specimen material is negligible and it is with reversed (compared to sputter coating) polarity! Actually I use my sputter coaters for making carbon film on TEM grids hydrolytic. Anyway, it is tested – no negative influence on specimen topography or chemistry (detectable with EDS).

In our lab we are using air for sputter coating for more than 15 years now. We are characterizing polymers, nanomaterials, nanocomposites and biological samples regularly. We do have a separate Ar attached sputter coating unit which we seldom use. We do not find any remarkable change in resolution for samples coated in argon. Practically its far more easier to maintain and run sputter coating unit using air rather than the one using argon cylinder.

18th October 2013

Dear Vladimir,

I have extensive experience in using a scanning electron microscope to examine metaplastic changes and early signs of pre-neoplasia in the uterine cervix. However, my experience in specimen preparation is limited (mostly done by my co-researchers). Therefore I cannot answer your question with authority.

It would be interesting to hear more about your work.

Kind regards,

Johan (JT) Nel.

As opinion A.Bogdanov (engineer):

I think that there is no exact answer to the question "What is better for ion sputtering coating - argon or air" because the process of Ion coating of SEM specimen is really two different physical processes: A) ion sputtering of metal and B) sputtered metal deposition, going in parallel in the chamber of Ion Coater.

The physics of Ion sputtering of metals is well studied and described in details (for example in L.I.Maissel, R.Glang Handbook of Thin film Technology. It is different of coarse depending on sputtered metal and sputtering gas, but is well reproducible for each metal/gas pare if we can stabilize such parameters as gas pressure, ionizing voltage, anode-cathode distance and ion current.

The results of sputtered metal deposition to specimen surface are strongly dependable on:

1) chemical composition of surface, 2) surface temperature, which depends on metal particles energy, ion current, electron bombarding density on surface (different in simple diode type and magnetron type Ion coaters), 3) surface thermo conductivity, 4) gas emission from specimen surface and some other factors.

In case of biological specimens these factors are so widely different, that the common discussion is impossible. If You wish precisely study Ion coating processes, it is necessary to standardize

coated specimens, for example use graphite or mica. Dried biological specimens are porous, have low thermo conductivity and as a result often give gas emission due to surface temperature rise.

Our practical choice, based on 30-year using of EIKO IB-3 Ion Coater in MSU Biofac EM lab, is of coarse Argon. Some years ago we had a period of Au coating in air (had no money for argon), and the coated specimens were sometimes colored in blue.

A.Bogdanov

Sputter gas, Always use high purity argon gas of the grade known as "White spot" this will ensure fast sputter rate and good pump down time.

I strongly disagree. While Ar can give a bit faster sputtering, in most cases it does not matter. As for quality of coating it does not matter. Air is just as good for noble metals, no tank replacements required, and it's free.

From my experience, there are no any practical differences between sputter gases. I've seen one very old sputter (JFC 1100) which was used for 40 years with one gold target in air. The target was looking pure and coating was great - I used it for some "exotic" applications including AFM probes coating with TEM (up to x300k) visualization. No any differences in comparison with a brand new Q150T operating with pure N2.

Thus, sputter gas can be important only for some special applications like quantitative EDX. For conventional sample preparation in routine SEM, air (don't forget about the intake filter) looks Ok.

The simplest method to estimate the sputtering quality - slightly (20-30 sec) coat a TEM grid (pure carbon is better, but formvar is Ok too) then examine it by TEM. I didn't see any differences between air, pure N2 and argon. In addition, it is a good method to make TEM grids for astigmatism correction: if the film has small holes (water addition to formvar, heating, etc.), after such coating it becomes semi-transparent under the beam and the borders of the holes will have very good contrast.