I know, this is plunging in deep, but semantics is something that has come to bug me and haunt me. Presenting ideas to international collaborators from academia and industry - orally or in writing - I want to be understood correctly. When we exchange thoughts and juggle concepts with like-minded researchers (often on the other side of the world) there is a lot of potential for greatness. Isn't it a great feeling to go to bed pondering about a problem and to wake up to find the answer in your inbox, neatly written out by your friends in Japan, Australia or America?We all speak English, we are wrecking our brains looking for answers to the same questions, we worry about the same problems. It keeps amazing me!
But then, who of us has not seen two brilliant professors arguing across a lecture room, not listening to each other, and mixing everything up. Hopefully, they will realise quickly that they mean the same thing! They are just calling it something different.
In the stress corrosion cracking community there are a number of terms/phrases/expressions which I find keep causing confusion. I may have to add to the list later, but here is the ones that I feel keep coming up.

Stress Corrosion Cracking (SCC) or Hydrogen Embrittlement (HE)
It has been suggested that both expressions refer to the same phenomenon. Both can refer to a mode of failure, that would not occur without the coaction of mechanical forces and changes in the local chemistry. However, HE can definitely occur without SCC - in hydrogen charged samples with no corrosion to be seen. It is more difficult the other way around. H gas tends to be released in corrosion/oxidation of metals in aqueous environments
M(2+) + 2 H2O ⇌  M(OH)2 + 2 H(+)
M(3+) + 3 H2O ⇌ M(OH)3 + 3 H(+).
This means that there will be H gas is almost always present in corroding areas and there is a possibility that HE is responsible for cracking.
There are many other good theories competing with the one of HE though. The corrosion products themselves form a brittle phase prone to cracking. The adhesion between metals and oxides is another weak point.
It was even suggested that corrosion causes vacancy injection and local enhancement of the metal plasticity, thus facilitating slip leading to exposure of fresh metal and crack growth.
The two expressions, SCC and HE, are definitely not synonyms for one and the same mechanism and it remains to be determined if SCC is aided by HE.

Ni-enrichment or Cr(Fe)-depletion?
I think I may be changing my mind on this. :)
In an Fe-Cr-Ni alloy surely both expressions mean exactly the same thing. If you talk about one, the other is automatically implied. However, there is a subtext. Calling it one thing or another, we suggest, which elements we believe are moving and which ones are staying behind. The tendency seems to be, that people working on Fe-base steels talk about Ni-enrichment, while people working on Ni-base alloys prefer Cr-depletion.
I suppose it is not that surprising that I am changing my preference having moved from one materials to the other. As both expressions essentially mean the same, you may say this is not really an issue. However, I have confused myself and others frequently with this. Maybe it is necessary to say both every time.

Testing and Exposure
This is something that I believe is born, because people (i.e. me)  talk about things they don't understand. I work on materials characterisation. I don't expose things. I don't test things. Anyway. Testing - so I was told - refers to SCC testing (or tensile testing, hardness testing, etc.) where there is a ASTM standard defining the TEST procedure. If you're just boiling a piece of metal in an autoclave for a few month, to test - erm... find out - what sort of oxides are forming, that is an exposure.

Internal Oxidation

My suggestion is: Let's never say these words again! The problem is, we have no idea what this means anymore. Traditionally metallurgists speak of Internal Oxidation, if less noble minor alloying elements for oxides inside a more noble matrix at high temperature (>500 C) . These oxides can be intergranular or transgranular. They often form disconnected precipitates throughout the entire bulk of the material.
When Scott and LeCalver saw discontinuous Fe- and Cr- oxides at grain boundaries in Ni-base alloys exposed to water at PWR operating temperatures (~300 C), which are MUCH lower, they concluded Internal Oxidation is contributing to SCC.
Other groups have later found discontinuous oxides at grain boundaries in stainless steels. They saw the resemblence with Scott and LeCalver's results and Internal Oxidation was mentioned in connection with steels as well.
This is problematic, because the original definition does not apply in a steel. The oxides, continuous or not, are Fe- and Cr-oxides. These are the main alloying elements, not some minor additions. The reason for the preferential oxidation of discrete areas are higher defect densities and fast diffusion paths. There is no difference chemically.
In some cases there are Cr-depleted/Ni-enriched areas at the oxidation front. It has been argued that locally the definition of Internal Oxidation still makes sense. And it does sort of. We can probably agree that we cannot talk about SCC austenitic steels and Ni alloys without studying the Fe-Cr-Ni Pourbaix diagram.

I think we have thoroughly confused ourselves and we need to stop talking about Internal Oxidation in
connection with SCC. I think it makes more sense to talk about discrete intergranular oxides at least until we have a better idea.


Today, I am busy operating the atom-probe here at Oxford.

There are a number of texts and figures on the internet describing the basic principles of APT [1,2], so I am not going to reinvent the wheel here. However, I always like to know a little bit about the historical development of a technique, because the circumstances leading to it's invention often teach us a lot about the science involved and problems to expect. This is also true in this case. As it happens I already wrote a short introduction (including the above image) in my DPhil thesis [3]...

"In his PhD thesis, which initially was a general study of field electron emission [4], Erwin Muller designed the rst eld electron microscope [5], which consisted of a needle specimen in an evacuated glass tube and a phosphor screen. In 1951 the addition of hydrogen as imaging gas and the reversion of the polarity of the tip from negative to positive, led him to the invention of the fi rst field ion microscope [6]. M uller reported,

"Das Feldionenmikroskop ermoglicht zum ersten Male eine Abbildung mit genugendem Auflosungsvermogen,  um das atomare Gitter sichtbar zu machen."

English: The field ion microscope allows for the first time a projection with sufficient resolution to make the atomic lattice visible. He also suggested cooling the tip and discusses possibilities to decrease the necessary desorption field strength. This was the fi rst time that imaging gas atoms, field ionised by electrons tunnelling through the free space between the ion and the surface of the needle, were made visible and revealed the atomic structure of a tungsten tip surface [4].

The strength of the fi eld, F, on the apex of the needle is given by
F = V/krt

where V is the applied voltage, rt is the radius of the tip and k a numerical constant (k = 2-5) [4], dependent on the shape of the needle. As it was now possible to achieve a fi eld large enough to cause desorption of positively-charged ions, it did not take long before Muller reported the eld evaporation of other elements [7]. Finally the addition of a mass spectrometer led to the construction of a prototype 1D atom-probe in 1967. At this time an aperture was used to only allow atoms from selected regions to enter the mass spectrometer and be analysed. A high voltage pulse on the needle is superimposed on the standing voltage to stimulate field evaporation, so that the time of flight could be measured accurately.

If the voltage at the needle is V0 and the charge of the evaporating ion is n x e with e being the elementary charge, the electrostatic potential energy of the ion at the tip can be expressed as Epot = neV0. The kinetic energy at the detector is approximately Ekin = 1/2m d^2/t^2 if m is the mass of the ion, d the distance to the detector and t the time of flight. This approximation is correct, if we assume a grounded counter electrode directly in front of the tip, so that the ions reach their final speed instantly [8]. With the laws of energy conservation it follows that

neV0 = 1/2m d^2/t^2.

From this it is easy to see that the mass-to-charge ratio is proportional to the square of the time-of-flight t^2:

m/n = 2eV0 t^2/d^2.

This way single selected atoms could be identi ed for the fi rst time.
This concept was picked up by Cerezo et al., who added a position sensitive detector and constructed the fi rst fully operational 3D atom-probe in 1988 [9]. Several improvements to the technique have been implemented and become commercially available since. Tsong et al. showed that fi eld ionization and eld evaporation can also be photon-stimulated [10] and laser fi eld ion microscopes and atom-probes were implemented.

The addition of a local electrode has improved mass resolution and fi eld of view and requires lower applied voltages to achieve sucient eld strengths for ion evaporation. It also allows the examination of multi tip coupons, as sucient evaporation elds are only achieved for one tip at a time [11,12]. A disadvantage of the local electrode is that material from fractured specimens can contaminate the electrode, which will then start
field evaporating itself. Therefore these electrodes have to be easily exchangeable in a sensible set-up.

The Local-Electrode Atom-Probe

Although the Cameca LEAP 3000X HR (TM) [13] (the instrument I am using at Oxford) looks a lot more complicated than the simple apparatus Erwin Muller designed, the basic principle remains the same. Atoms and molecules are eld evaporated and eld ionised from a thin, needle specimen. This can be achieved by traditional voltage pulsing or laser pulsing. As mostly less-conductive oxides were studied in this thesis, laser-pulsing was used throughout. The ions are accelerated in the field of an exchangeable local electrode. They pass the reflectron, which further improves mass resolution. The time-of-flight is recorded when the ions hit a MicroChannel Plate which locally ampli es the signal at the point of ion impact before it is detected with a position sensitive detector. With a thorough knowledge of the instrument geometry and an estimate of the field at the tip it is possible to reconstruct the sample in 3D."

[1] http://www-fim.materials.ox.ac.uk/
[2] http://en.wikipedia.org/wiki/Atom_probe
[3] K Kruska, DPhil thesis: Understanding stress corrosion cracking mechanisms, Oxford University 2012
[4] M Drechsler. Erwin Muller and the early development of fi eld emission microscopy. Surface science, 70:1-18, 1978.
[5] E W Muller. Zeitschrift fur technische Physik, 17:412, 1936.
[6] E W Muller. Das Feldionenmikroskop. Zeitschrift fur Physik, 131:136-142, 1951.
[7] E W Muller. Feldemission. In Ergebnisse der exakten Naturwissenschaften, volume 27 of Springer Tracts in Modern Physics, pages 290-360. Springer Berlin / Heidelberg, 1953. 10.1007/BFb0110808.
[8] M K Miller. Atom-probe Tomography : Analysis at the Atomic Level. Kluwer Academic/ Plenum Publishes, 2000.
[9] A Cerezo, T J Godfrey, and G D W Smith. Application of a position-sensitive detector to atom-probe microananlysis. Review of Scienti c Instruments, 59:862-866, 1988.
[10] T T Tsong, J H Block, M Nagasaka, and B Viswanathan. Photon stimulated fi eld ionization. The Journal of Chemical Analysis, 65:2469-2470, 1976.
[11]T F Kelly, V Patrick, P P Camus, D J Larson, and L M Holzman S S Bajikar. On the many advantages of local-electrode atom probes. Ultramicroscopy, 62:29-42, 1996.
[12]T F Kelly, P P Camus, D J Larson, L M Holzman, and S S Bajikar. Us patent 5,440,124, 1995.


    Science blog, just writing about some experiments I'm doing, conferences I'm visiting and random things that may come to my mind.


    June 2013