The following is part of a discussion between scholars concerning the mechanical processes which went into the medieval Islamic creation of Damascus steel sabers using raw wootz steel from the Indian subcontinent. In 2006, a paper appeared in Nature describing what Peter Paufler and colleagues identified carbon nanotubes as part of the underlying mechanical process. I covered that research here in 2006, and it remains a popular article here at Archaeology at About.com. In late 2012 I was contacted by Madeleine Durand-Charre, a scholar who expressed her doubts about the identification of nanotubes; and in April 2013, Professor Paufler responds here.
- Damascus Steel - Sword Makers of the Islamic Civilization, About.com coverage
- Nanoscale Precipitation in Genuine Wootz Steel Blades, Madeleine Durand-Charre
Dear Madam,
Thank you for sending a copy of your review 'Les aciers damassés décryptés' and for the comment - dated 7 February 2013 - on your view about our papers. Let me first apologise for my late response. I enjoyed the review very much and - stimulated by Kris Hirst - want to add a few words concerning our point of view.
(i) Your objection against our description of Fig. 1
Looking more closely at Fig.1 (taken from our publication Kochmann et al. 2004) with the arrows included in your letter of response, you could convince us that this microstructure resembles a bainite reaction. However, this micrograph stands out, because we had employed there several sophisticated techniques of ion beam etching to prepare the surface. So the resolution is far better than that of other micrographs we got. We thank you for drawing our attention to this. As you noticed yourself, the correct identification of bainite and related microstructures remains a matter of discussion in the literature. We relied more on X-ray diffraction analyses and reported that bainite had appeared in our sabre, however, among other phases (Reibold, Levin et al 2006).
(ii) Is Fig.1 representative?
Unfortunately, the situation recorded by other micrographs is not that clear. Note that we have found definitely distinct microstructures as a function of distance from the cutting edge, regardless of their resolution (see also (iii)). Hence, not knowing the thermal history of our historic specimens it remained difficult to distinguish between bainite and pearlite in cases of poor resolution of the micrographs. When bainite had appeared in our X-ray diffraction analyses, then the signal was averaged over the entire height of the blade of 16 mm. Regions like that of Fig. 1 might have contributed to the signal while the rest did not. The answer to (ii) is no.
(iii) Results of other authors concerning the identical Damascene blades
As you know, we had specimens of two historic sabres at our disposal (delivered by the Berne Historic Museum), parts of which were also examined by previous authors B. Zschokke (Zurich 1924), J. Piaskowski (Krakow 1978), J.D. Verhoeven, A.H. Pendray et al. (Ames 1998), and J.D. Verhoeven (2002). This is a rare fortunate circumstance for comparison, some kind of unintended round-robin.
Zschokke identified pearlite along with cementite in them. At his time neither bainite nor sorbite were known. Piaskowski then recognised cementite 'in a sorbitic matrix'. From metallographic studies, Verhoeven et al. (1998) concluded that pearlite dominated near the cutting edge and the so called divorced eutectoid transformed (DET) matrix occurred near the fat end of the blade. Verhoeven and Gibson (1998) insisted that DET is different from bainite but admitted that other authors take the opposite view. That variation of microstructure as a function of distance from the cutting edge (due to changes in the cooling rate) is just what we have observed, too. That means there was no unique microstructure throughout our sabres. On the other hand, the numerous TEM images we have taken both in near-edge and far-edge regions looked similar at nanoscale, i.e., we could not attribute characteristic differences to them.
(iv) Consequences of (i) through (iii)
Now I return to the nanostructures observed by us. Our main points were that (a) they exist, (b) they are arranged in colonies of parallel wires or tubes, (c) traces of plastic deformation occur only occasionally inside those colonies, and (d) various phases of Fe-C forming the wires have been identified by comparing X-ray diffraction (XRD) pattern and Fourier transforms of High Resolution Transmission Electron Microscopy (HRTEM) photographs, among them cementite Fe3C dominating. These are plain facts without reference to a mechanism of formation.
However, we also have expressed, in the discussion sections of our publications, ideas concerning the formation of nanowires and nanotubes. These are proposals or speculations linking experimental data of other authors with our results. And this is the point where we agree with you: We have to differentiate between matrices of bainite, pearlite or other type, when we try to model those formation processes. By now it seems that most of the models fit the situation pearlite better, although I am not aware of concrete models connecting the bainite microstructure with the kinetics of cementite colony evolution. That means, we have implicitly assumed a pearlite microstructure unless the model is independent of the structure.
(v) Conclusion
We are thankful for your critical remark drawing our attention to the bainite microstructure and mechanism. The microstructure of our blades studied was inhomogeneous, i.e. samples taken for TEM originated from either pearlitic or bainitic or transitional regions. On the other hand, the nanostructures observed did not reflect the distinct origin obviously. This is why most of the models used (see Appendix) do not refer to a definite microstructure. Knowing that bainite was partly present in our sabres, alternative models of the formation of nanostructures should be worked out for comparison. We would gratefully acknowledge relevant hints to the literature, if existing.
Yours sincerely,
Peter Paufler (on behalf of the co-authors)
See the appendix for additional considerations.
