String of Pearls Tekkotsu

[JohnTo]

One of my first tsuba was a tosho tsuba from the RB Caldwell collection with a couple of prominent linear tekkotsu on the rim and so I thought that I knew everything about ‘iron bones’.  However, I found ‘exploding’, ‘granular’, ‘fine’ and ‘speckled’ tekkotsu mentioned in books and catalogues, but examples of these were never actually illustrated (the tsuba were always photographed face on).  Consequently, my knowledge of tekkotsu is from poorly described books and my own observations of my collection.  I just love the various lumps and speckles that give iron character and so I would like to present, for your critical comments, a new phenomenon that I observed recently, namely String of Pearls tekkotsu (named after a Glenn Miller tune).

The tsuba in question is a mokko shaped Kyo sukashi tsuba with a bobbin (or is it a central part of shippo circles?) with karagane links.  Identical tsuba to this have been offered for sale by Aoi Art and Ebay, both with NBTHK Hozon for Kyo Sukashi around 1600.  Mine came from a job lot from an old collection bought in 2017, which also had some other nice pieces, so I believe that it has a ‘provenance’.  Recently, one late Friday afternoon, I was looking at some tsuba while having my usual end of week indulgence of a pint of Old Peculiar, which I find heightens perception and enables me to hear the ‘iron speaking to me’.  This particular day was spent outside as it was a warm summer’s day and the evening sunlight caught some shiny spots on the rim of this tsuba that I had not noticed before.  ‘Fine granular tekkotsu’ I thought, but on close inspection I noticed that the spots were in lines and appeared on all four lobes around the mimi.

At this point I expect that some of you guys are jumping up and down and shouting ‘cast iron!’.  While I would not profess to be an expert on cast iron tsuba I have identified five in my collection, which I inadvertently bought in mixed lots from on-line auctions (poor photos is my excuse).  This tsuba does not have the seams inside the sukashi, the characteristic look of cast iron, odd looking seppa dai, etc. of my cast examples.  That is not to say that some Japanese artisan of old was not skilled in casting iron which looked exactly like forged iron.  However, it looks genuine to me.

Here is my explanation of what I’m seeing.  The tsubako folded an iron blank plate in on itself, in the same way as a swordsmith would (attached to a handle, tekkobo?).  Before folding, the outer surfaces of the iron absorbed carbon from the charcoal fire (maybe left in too long) and when folded and welded together formed a thin layer in the middle of the plate of high carbon iron.  The outer surface of the plate would also be expected to be of a similar composition, but the two outer surfaces would be subject to subsequent hammering, loss of iron from scaling in the forging process and then filing as the tsuba was made.  Cutting out the blank tsuba from the plate would result in loss of an iron strip on all sides, including the folded edge.  Thus all sides would be of identical forms of iron and contain a thin layer of high carbon iron right up to the edges.  After the tsuba was shaped (including cutting the sukashi) it was given a final heat treatment.  I don’t have any evidence (literature is blank and 17thC tsubako left no visual examples of their working practices on YouTube), but I believe that some tsubako protected the tsuba against burning in the final heat treatment by covering with different thicknesses of clay, like swordsmiths do.  A final quench produced what would be called on a sword, sunagashi (drifting sand), i.e. lines of martensite particles, nie, along the weld lines between the iron layers.  The difference between nie seen on a sword and a tsuba being that a sword polisher produces a flat surface and the nie becomes visible as changes in the colour of the iron.  Such features in tsuba are normally hidden by the patina, but in this tsuba the nie are prominent because they stick out as granular tekkotsu and the patina has partially worn away.  The swordsmith Yoshindo Kajihara (The Craft of the Japanese Sword, p 91-92) states that ‘quenching will produce bright grainy effects in the steel called nie…and sunagashi-all patches of martensite but with different names according to their size or location.  The longer a blade is heated the larger the size of the austenite grains that form and thus the larger the size of the martensite crystals that remain in the final blade.’  I believe martensite has a lower density than the softer pearlite, so would stick out from the surface, as well as being the reason that a sword blade curves during quenching.

The String of Pearls is difficult to see, and more difficult to photograph, and is not visible as a continuous line all the way round.  One string stretches allmost the entire length of one of the lobes.  However, it is visible on all four lobes and had they been joined together ‘necklace’ would have been a better description.  To aid identification of the feature, I have included photos (with the spots circled) from a section of a blade by Midzuta Kunishige that has prominent lines of sunagashi (some tsuba collectors may be unfamiliar with features found on blades).

Inspired by this discovery, I have bought a pair of magnifier eyeglasses and checked other iron tsuba in my collection.  Some do in fact appear to have String of Pearl tekkotsu, some consisting of just four or five dots, but some are longer, though none are as prominent as in this tsuba.  Generally, they appear as a line of nanako type spots, about 0.2 mm wide.

I look forward to reading your critical comments.

Best regards, John

(just a guy making observations, asking questions, trying to learn)

 

[Stephen]

 

Interesting John!

[ROKUJURO]

John,

a nice TSUBA with a lot of KARIGANE (not karagane)! Does not look at all cast to me.

TEKKOTSU is a very interesting topic!

The problem is that all literature I have seen about TEKKOTSU is based on assumptions (made by non-metallurgists), and no scientific evidence has ever come up. I am working on this subject but I cannot present my thesis as it is not finished yet, but from my observations, I will give you some food for thoughts:

Carbon migration in iron/steel is a very slow process. You need a minimum temperature of 900°C to start it, and it takes hours to transfer very small amounts of carbon into an iron or steel surface. Time, temperature, the metal mass and surface, and the exclusion of oxygen are crucial factors for this. On the other hand, the carbon content of a steel layer imbedded in low-carbon iron will follow these physical laws and release its carbon content into the iron environment. We call that homogenization, and it works faster when the layers are thinner.

The viscosity of steel is considerably higher at forging temperature than that of iron. So there is no way of "pressing" out steel lumps to the surface of the iron, in the contrary, I have shown in my test samples that it is exactly the other way round. In a composite test piece consisting of iron and steel sheets, the softer iron protrudes at the edges.      

There is no evidence for quenching (= hardening) of TSUBA, in the contrary, TSUBA have to have some considerable resilience and plasticity to be able to parry a stroke without breaking. Instead there is the story of one YAGYU Samurai (don't have the name at hand) who tested TSUBA for their quality in trying to break them (story goes about a mortar being used for that). 

This brings us to the use of a TSUBA. Some authors insist that a TSUBA is mounted on a blade predominantly to prevent the hands of the SAMURAI to slip onto the blade in a thrusting motion. While this will certainly work, I do not believe this to be the main reason for the mounting of a TSUBA on a sword. I think that catching an opponent's blade with the MUNE and having it slide down is a risk for the hand which will be reduced by a TSUBA. For this purpose, the outer dimensions of a TSUBA are not that important.

I have seen a good number of TSUBA with KIRI-KOMI, so that the protective aspect of TSUBA may not be considered too small.

Going a bit into metallurgy, you are correct in that the density of hard martensite is lower than that of pearlite, which causes some stress in the blade in the YAKIIRE process. On the other side this does not cause the martensite to come to the surface like air bubbles in water! Martensite is a structure that goes relatively deep into the steel - we are talking about a max. of 4 mm from each side. Repeated polishiing can reduce the width and the appearance of HAMON.

In the TOGI process, very special natural stones are used which are chosen by the TOGISHI following his experience and knowledge about the given blade. In the later steps of a good polish, ideally soft stones (and a lot of time) will be used to remove steel from the blade surface without grinding the martensite lumps (NIE and NIOI) completely flat. In magnification, you can see NIE and NIOI protruding out of the surface like very small spheres. In cheap polishes, everything is flat and level and only visible by heavy HADORI.

There is a big difference in the final treatment of blades and TSUBA: While the former should never be treated with acids, TSUBA ofter undergo an etching process necessary to create the desired 'old' or 'used' appearance which is not always the same as patina. In my experience, carbon steel is more easily attacked by acids than low-carbon iron, so if the superficial phenomena we call TEKKOTSU were carbon steel, they would be attacked by the acids in the first place.   

Some authors have speculated about TEKKOTSU being lumps of hard steel which remained when the surrounding iron of a TSUBA was worn away in the use by rubbing on clothes. This is very far away from any probability.  

I will not go further down in the abysses of steel and forging,   

Kind regards,

Jean C. 
 

[christianmalterre]

using acids wont fruit successfully enough to create a so called "Bone"....

you had to use a Base!

 

especially on low tempered higher carbon iron.

 

in sum but, i do fully agree with you.

 

Christian

[zanilu]

Jean. Your thesis looks very interesting. It will be available for the public when finished?

 

As for the function of tsuba, in certain Japanese sword schools the tsuba is specifically used for parring sword slashes. This require resilience not hardness... This school dates back to Toyotomi Hideyoshi times, the founder of the school was his sword master.

 

Just my two cents thoughts...

 

Luca

[Japan2112]

That's a really nice tsuba. Usually if you see karigane you have Owari, but the overall feel is Kyo. I like the iron in your pictures.

Sunlight does strange and wonderful things to tsuba. I think early morning or near sunset light usually brings out the "best"

[ROKUJURO]

For John: Different form but similar design

[JohnTo]

Thanks Steve, nice reply, so here are some more of my thoughts.

Steve states that there is no evidence that tsuba were subjected to yakiire in similar way to swords.  He may be right, but then is there any evidence no tsubako used yakiire?  As a scientist I was always taught to take note of existing dogma, but to listen to the voice of the system and then form your own theories.  I would like to present what I think is evidence for yakiire from one of my favourite tsuba (I have posted this tsuba before in a different context).

I believe that this tsuba is from the Kanayama School, its simplicity IMHO is just pure Zen.  I just love it, especially as I got it cheap as a previously unsold lot from an old collection, complete with old kiri box.  A simple star, bobbin or shippo component forms the basis for the design (like the one in the original posting).  There are stumps from pairs of bars going from the star to the mimi, which appear to have been cut off in historical times.  Holding this tsuba in my (gloved) hand I can hear the iron speaking to me.  Of course it might be telling me porkies, but this is what it is saying.

The first thing of interest in this tsuba is the mimi (the outer rim).  It is heavily pitted, like a teenager with bad acne, but there are no signs of rust scabs.  This pitting is uniform around the whole rim.  Unfortunately, I have seen more rusted mimi than I care to but I have never seen one with rust evenly distributed around the whole tsuba.  Corrosion has always been in patches and usually around the whole mimi, not just the outer rim.  This pitting must have been caused by over heating in a fire causing the iron to burn (done this myself on a one day blacksmith course!).  So why is the pitting only on the outside of the mimi?  I have no other explanation than to assume that the rest of the tsuba was protected from the fire in the forge either by coating with clay (as in sword making) or masking with a couple of iron discs (unlikely).

The ji (faces) of the tsuba are in good condition and do not show signs of corrosion.  A little ishime finish, but not pitted or scabby rust.  The other feature of the ji is ‘stars’, tiny specs less than 1 mm that sparkle in the light.  I assume that these are hard martensite (nie) particles that probably cannot be called tekkotsu as they are flush with the surface.  They are scatterred all over the ji, on the arms of the shippo/star and around the seppa dai, on both sides.  Maybe these particles did not take on a patina as readily as the surounding iron; I doubt that they lost their patina and became shiny due to rubbing, as tekkotsu on the mimi of some tsuba may have done.

The edges of the central star are sunk in the centre.  I cannot imagine that the tsubako cut them this shape.  They would have been cut and filed straight and vertically to the face of the tsuba. I believe that the middle part of the iron shrunk during yakiire, the central iron being softer than the hardened surfaces.  There is no evidence of san mai awase construction of the plate that I can see and no evidence of lamination, as in Akasaka tsuba (of course the tsubako may have been very skilful in welding plates together).  Interestinly, the cut edges of the shippo/star also have small (less than 1 mm) specks standing proud from the surface.  They look like grains of sand and I have to resist the temptation to pick at them to see if they would fall off like sand would, or whether they are iron and welded to the surface!

All this seems to me to point to a tsuba that has been covered in clay and subjected to yakiire as a final stage in its manufacture.  If Steve has another explanation of these phenomena I would pleased to hear them.

I have often heard that iron tsuba were not hardened as they might shatter during impact.  But consider the main purpose of the tsuba.  In some fencing schools emphasis may have been to stop the hand slipping down the blade in a thrust attack, particularly if the oponent was wearing armour.  In this case a cheap cast iron tsuba would have been sufficient for the task, despite being likely to shatter if hit with a metal object (sword).  Note: I am not advocating the use or collection of cast iron tsuba.  Ugh!

Considering the rather insubstantial tsuba presented herein, a hardened mimi may may indeed break under a sword blow, but unless the cut was well executed, something unlikely during combat, the outcome may equally well be a chipped or broken blade.  Edven if the mimi was cut through the blade would then come up against the star/shippo centre, with a hard outside and soft impact absorbing middle, again with an equal possibily of the blade, rather than the tsuba, coming off worse.

Also, please note that I did not say that martensite floats to the surface like bubbles.  Iron is pretty solid, even when hot, so not much floats around!  I am quoting Yoshindo when I say that iron with 1-2% carbon will tend to form large crystals of austenite when subjected to long periods of heating (the longer, the bigger).  It is only upon yakiire that these crystals become martensite, forming the iron lumps that we, tsuba collectors, call tekkotsu and sword collectors call nie (in their various guises).  I believe that austenite only forms at high temperatures (above 723°C according to an on line iron phase chart), quenching from a high temp causes the austenite structure to ‘flash freeze’ as the martensite crystals we love to see, whereas slow cooling of iron reverts austenite to the softer cementite and pearlite; if I remember correctly from my student days.

You also mention the old howler about tekkotsu only developing after rubbing away the soft iron.  I used to think this was ridiculous as well, but realised that what rubbing actually does is to preferentially remove the brown/black oxide patina from tekkotsu making it more evident to the naked eye, i.e. rubbing does not form tekkotsu, but it might make it visible, as im my original post regarding String of Pearls tekkotsu.

In summary, I have three observations that point to a final heat treatment (yakiire) of this tsuba:

1.  Burning of the iron outer surface of the mimi
2. Shrinking of the (softer) core iron with relation to the outer (hard) surfaces
3. Presence of martensite stars in the ji.

As I said, I have no evidence for any of the above; I’m just carrying out what we scientists call ‘A thought experiment’.  But that’s what studying and collecting is all about, challenging existing dogma, engaging in debate, even if your ideas are wrong.

Best regards, John

(just a guy making observations, asking questions, trying to learn)

 

[ROKUJURO]

John,

an interesting but unfortunately quite corroded TSUBA. I cannot see the bright speckles properly, so no attempt for an analysis here.

Concerning your text:

... I am quoting Yoshindo when I say that iron with 1-2% carbon will tend to form large crystals of austenite when subjected to long periods of heating (the longer, the bigger).  It is only upon yakiire that these crystals become martensite, forming the iron lumps that we, tsuba collectors, call tekkotsu and sword collectors call nie.... ,

l will only add that in this statement. YOSHINDO is correct (it applies as well to low-carbon iron), but your assumption is not. I will shoot some sample photos as soon as I am able to, and you will see what he meant.

As I wrote in my long comment above, we cannot be sure that martensite particles we find in swords as NIOI or NIE are the same as TEKKOTSU in TSUBA. You are of course right to say that YAKIIRE could possibly be executed on TSUBA as well, but there is no evidence and no good reason for that. A major thought is that to produce martensite effects in a workpiece, you have to make sure that there is sufficient carbon content in the metal. This starts theoretically at 0,22% of carbon content, but you would need about 0,5% C minimum to really see some effect. An item made from raw iron will never produce any martensite irrespective of how often you quench it. 

[ROKUJURO]

I have finally shot some (unfortunately not so good) photos of the surface of broken wrought iron pieces wich were submitted to long-term heating. Both are old; the thicker piece with the crystals may date to late medieval times or early 17th century. It was part of a punching tool to make holes in a workpiece. The other flat piece is a part of an old wagon wheel, dating at least into the pre-industrial times or before.

Both items show pronounced crystal forming caused by repeated heating and slow cooling. This has nothing to do with martensite as both materials do not contain any carbon at all.