50

A Katana needs to be made of steel folded 1000 times, according to eHow:

Pound the tamahagane into two separate blades. These need to be heated, pounded by hand, and folded 1,000 times each to remove the impurities. Pound one of them another 1,000 times to make it harder than the other.

But, according to simple mathematics and physics, this would mean that a Katana is made of 21000 layers, or 10 followed by 300 zeros!

Note that this doesn't really mean it was not done, though.

So - were Katanas really made of folded steel? How many times was it typically folded?

Mad Scientist
  • 43,643
  • 20
  • 173
  • 192
Sklivvz
  • 78,578
  • 29
  • 321
  • 428
  • 21
    I think they meant 1000 layers which would only mean 10 folds – ratchet freak Feb 05 '12 at 13:10
  • 5
    It doesn't seem a necessary thing that the every bit of steel is folded in each fold. In fact, that's part of the process - they would have [patterns folded into them](http://www.trueswords.com/musashi-sharp-damascus-steel-katana-sword-4096-layers-p-4992.html). [Wikipedia](http://en.wikipedia.org/wiki/Japanese_swordsmithing#Forging) suggests up to 65000 layers. – Jivlain Feb 05 '12 at 13:33
  • **Note**: this question was asked as part of the "[topic of the week](http://meta.skeptics.stackexchange.com/questions/1371/topic-of-the-week-weapons)" initiative to raise our questions per day stat. Please contribute some great questions! – Sklivvz Feb 05 '12 at 14:22
  • 65000 layers makes sense, that would mean 16 folds (2^16 = 65536) – vartec Feb 05 '12 at 14:44
  • 27
    ehow has got to be one of the worst sites on the web. – Sam I Am Feb 05 '12 at 15:04
  • 2
    I seem to recall that modern metallurgists figure that between 10 and 20 folds is optimum depending on a lot of things (metallurgy being a ridiculously complicated business), and that the best Japanese sword smiths had come very close to the optimum by craft methods. – dmckee --- ex-moderator kitten Feb 05 '12 at 17:19
  • 1
    Documentary on Disc or NGC fa few years ago followed Japanese swordsmiths and showed their work. They fold the metal 10-20 times (can't remember exactly, might depend on the smith and/or the customer/purpose of the weapon). Would make this an answer but can't find a video. – jwenting Feb 06 '12 at 07:00
  • 8
    You could fold the metal 1000 times without the result being an unfeasible number of layers: the layers merge slightly on folding and don't remain strictly distinct. This could, in principle result in changes to the alloy crystal structure and chemical makeup. The 1000 folds might be required as the changes only happen slowly. The argument about the resulting number of layers is irrelevant. – matt_black Feb 12 '12 at 16:54
  • 1
    Homeopathic swords, perhaps? Folding it 1000 times would result in layers **much** less than one atom thick. What would be the point? – Loren Pechtel Apr 01 '15 at 21:47
  • 6
    @Loren Pechtel: I'm neither a metallurgist nor a swordsmith, but perhaps the repeated folding and stretching alters the microstructure of the metal? As a parallel, consider that when kneading bread I fold it roughly 300 times, a process which forms gluten into long chains, and allows the bread to rise: http://foodreference.about.com/od/Tips_Techniques/a/The-Science-Of-Kneading-Dough.htm – jamesqf Apr 02 '15 at 18:06
  • 1
    Looks like they are conflating "pounds" and "folds," also. – PoloHoleSet Jan 06 '17 at 15:05
  • 1
    The real key is to get equally thick layers of butter and pastry. Fold, then roll to original size, fold then roll to original size, and repeat ... Wait a minute, I think that's puff pastry not swords. I got confused cos of "mille feuille" and "1000 folds" ... – Grimm The Opiner Feb 24 '17 at 12:32
  • Law of diminishing returns probably kicks in long before a thousand folds. –  Apr 03 '17 at 20:29
  • The Swords were folded and hammered mainly to get rid of impurities and to strengthen the blade...9 folds would produce over 1000 layers – rex Apr 03 '17 at 20:08
  • 1
    The folding, as stated, is part of the process to remove impurity, and likely to add carbon to the steel. The entire point is to get a homogeneous, pure, lump of steel at the end. Any Damascus and steel grade layering is done after, and in addition, to this step. When you are working with something illiterate peasants just mined out of the ground (instead of modern steel ingots), you need to spend a long time purifying and fortifying it. – Jonathon Apr 04 '17 at 20:36

1 Answers1

58

According to SamuraiSwords.com, there are two folding steps, in first traverse and longitudinal folding is repeated:

The difference between these two folding procedures lies on that the small hammer is used to fold in transverse folding while the big hammer is used in longitudinal folding. Generally, about 12-15 folings are repeated.

In second (optional one):

another forging step can be added (Age-kitae). Tamahagane is forged again as a shape of stick once more and cut uniformly about 7.5cm long. This age-kitae is an optional step and sometimes omitted. A tight microstrucutre and hada (grain pattern) can be expected after this step and the uniformity of the carbon content is improved.

Likewise the shita-kitae, tamahagane is heated and folded 7 to 8 times.

Of course this is just process of creating 2 or 3 grades of steel (hard, soft and optionally medium). Then they are combined, making greater total number of layers, how many depends on the style:

katana fabrication

As for "1000 folds"...

shita-kitae, is repeated from 8 to as many as 16 times. After 20 foldings, (220, or about a million individual layers), there is too much diffusion in the carbon content, the steel becomes almost homogenous in this respect, and the act of folding no longer gives any benefit to the steel.

(source: Wikipedia, referencing A History of Metallography by Cyril Smith - The MIT Press 1960 Page 53-54)

Also, anything above 25-26 folds would make no sense at all, as by then the layers' thickness would have to be less than radius of iron atoms, which is 126 picometers (= 1.26 × 10-10 meters)

Community
  • 1
vartec
  • 26,581
  • 5
  • 97
  • 155
  • 17
    The argument about how thin the layers get is irrelevant: they don't remain distinct when folded. The whole point of the repeat folding is to homogenize the metal and encourage surface contaminants/additives to migrate into the bulk metal improving its properties. – matt_black Apr 05 '12 at 12:02
  • 4
    You could add that folding it 10 times would imply 1000 layers. – ChrisW Nov 25 '14 at 22:12
  • 9
    1024, if you're going to be really pedantic. :) – GordonM Apr 01 '15 at 09:50
  • 1
    @matt_black While they may not be distinct down to the individual layer, you can get interesting patterns of at least some of the layers from acid etching a folded sword, IIRC. – SGR Feb 24 '17 at 15:26