Phillipe, and everyone-

    This has been a fascinating discussion for me, as Skip and I have
recently begun experimenting with (intentional) manganese additions in our
bloomeries. Many of the questions we've been asking have been answered in
this discussion.
    The point that may not have emerged too clearly, though, is that the
main influence of Mn on carburisation is not from reduced Mn in the metal. I
think it's main effect is indirect, from the characteristics of an Mn rich
slag.

    In the hearth zone of a bloomery, carburised iron in contact with iron
rich slag will reduce iron from the slag, and decarburize the metal. In a
hot slag bath, decarburization and reduction are two sides of the same coin.
    In past direct bloomery steelmaking attempts, we found that even when we
created a high carbon bloom, much of the carbon would be lost in the early
stages of bloom consolidation, while the metal is still surrounded by lots
of high iron slag. If we purposely smelted until we had a low iron slag, the
high melting temperature of that slag caused great difficulty in welding the
bloom into a good solid billet that wouldn't delaminate.
    So we see two basic approaches to overcome this problem-- either 1)
eliminate a large portion of the slag from the iron, or 2) create a fluid
slag that doesn't continue to reduce/decarbrize the metal. We are finding
that the manganese helps to do both these things.

     The first approach, of just eliminating the decarburizing slag, is to
consolidate the bloom to eliminate most slag, and then remelt and carburize
it in the bloomery or a charcoal fire, as described by Aristotle, and a
millenia or so later by Evenstad. But this requires a two-step process, with
significant waste. But it works well, we've tried it.

    The second approach: with an Mn-rich slag, the Mn is much less available
to reduce/decarburize, and the bloom can retain more carbon. We tried
manganese dioxide additions  in a smelt a few weeks ago, and though we did
not have highly reducing conditions (only 14 inches of stack height), a
discrete half of the bloom was steel that retained its carbon through
forging, and was very similar to the remelted steel mentioned above.
Estimated .8 C by spark test.

    What was most surprising, though, when we made a bloom with manganese,
was how fluid and unsticky the slag was. (I guess the scientific term would
be non-wetting). When the bloom was removed from the furnace, the slag just
ran right out of the bloom. When cutting or forging the bloom, any remaining
slag ran out immediately into a puddle on the anvil, rather than squirting
out with each blow as a regular fayalitic slag does. So that the slag is
almost self-eliminating, accomplishing the same thing as the Aristotle and
Evenstad processes.

Hope that's reasonably to the point...


Lee

Lee Sauder
Woods Creek Forge
229b McLaughlin St.
Lexington, VA 24450
(540) 463-6632
web: http://iron.wlu.edu


----- Original Message -----
From: "dillmann" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Sunday, May 15, 2005 5:30 AM
Subject: Manganese in bloomery process


> Dear all,
>
> For several month, I'm analysing ferrous artefacts coming from direct
> process and that are made from an ore containing high levels of Manganese
(a
> 14th c. forge in the French Pyrenees). The slag inclusions contain up to
> 30mass% of MnO and Mn is not detectable in the metal (<100ppm). Indeed,
the
> metal is highly carburised (over eutectoid).
>
> Several authors talk about the carburising role of Mn and, indeed, several
> historical zones of producing steel were associated with high Mn ores
> (Italian alps, Pyrenees...)
> Nevertheless, I have several questions about the exact metallurgical role
of
> this element in the carburising process during bloomery operating chain.
> Indeed, in the thermodynamic conditions of the shaft furnace, the MnO
oxide
> can not be reduced into metallic Mn. Thus, this element can not play a
role
> in the metal. Moreover, even if this element were in the iron, it seems
that
> the influence of Mn on the diffusion coefficient of carbon is not proved.
>
> So, I ask the questions:  does anyone study the exact role of manganese on
> the carburisation of iron in the direct process. Does Mn play a role
indeed?
> Can you help me?
>
> Best regards
>
> Philippe Dillmann
>