Note: Descriptions are shown in the official language in which they were submitted.
~036~44
The subject invention is addressed to ferrous metallurgy, and
more particularly, to addition agents for incorporating magnesium in ferrous-
base melts.
As is known, it is virtually the most conventional of metallurgical
practices to innoculate or otherwise treat molten irons for purposes of deoxi-
dation, desulfurization, degasification, modification of the as-cast morphological
structures, etc. Magnesium has seen extensive use in this connection and has
been largely used in the production of "ductile iron", i.e., cast iron in which
at least part of the graphite is present in spheroidal form. This latter develop-
ment, early described in U.S. Patent No. 2,485,760 to Millis, Gagnebin & Pilling,
significantly closed the gap between low-cost but brittle cast irons and the more
expensive but less brittle steels.
In any case, over the years research has continued in respect of the
development of improved procedures for introducing magnesium into a cast
iron melt owing to its high degree of reactivity and propensity toward violent
reactions. Too, in recent years increased emphasis has been and continues
to be given to ecological considerations, namely, smoke emissions (attributable
largely to evolved magnesium oxides) . Since most approaches to the problem,
as does the present, involve the production and use of addition agents, "crusha-
bility" of the additive as will be explained herein, also plays an important role.
All these factors, in turn, focus attention on the economic picture.
In accordance with the present invention, it has been discovered that a
hereinafter more fully described addition agent, provided that it is controlled
to contain special and correlated percentages of magnesium, nickel, silicon
and iron, results in a material having a unique "combination" of attributes,
particularly for ductile iron production, including (i) relatively low reactivity
with molten iron and thus low smoke emission, (ii) excellent magnesium
recovery both in terms of preparation of the additive and in respect of the
ferrous-base product produced, (iii) and good crushability such that excessive
--1--
1036844
fines can be avoided, these obtaining (iv) with the simultaneous capability of
using relatively low silicon levels and (v) re]atively low cost. Prior art
ductile iron addition agents have possessed one or more of these characteristics
but, insofar as we are aware, none has combined all of them in one additive.
Moreover, apart from ductile iron production, it is deemed that these novel
addition agents also can be used in treating both wrought and cast low alloy
steels, particularly in minimizing the difficulties occasioned by reason of the
contaminant sulfur.
Generally speaking, the present invention contemplates addition agents
containing about 3%to 5.5%magnesium, about 19%to 24%nickel, about 15%to 28%
silicon and the balance essentially iron, the iron preferably being not less than
45%. Elements such as copper and manganese are not essential. Manganese
need not exceed 2 or 3% although it can be as high as 12%.
In carrying the invention into practice, should the magnesium be too
low, e . g ., 2%, the addition alloy is rendered too costly since a larger alloy
addition would normally be required to produce a given magnesium level in a
treated iron, notwithstanding a possibly higher magnesium recovery. On the
other hand, based upon prior metallurgical principles it would be expected
that as the magnesium is increased greater would be the reactivity and larger
would be the obnoxious emissions, (i.e., magnesium recovery and smoke genera-
tion would be expected to be roughly inversely proportional), this at the expense
of smaller magnesium recoveries and higher cost. This need not be the case
particularly if the nickel content is correlated with the magnesium percentage.
Within the chemistry contemplated herein, nickel promotes an increase in
magnesium solubility in the addition agent. And, accordingly, if sufficient
nickel is present such that the magnesium is soluble or substantially so in the
addition agent, less smoke emission is encountered and magnesium recovery is
enhanced in treatment of, say, a ductile iron. While the magnesium level need
not exceed 5 . 5%, it can be present up to 6 or 7%.
--2--
- 1036844
With further regard to the nickel content, it need not exceed 23.5%
although up to about 25% can be present. And while it can be as low as 15
or 16%, particularly at the lowest magnesium levels, it is deemed that a range
of 19.5% to 23.5% is most advantageous .
Concerning the element silicon, it should, at least in accordance here-
with, be maintained at low levels. There is no significant reason why it should
exceed about 30 or 32%; however, by keeping it below this level important commer-
cial advantages are derived. For example, if in the production of ductile iron large
quantities of silicon had to be added in the treatment and innoculation steps,
then the silicon content of the ferrous melt would have to be kept low. However,
this runs counter to desired commercial practice since it demands more careful
selection of a base melt charge and would ostensibly occasion undue refractory
wear. Too, excessive silicon detracts from a friable addition agent and tends
to promote an unwanted amount of fines during crushing. These fines must be
remelted. Thus, product yield is reduced and cost increased. A silicon range
of 19 or 20% to 25 or 27% is quite satisfactory. Contributing to these desiderata
is a low silicon (added) to magnesium (recovered) ratio. This ratio preferably
does not exceed about 12: 1 and is desirably less than about 10: 1.
The following data will help serve as illustrative of the instant
invention.
A number of addition alloys, both within and without the invention,
were prepared as follows: electrolytic nickel, ferromanganese (when used)
and iron were melted together (about 3% carbon was added in Alloy 1 to lower
the melting temperature) and small amounts of FeSi were added as required
to keep the baths deoxidized and quiet. To assure complete solution of the
elements the melts were heated to about 2850F. and then cooled to about 2600F.
whereupon the remainder of the silicon was added.
The baths were then cooled to a temperature (about 2350F . ) near the
freezing point and the magnesium was added either in the form of pure magnesium
sticks or 50 Ni-50 Mg master alloy. The alloys were cast into small truncated
--3--
~036844
- cone-shaped pig molds (1, 2 and 5 lb. sizes) and subsequently crushed to
provide generally uniform equiaxed shaped pieces of roughly 1/2" to 1/4" in
diameter.
Often overlooked is the magnesium recovery in simply producing the
addition agent, the emphasis usually being accorded to magnesium retention
in the final ferrous base melt to be produced. However, this is an important
adjunct to cost and therefore "magnesium addition agent recovery" was deter-
mined for most instances (Table I).
Addition alloy #1, Table I, was added to a molten ferrous base nominally
of 3.4% C, 2%Si, .45%Mn, Bal. Fe, prepared using pig iron,commercial iron,
ferromanganese and ferrosilicon, and heated to 2800F. The above-described
magnesium addition alloy (enough to provide approximately 0.05% Mg to the bath)
was placed in a cavity in the bottom of a specially lined treatment ladle
(100 lb. melts treated except Alloy 1 which was 30 lbs.) and the cavity was
covered in the case of Alloy 1 with a 1/8" thick steel plate wéighing 0.6 lb.
A cover of crushed FeSi (50-50) equal to about 1% by weight of the bath was
used for the remaining addition agents. The iron melt was tapped into the
ladle at 2800F. On the basis that smoke emissions were relatively proportional
to flare, which in retrospect is seemingly reasonably true for very high
nickel (93-95%) addition agents, a camera was used to photographically judge
the amount of smoke emission. In this instance, during tapping into the ladle
a still camera was opened and when all visible sign of reaction ceased, the
shutter was closed. After the reaction, a chill slug was poured for chemical
analysis of magnesium. The bath was poured into a second ladle and innoculated
with 0.5%Si as standard foundry grade ferrosilicon containing about 85%Si. A
second chill slug was analyzed for magnesium.
~ecause it was thought that the flare test was not sufficiently accurate
for smoke generation in the more reactive agents, a different smoke test was
devised for the remaining alloys (and also a repeat of Alloy #1). A Hi-
Volume Air Sampler was used to sample a portion of the smoke drawn off through
an exhaust vent. The exhaust hood was placed to encompass practically
~03684~ l
all the generated smoke, the exhaust being sampled at a distance 20 feet from
the ladle. A fraction of the air and likely a similar proportion of the MgO smoke
was drawn through the sampler by a small fan. The smoke was collected on a
filter which was weighed both before and after test. The weight gain was taken
as the measurement of smoke emitted. Since clogging of the filter occurred in
the more reactive alloys, a correction factor,S2 = Sl(2 Fo/Fo + Fl), was used
to compensate for the drop in air flow rate through the sampler. S2 represents
the corrected weight gain, Sl the measured weight gain, and Fo and Fl, respec-
tively, correspond to the air flow from the smoke tester before aEld after test.
Various addition agent compositions are given in Table I below
together with the percent magnesium recovery in preparing the same. It
will be noted that a high magnesium recovery was obtained in most instances.
However, magnesium recovery was low in respect of Alloys A, B, and C
(alloys outside the invention) due largely, it is believed, to low nickel levels.
TABLE I
% Recovery
Alloy Ni Si Mg FeOther Magnesium
21.530.2 4 31.611.8Mn 93
2 22.524.6 3.35 bal . -- 67
3 21.727.0 3.81 bal . -- 76
4 23.829.2 3.4 bal. -- 68
18.224.3 3.66 bal . -- 61
6 22.121.6 3.64 bal. -- 72
7 22.130.8 4.50 bal . -- 57
8 24.416.2 3.70 bal. -- 62
9 21.617.4 4.41 bal. -- 55
24.223.0 6.41 bal . --
11 24.523.4 6.24 bal . -- 78
12 24.330.1 5.10 bal . -- 85
13 20.920.2 3.04 bal.9.64Mn 63
14 19.729.9 3.85 bal . 9.60Mn 80
A 10.825.6 1.21 bal. -- 20
B 10.930.6 0.61 bal.9.27Mn 13
C 14.926.3 2.11 bal . -- 35
D 94 -- 4.5 bal.1.5C --
E -- 40 5.5 bal . -- --
-
1036844
The data obtained using the above agents are reported in Table II,
the ductile iron base melt nominally having contained about 3.6% C, 1.7% Si,
0.4% Mn, balance iron and impurities .
TABLE II
% Mg Weight
Alloy Ni Si MgRecoveredSi/MgGain, S2
22 30 4 71 10.650.41
2 23 25 3.35 62 12.0 0.185
3 22 27 3.862 11.4 0.19
4 24 29 3.470 12.2 0.20
18 24 3.66 48 13.7 0.19
6 22 22 3.660 10.0 0.23
7 22 31 3.570 9.9 0.29
8 24 16 3.770 6.2 0.27
9 22 17 4.466 5.9 0.31
24 23 6.470 5.1 0.38
11 25 23 6.24 70 5.3 0.43
12 24 30 5.160 9.8 0.37
13 ---- _ _ _ _
20 14 ---- ---- ----
A 11 26 1.272 29.9 0.09
C 15 26 2.166 18.6 0.14
D 94 -- 4.595 0 0.10
E -- 40 5.544 16.5 1.00
.
NOTE: The percentages of Ni, Si and Mg are rounded off in Table II.
Concerning the alloys beyond the invention, Alloys A, C, D and E, either
the S2 factor, the Si/Mg ratio or cost left something to be desired. As to the
alloys within the invention, a number of them had a desirably low smoke emissionfactor, S2, of not more than 0.4, a magnesium recovery of 60% or more, and a
Si/Mg ratio below about 12. The photograph flare test for Alloy #l indicated
the alloy to be less reactive than a number of commercially available additives,although it was difficult to quantitatively determine the result. It is considered
that a higher nickel content would have proven beneficial for alloys such as
Alloy #5 (Mg recovery 48%) as evident from Alloy #6 (Mg recovery 60%) . Loweringthe magnesium level of Alloy #5 would also have likely been helpful. An additionagent containing 19.5% to 23.5% nickel, 20 to 27% silicon, 4 to 5.5% magnesium
is deemed particularly beneficial.
-6
1036844
As will be understood by those skilled in the art, in referring to the
iron content as constituting the ~balance~ or "initially the balance~ of the addition
agents contemplated herein other constituents can be present in amounts which
do not adversely affect the basic characteristics of the additives. In this connec-
tion, elements such as calcium, cerium, rare earth metals, carbon, cobalt, etc.,
can be present though they need not exceed up to 1% calcium, up to 1% cerium,
up to lQ6 of other rare earths, up to 1% carbon, up to 2% cobalt, etc. Copper, if
any, preferably does not exceed about 1% or 2%, since it is a pearlite stabilizer
and can deleteriously affect graphite shape.
Although the invention has been described in connection with preferred
embodiments, modifications may be resorted to without departing from the spirit
and scope of the invention, as those skilled in the art will readily understand.
Such are considered within the purview and scope of the invention and appended
claims .
--7--
