Significance of Alloying Element Levels in Realizing the Specified Tensile Properties in 18 Wt % Nickel Maraging Steel

doi:10.4236/msa.2011.28150

Paper Menu >>

Journal Menu >>

Materials Sciences and Applicat ion, 2011, 2, 1116-1120

doi:10.4236/msa.2011.28150 Published Onl ine August 2011 (http://www.SciRP.org/journal/msa)

Significance of Alloying Element Levels in

Realizing the Specified Tensile Properties in 18

wt% Nickel Maraging Steel

Muktinutalapati Nageswar a Rao 1*, Krishnan Sivasubraman i an2

1Andhra University, Visakhapatnam, Indian Institute of Technology Bombay, University of Pennsylvania, Philadelphia, School of

Mechanical & Building Sciences VIT University, Tamil Nadu, India; 2PSG College of Technology, Coimbatore, Mishra Dhatu Ni-

gam Limited, PO Kanchanbagh, Hyderabad, India.

Ema il: *m[email protected]

Received J anuary 14th, 2011; rev ised May 19th, 2011; ac c epted May 26th, 2011 .

ABS TRACT

Among the various grades of commercially available 18 wt% nickel maraging steels, the one with nominal 0.2% proof

st re ngt h i n the rang e 1700 - 1750 MPa is the most commonly used and is distinguished by an excellent combination of

high strength and high fracture toughness. The main alloying elements are nickel, cobalt, molybdenum and titanium.

The first t hree of the se are present at relatively high conce ntrations i n t he chemical composition. The high c ost of these

metals leads to a high cost of production and thi s become s a deterrent to extensive use of the steel. In the pre sent study,

an attempt was made t o produce the steel by pegging the l evels of these al loy ing eleme nts in t he lower half of the speci-

fied range. The objective was to save on the raw material cost, while still conforming to the specification. The steel so

produced coul d not, however, attain the specified tensile properties af te r fi nal heat treatment. The observed behavi or is

explained based on the role played by the different all oying elements in driving the pre cipitation hardening reaction.

Keywords: 18% Ni Maraging Steel, Chemical Composition, All o ying Ele ments, Age Hardening, Tensile Proper ties

1. Introduction

18 wt% Ni mar aging steels based on iron -ni ckel mar ten-

site constitute a very important family of high strength

steels. They distinguish themselves by demonstrating an

unparalleled combination of high strength and high frac-

tur e t oughn es s in hea t tr eated condi ti on and excel lent h ot

and cold workability and weldability. Because of their

high str en gth to weight ratio, th ey find ext ensi ve applica-

tion in aerospace sector. Different grades of maraging

steel are commercially available, covering the strength

range of 1400 - 2400 MPa. With increasing strength level,

the tensile ductility and fracture toughn ess decrease. Ac-

cor dingly the lower stren gth variants are used wher e hi gh

ductility and fracture toughness ar e important for design.

The higher strength variants are used where high strength

is of pa ra moun t im por tan ce for de sign an d on e can ma n-

age with moderate levels of fracture toughness. Titanium

is used as the primary strengthening element in these

steels; pr ecipitation of titanium bearing intermetallic par-

ticles in martensitic matrix in a uniform and finely dis-

persed mann er during aging leads t o development of ver y

high strength levels. Aging has to be optimally carried

out to realize the maximum strengthening effect. The

1988 symposium [1] deliberated on important develop-

ments and applications of maraging steels and more re-

cently the metallurgy of 18% Ni maraging steels has

been r eviewed by Rao [2].

An im portant factor that has come in the way of exten-

sive use of 18% Ni maraging steels is their high cost.

Cobal t is an expen sive alloying elem ent an d is present in

the range of 8 to 12 wt% in these steels, contributing

importantly to the cost. Accordingly efforts have been

made t o de vel op cobalt -fr ee maraging st eels with a com-

parable strength-fracture toughness combination. Co-

balt-free maraging steel grades with strength levels cov-

ering the ran ge 1400 - 2000 MPa are now commercially

available.

In addition to cobalt, the composition of 18% Ni ma-

raging steel prominently includes nickel and molybde-

num; these elements are also costly and present in sub-

stantial quantity in the steel. Thus they contribute signif-

icantly to the pr oduction cost of the steel.

Significance of Alloying Element Levels in Reali zing the Specified Tensi le Properties in 18 wt % N ickel Maraging St eel

1117

With the levels of alloying elements in the middle /

upper half of the specified range, material could be pro-

duced meeting the specified properties. An effort was

made to contain the production cost of bar material by

processing a batch of steel with the levels of the costly

alloying elements - Co, Mo and Ni - in the lower half of

the range allowed by the specification for chemical

composition. The steel so produced has been evaluated

and found to be not meeting the specification for tensile

strength and yield stren gth. Th e paper gives the details of

processing and evaluation of this batch and provides an

explanation for the failure encountered in meeting the

specifi ed tensi le properties.

2. Material

Th er e was a r equir em ent for 33 mm di am et er bar s of 18%

Ni maraging steel grade conforming to Aerospace Mate-

rials Specification (AMS) 6512. The chemical composi-

tion of t he steel as per this specificat ion is given in Table 1.

The tensile properties specified in AMS 6512 for the bar

material are given Table 2. The maraging steel was pro-

duced by double vacuum melting - vacuum induction

melting (VIM) foll owed by vacuum ar c r emelting (VAR).

Scrap of 18 wt% Ni maraging steel was consolidated by

melt ing in el ectri c arc furna ce. The liquid m eta l, at the end

of arc furnace melting, wa s cast into a cylindr ical electrode.

The electrode material was subjected to electroslag re-

melt ing (ESR). The E SR process ed material was then used

as charge for va cuum induc ti on melting. Small am ounts of

pure iron, molybdenum pellets, nickel shots and electro-

lytic cobalt were added to adjust the composition during

va cuum ind uct i on m elting . Stand ar d pr a cti ces for m elting ,

refining and casting wer e followed for VIM pr ocessing. In

order t o expl ore t he possibility for cutti ng down the raw

Table 1. Chemical composition of maraging steel, as speci-

f ied i n AMS 6512.

Element wt%

Ni ckel 17 - 19

Cobalt 7.0 - 8.5

Molybdenu m 4.6 - 5 .2

Titanium

0.3 - 0.5

Aluminum 0.05 - 0.15

Carbon

< 0.03

Iron

Balance

Tab l e 2. Mechanical properties of the bar material as speci-

f ied i n AMS 6512.

Property

Specification

Ultimate tensile stren gth (MPa)

1758 m in

0.2% Proof strength (MP a )

1724 m in

% Elongation

6 min

% Reduction in area

40 m in

material cost and in turn the production cost, while still

conforming to the specification, one batch of the steel

was processed through VIM with levels of the costly

elements - Co, Mo and Ni - ai m ed i n the lower hal f of the

specified range for the respective elements. Detailed

chemi cal analysis of the batch at this stage was carried

out and the r es ul ts ar e given in Table 3. Th e molten met-

al was tapped in to a 480 mm diameter mould. The 480

electr ode was con diti on ed an d r emelt ing was done

in a VAR furnace to pr oduce a 550 mm

ingot.

The VAR ingot was subjected to hot working, com-

prising of hot forging in a press, hot forging in hammer

and finally hot rolling to realize the mater ial in the form

33 mm

bars. Conditioning of the material during hot

working was carried out as necessary. Detailed ch emical

analysis of the material was again carried out at the bar

stage; the results were found to match with those ob-

tained at VIM stage, except for a small drop in titanium

l e ve l from 0.43 to 0.41 a nd alumi num level from 0.095 to

0.090.

3. Re s ults

The 33 mmφ hot rolled bar material was taken up for

heat treatment. The treatment comprised of two stages -

(1) Soaking at 950℃ foll owed by for ced air cool in g and

(2) Soaking at 820℃ followed by air cooling. Tensile

properties were evaluated after carrying out aging of the

mat er ial . Th e tem perat ur e us ed for aging was 485℃. The

a gin g ti m e n or m all y a do pt ed i s 3 h our s, but th e spe ci fi ca -

tion allows aging time up to 6 hours. Aging for

hours resulted in the UTS and 0.2% PS values not meet-

ing the specification. Aging was then continued and ten-

si l e pr oper ti es eval ua t ed aft er c um ula ti ve aging ti m es of 6,

12 and 15 hours. There was steady improvement in both

0.2% PS and UTS values with increasing aging time, but

even after 15 hours the proof strength values were not all

above the specified minimum. Table 4 gives the details.

On the other hand, heats made with standard practice,

where the levels of individual alloying elements are

maintained in the middle/upper half of the specified

range, show ed acceptable 0.2% PS and UT S values after

Table 3. Chemical composition of the batch produced with

l ow le vels of all oying elem ents, aft er VIM (wt%).

Element

Wt %

Carbon

0.009

Sulfur

0.0016

Phosphor ous

0.006

Ni ckel

17.6

Cobalt

7.6

Molybdenu m

4.7

Titanium

0.43

Aluminum 0.095

Iron

Balance

Significance of Alloying Element Levels in Reali zing the Specified Tensi le Properties in 18 wt % Ni ckel Ma ra gi ng S tee

1118

Tab l e 4. M ec han i cal pr ope r ti es obt ai n ed on 33 mm

hot rolle d and heat tr eat ed bars from the batch under study, as a func-

tion of aging time. Solution treatment: 950˚C 1 hour followed by forced air cooling to room temperature; 820˚C 1 h our fol-

l ow ed b y air cooling Aging temperature: 485˚C A ging i s followed by air cool ing.

Aging time (Hours)

0.2% Proof strength

(MPa )

Ultimate tensile stren gth

(MPa )

% Elongation % Reduction in area.

3.5

1606 - 1640

1670 - 1690

12-13

1668 - 1679

1736 - 1740

57 - 61

1709 - 1730

1762 - 1772

11 - 14

57 - 62

1718 - 1742

1775 - 1785

12 - 13

58 - 62

3.5 hours of aging and the values remained essentially

constant until cumulative aging time of 15 hours, for

which data are available. Figure 1 shows variation of

0.2% proof strength and ultimate tensile strength as a

function of aging time on a comparative basis for the

batch under study and heats made as per standard prac-

tice. It is to be emphasized that aging of the samples

from the batch under study has been done under condi-

tions identical to aging of samples from heats made as

per standard practice. The difference in the aging beha-

vior, as shown in Figure 1, is hence indeed due to dif-

ferent response of the material from the batch under

study to aging treatment.

Microstructural examination was carried out on sam-

ples drawn fr om th e bar material after agin g. The micro-

structure comprised of aged martensite. The prior auste-

nitic grain size was fine, 7 to 8 on ASTM scale. Similar

grain size values were obtained on bars from heats

processed as per standard practice. Inclusion rating was

carried out in the unetched condition as per ASTM E45.

Th e r ating of th in oxi des was 0. 5; th e r atin gs for th e sul-

fide, silicate and alumina type inclusions wer e z er o. Car-

bide / carbonitride particles were seen, but their content

was well within acceptable limits. Similar inclusion rat-

ings were obtained in heats made with recycling scrap

forming only a minor part of the total charge weight.

4. D i sc u ssi on

Even though it i s a hea t made of 100% scr ap of m a raging

steel, this is not believed to be responsible for the ob-

served failure in realizing the specified strength level.

Th e high er per cent a ge of scr ap, if it ma de a differ en ce, is

expe cted to get r eflect ed in th e incl usion r atin g. How ever,

the inclusion rating in the batch under study was found to

be similar to the rating observed in heats made with re-

cycling scrap forming only a minor part of the total

charge weight. Further, experience has shown that inclu-

sion rating affects more the percent elongation and per-

cent reduction in area values and for the batch in ques-

tion, measured values for these two attributes wer e com-

fortably above the respective minimum specified values.

It is to be noted tha t the 0. 2% pr oof strength and ulti-

mate tensile strength values as per AMS 6512 are consi-

derabl y h igher tha n those spe cifi ed in other specifications

cover ing this material. Tab l e 5 bring s out this com parison.

What this means is that the bar material under discussion

had to be produced to a relatively high strength level.

The observed grain size in fully heat treated condition

wa s v er y f ine - 7 t o 8 on AST M s cal e. Figure 2 sho ws the

typical microstructur e in fully h eat treated conditi on. Sim-

ila r gra in siz e was obser ved in h eats made as per standard

practice. Hence the Hall-Petch strengthening in the batch

under study is the same as that in the heats made as per

standard practice. In spite of this the measured strength in

the batch under study is f a ll i ng s hort of s pe cif icat ion.

Nickel level in the batch studied (17.6 wt%) lies in the

lower half (17 - 18%) of the AMS 6512 specification.

The Ms a nd Mf temperat ur es are influen ced by the Ni level.

Ni ckel level al s o influen ces the aging temperatur e / time at

which austenite reversion sets in. The composition of the

maraging steel is designed such that (i) martensitic trans-

formation is complete well before the steel cools down to

room temperature and (ii) austenite reversion does not

occur with the time / temperature combinations normally

empl oyed for aging. The micr ostructure of the batch under

study in the solution treated condition was found to con-

tain < 2% ret ai ned a us t enite. T he m i cros tr uc tu r e e ven a f t er

agi ng for 15 h our s at 485 oC ha d a vol ume fr a cti on of a u s-

tenite <2%. The design requirements are thus fully met

even at this slightly lower level of Ni. The lower Ni level

could possibly mean lower activity of Ni in Fe-Ni marten-

site, and a reduced driving force for Ni3(Ti,Mo) pr ecipi ta-

tion to occur, with the consequence of a lowered strength

l evel obta in ed a fter a gi ven agi ng t em pera tur e / ti me com-

bination.

The range for molybdenum in the steel, as per AMS

6512 is 4. 6 to 5. 2. The r ange for Mo for equi valent grade

supplied by Carpenter Steel (Carpenter NiMark Alloy 250)

[3], for exam pl e, is 4. 7 t o 5. 0 wt%. The le vel of Mo in the

batch under study corr esponds t o l ower limit of the range.

This again mean s a reduced driving force for Mo to preci-

pitate out as Ni3(Ti,Mo); the consequence again would be

attainment of a reduced strength level after aging for a

given time / temperat ure combination.

The cobalt level in the steel under investigation (7.6

wt%) is in the lower half of the range specified by AMS

6512. This level appears to be somewhat lower than re-

quired. There is evidence that Co le vel in this gra de is peg-

Significance of Alloying Element Levels in Reali zing the Specified Tensi le Properties in 18 wt % Ni ckel Ma ra gi ng S tee

1119

Table 5. Minimum 0.2% PS and UTS values specified for the maraging steel grade under discussion as per different stan-

dards.

Standard

0.2% PS (MPa)

UTS (MPa)

DIN EN3529 (1999) for Aerospace forgings

1620

1720

MIL-S-46850D (1991) for bars, forgings, sheets, str ips, plates

1655

AS TM A538 (1982) fo r press ure vessel plates (withdrawn in 1987)

1580

1650

AM S 6512E (2005) for bars, f orgings, rings

1724

1758

Figure 1. Variation of 0.2% PS and UTS as a function of aging time for the batch under study (with low levels of alloying

elements) an d for the heats made with standard practice.

Figure 2. Ty pic al microstructure of the maragin g st eel bars in fully heat tre ate d condi tion.

ged at 7.8 to 8. 0 wt% [4-6] . T h e r ol e of C o on the l ower-

ing of solid solubility of Mo in Fe-Ni marten sitic matrix

has been well established [1,7]. With Co present at a

lower level, there will be, to that extent, a r educed effe ct

of lowering of the solid solubility of Mo in martensite

and correspon dingly a reduced precipitation of Mo bear-

ing age-hardening precipitate Ni3 (Ti,Mo) after aging for

a given time / temperature combination.

Titanium level in the steel under study is 0.41 wt%.

There is eviden ce that Ti in this gra de is pe gged a t a level

of 0.45 wt% [4,5]. Values as high as 0.55 wt% Ti have

been adopted by Boehler Edelstahl for their equivalent

grade [6]. Titanium contributes importantly to streng-

thening by precipitating in the form of titanium bearing

1120 Significance of Alloying Element Levels in Realizing the Specified Tensile Properties in 18 wt % Nicke l M araging Steel

par ticl es Ni3(T i,Mo), l eaving unde tecta ble a moun t ( < 0.1

wt%) of Ti in th e matrix [7]. It is hen ce concluded that a

somewhat higher level of Ti would have facilitated

r eaching the specified str en gth l evel . Th e level of alumi-

num (0.09%) is slightly below th e middle of th e specified

range (0.1%). This is expected to have very small effect

on the strength, considering that strengthening effect of

Al is 5.6 - 6.3 kg/sq·mm (55 - 62 MPa) per 0.1 wt% [8].

If the microstructure a ft er a ging contain s con sider able

amount of austenite, this could lead to a relatively low

level of yield strength, as austenite is a soft phase com-

pared to aged martensite. In the present case, however,

austenite is present in the aged micr ostructur es a t a l evel

of < 2%. Hence th is i s not a caus ative factor for the steel

not respo nd ing sa tis fac torily to the aging treatment.

Aging has been continued till 15 hours and tensile

properties evaluated. There is a steady increase in the

strength with aging time; however, even after aging for

15 hours, not all values of 0.2% proof strength met the

speci fi ca ti on. Furth er it i s n ecessa r y i n in dus tr ial pr a ctice

to have a steel composition which will respond to the

aging treatment and lead to attainment of specified me-

chanical properties in a relatively short time, say 3 to 6

hours, from the product ivity point of vie w.

It th us becomes clear that adhering to AMS 6512 with

respect to chemical composition, by itself, is not suffi-

cient to meet th e specification with r espect t o mechan ical

properties. As mentioned in the results section, heats

made with the standard practice, where the levels of in-

dividual alloying elements are maintain ed in th e middle /

upper half of t he specified range show ed acceptable 0.2%

PS an d UTS val ue s a ft er 3. 5 h ours of a gin g. Th e s ol uti on

to consistently producing the material with total confor-

mity to AMS 6512 hence lies in producing the melts with

levels of Ni, Co, Mo and Ti in the middle or even in the

upper h alf of the range specified in the AMS.

5. Conclusions

1) Maintainin g the levels of alloying elements within the

range specified in AMS 6512 for 18 wt% nickel marag-

ing steel is not sufficient to realize the strength levels

specifi ed in the sam e Standar d.

2) I t is believed that the r elatively low levels of Ni, Mo

and Ti within the specified range, tried out for reducing

the cost of production, lead to a relatively low volume

fraction of Ni3(Ti,Mo); this causes lower than the re-

quired precipitation strengthening effect.

3) The relatively low level of Co tried out appears to

be r esul ting in a less decr ease in solid solubi lity of Mo in

the martensitic matr ix, thereby leading to reduced extent

of precipitation of Mo and consequently reduced amount

of preci pitation str ength ening.

4) The standard practice of melting, with particular

emphasis to pegging the levels of Ni, Co, Mo and Ti in

the middle or upper half of the range specified in AMS

6512, has to be adopted if material meeting this specifi-

cation in all respect s is to be produc ed.

6. Acknowledgement s

M Nageswara Rao is grateful to the Management of VIT

Un iversity for their kind consent to publish this paper. K

Sivasubramanian is indebted to the management of

MIDHANI for encouragement and per mission to publish

the results.

REFERENCES

[1] R. K. Wilson, “Maraging Steels: Recent Developments

and Applications,” The Minerals, Metals & Materials So-

ciety, Warrendale, Pennsylvania, 1988.

[2] M. N. Rao, “ Pr ogr es s in Unde r s ta ndi ng t he Me ta ll ur gy of

18% Nickel Maraging Steels,” International Journal of

Mater ials R esear ch, V ol . 97, No. 11, Nove m ber 2006 , pp.

1594-1607.

[3] “NiMark Alloy 250,” Internet Available:

http://www.cartech.com/ssalloysprod.aspx?id

[4] P. P. Sinha, D. Sivakumar, T. Tharian, K. V. Nagarajan

and D. S. Sar ma, “Thermal Embrittlement in 18Ni Cobalt

Free and 18Ni-8Co-5Mo Maraging Steels,” Materials

Science and Technology, Vol. 12, November 1996, pp.

945-954.

[5] P. P. Sinha, K. Sreekumar, N. S. Babu, B. Pant, A. Na ta-

rajan and K. V. Nagarajan, “Development of Heat Treat-

ment Parameters to Improve Fracture Toughness and

Grain Size of an Embrittled Maraging Steel,” Journal of

Heat Treating, Vol. 9, No. 2, 1992, pp. 125-131.

doi:10.1007/BF02833148

[6] P. Wuer zing e r , R. Rabi t s ch R a nd W. Me ye r , “ Pr oduc t i on

of Maraging Steel Grades and the Influence of Specified

and Nonspecified Elements for Special Applications,”

Journal of Materials Science, Vol. 39, No. 24, 2004, pp.

7295-7302. doi:10.1023/B:JMSC.0000048744.03279.ec

[7] W. Sha, A. Cerezo and G. D. W. Smith, “Phase Chemi-

stry and Precipitation Reactions in Maraging Steels: Part

IV, Discussion and Concl usions,” Metallurgical Transac-

tions A, V ol. 24A, No. 6, J une 199 3, pp. 1251 -1256.

doi:10.1007/BF02668193

[8] A. G. Haynes, “The Making, Shaping and Heat Treating

of 18% Ni Maraging Steels,” Proceedings of ASPA Se-

minar, Madras, 1978, p. 38