Solution and Aging of MAR-M246 Nickel-Based
Superalloy

Renato Baldan, Antonio Augusto Araújo Pintoda Silva, Carlos Angelo Nunes, Antonio Augusto Couto,
Sinara Borborema Gabriel, and Luciano Braga Alkmin

(Submitted June 1, 2016; in revised form September 23, 2016; published online December 9, 2016)

Solution and aging heat-treatments play a key role for the application of the superalloys. The aim of this
work is to evaluate the microstructure of the MAR-M246 nickel-based superalloy solutioned at 1200 and
1250 �C for 330 min and aged at 780, 880 and 980 �C for 5, 20 and 80 h. The c¢ solvus, solidus and liquidus
temperatures were calculated with the aid of the JMatPro software (Ni database). The as-cast and heat-
treated samples were characterized by SEM/EDS and SEM-FEG. The c¢ size precipitated in the aged
samples was measured and compared with JMatPro simulations. The results have shown that the sample
solutioned at 1250 �C for 330 min showed a very homogeneous c matrix with carbides and cubic c¢
precipitates uniformly distributed. The mean c¢ size of aged samples at 780 and 880 �C for 5, 20 and 80 h
did not present significant differences when compared to the solutioned sample. However, a significant
increasing in the c¢ particles was observed at 980 �C, evidenced by the large mean size of these particles
after 80 h of aging heat-treatment.

Keywords aging, heat-treatment, MAR-M246, microstructure,
solution, superalloy

1. Introduction

The MAR-M246 is a polycrystalline nickel-cobalt-tungsten
superalloy developed by Martin Marietta Corporation, which
became part of Lockheed Martin Inc. (USA) in 1995. This
superalloy has been widely employed in rotors of automotive
turbochargers and in high-pressure turbo pumps (Ref 1). The
density of MAR-M246 superalloy is 8.44 g/cm3 and the
melting range is between 1315 and 1345 �C (Ref 2). The
MAR-M246 superalloy generally solidifies with a dendritic
structure containing carbides and exhibits extensive interden-
dritic segregation. A variant alloy, called MAR-M246 (Hf),
contains 1.5 wt.% hafnium, which is added as a carbide former
and grain boundary strengthener. The as-cast microstructure

contains carbides and the gamma (c)-gamma prime (c¢),
eutectic, as well as the nickel-based matrix which is strength-
ened by solid-state precipitated c¢ phase (Ref 3).

Solution heat-treatments are commonly used for nickel-
based superalloys and play a key role for the application of
these materials at high temperatures (Ref 4, 5). The purpose of
this heat-treatment is to solutionize the c¢ phase in the c matrix
and eliminate segregation generated due to non-equilibrium
solidification (Ref 6). Then, aging heat-treatments are applied
to provide higher creep resistance by generating very fine
secondary c¢ particles with suitable size and distribution in the
c matrix (Ref 7-11). To achieve the best results in terms of
mechanical properties, the temperature and duration of these
heat-treatments must be carefully controlled.

The creep-rupture strength of superalloys is one of the
most significant properties to consider during the material
selection process. The high-temperature (760-1200 �C) creep-
rupture characteristics of nickel-based superalloys are supe-
rior to those of either cobalt or nickel-iron-based superalloys.
Cast nickel-based superalloys such as MAR-M246 main-
tain high stress-rupture strengths at elevated temperatures.
(Ref 12, 13).

The aim of this work is to present the microstructural
evolution of MAR-M246 superalloy submitted to different
solutioning and aging heat-treatments conditions, correlating
and comparing the results with simulations from JMatPro
software.

2. Experimental Procedure

The nominal composition of the MAR-M246 nickel-based
superalloy employed in this work is (wt.%): 10 Co; 10 W; 9 Cr;
5.5 Al; 2.5 Mo; 1.5 Ti; 1.5 Ta; 0.15C; 0.05 Zr; 0.015 B; Ni
balance. Samples were extracted from the central part of turbo
charger rotors via electro-discharge machining for the heat-
treatment experiments. These samples were encapsulated in

RenatoBaldan, São Paulo StateUniversity (Unesp), Campus of Itapeva,
Rua Geraldo Alckmin 519, Itapeva, São Paulo, 18409-010,
Brazil; Antonio Augusto Araújo Pinto da Silva, Instituto de
Engenharia Mecânica - IEM, Universidade Federal de Itajubá -
UNIFEI, Avenida BPS 1303, Itajubá, MG 37500-903, Brazil;
Carlos Angelo Nunes, Escola de Engenharia de Lorena (EEL),
Departamento de Engenharia de Materiais (DEMAR), Universidade de
São Paulo (USP), Estrada Municipal do Campinho, Lorena, São Paulo
12602-810, Brazil; Antonio Augusto Couto, Instituto de Pesquisas
Energéticas e Nucleares (IPEN), Av. Lineu Prestes, 2242, Cidade Universi-
tária, 05508-000 São Paulo, Brazil; Sinara Borborema Gabriel,
Faculdade de Tecnologia de Resende, UERJ – Universidade do Estado
do Rio de Janeiro, Rodovia Presidente Dutra km 298 (sentido RJ-SP) -
Pólo Industrial, 27537-000Resende,Rio de Janeiro, Brazil; andLuciano
Braga Alkmin, Campus Angra dos Reis, CEFET-RJ - Centro Federal de
Educação Tecnológica Celso Suckow da Fonseca, Rua do Areal, 522,
23953-030 Angra dos Reis, Rio de Janeiro, Brazil. Contact e-mail:
renatobaldan@gmail.com.

JMEPEG (2017) 26:465–471 �ASM International
DOI: 10.1007/s11665-016-2462-0 1059-9495/$19.00

Journal of Materials Engineering and Performance Volume 26(2) February 2017—465

http://crossmark.crossref.org/dialog/?doi=10.1007/s11665-016-2462-0&domain=pdf
http://crossmark.crossref.org/dialog/?doi=10.1007/s11665-016-2462-0&domain=pdf


quartz tubes under argon atmosphere (min. 99.995%) and then
heat-treated in a tubular furnace followed by air cooling. The
solution heat-treatments were performed at 1200 and 1250 �C
for 5.5 h while the aging heat-treatments were performed at
780, 880 and 980 �C for 5, 20 and 80 h. A K-type thermo-
couple was placed near the sample to determine accurately the
true temperature of the materials. The heat-treatment conditions
were chosen based on data of liquidus, solidus and c¢ solvus
temperatures determined from simulations using the JMatPro
software (Ni database).

For microstructural characterization, the samples were
prepared following conventional metallographic techniques
and etched with an aqueous solution of 1% citric and 1%
ammonium persulfate. The microstructural analysis was per-
formed with the aid of a conventional scanning electron
microscope (SEM) Hitachi TM 3000 with OXFORD Swift
ED3000EDS detector and a field emission gun scanning
electron microscope (FEG-SEM) JEOL JSM-6701F. The

hardness values of all samples were determined using a
Buehler Micromet equipment applying a load of 500 gf for
30 s.

The c¢ mean diameters of aged samples were measured
directly from the micrographs (approximately 2000 particles
randomly distributed) with the aid of ImageJ software. The
shape of each particle was approximated to a square and an
equivalent diameter was calculated by the relationship A = pd2/
4, where A is the area of the square and d is the equivalent
diameter of the particle.

3. Results and Discussions

3.1 Thermodynamic Simulations with JMatPro

Figure 1 shows the simulation of phase amount (wt.%)
between 600 and 1400 �C for the MAR-M246 superalloy. It

Fig. 1 Simulation of phase amount (wt.%) between 600 and 1400 �C for the MAR-M246 superalloy

Fig. 2 Simulation of amount of minor phases (wt.%) between 600 and 1400 �C for the MAR-M246 superalloy

466—Volume 26(2) February 2017 Journal of Materials Engineering and Performance



can be seen that c phase and MC carbides precipitate almost
simultaneously near 1360 �C, c being the primary phase. The
liquidus, solidus and c¢ solvus temperatures are 1361, 1317 and
1181 �C, respectively. Based on this, on equilibrium conditions
the temperature range for solution heat-treatment of the MAR-
M246 superalloy is between 1181 and 1317 �C. It should be
pointed out that in this range the alloy is not c single-phase,
being also composed of MC carbide and M3B2/MB2 boride
phases. Figure 2 shows the amount of minor phases from
JMatPro simulations for the same temperature range. For the

solution temperature range (between 1181 and 1317 �C), it can
be seen that the MC carbide is stable and does not decompose
while the MB2 boride decomposes and forms the M3B2 boride
near 1230 �C. In addition, other phases (M23C6 and M6C
carbides, M3B2 boride and l phase) may be present in the lower
temperature region of aging treatment (between 780 and
980 �C). Table 1 shows the simulation (JMatPro software) of
expected amount of phases at 780, 880 and 980 �C. The
amount of c¢ and M23C6 phases increases and that of M6C
carbide decreases as the aging temperature is reduced. Finally,
the c¢ fraction for MAR-M246 superalloy decreases from
64.6 wt.% at 780 �C to 59.1 wt.% at 880 �C and 50.8 wt.% at
980 �C.

3.2 Microstructural Characterization of the As-Cast Material

Figure 3(a) shows the as-cast microstructure of MAR-M246
nickel-based superalloy. It can be seen a segregated dendritic
microstructure and MC carbides in the interdendritic regions,
with different sizes and morphologies. The mean value of
primary dendrite arm spacing (3 measurements) is about
68 lm. As expected, c¢ precipitates formed via a solid-state
reaction during cooling were observed in the primary c
dendrites, as depicted in Fig. 3(b), associated with a decrease
in aluminum and titanium solubility in c at lower temperatures.
Standardless EDS analysis of the MC carbide phase (13
measurements) identified the main metallic elements present as
follow (in wt.%): 27.4± 4.3 Ti, 8.6± 0.8 Cr, 8.3± 1.6 Mo,
10.9± 1.9 Ta, 12.7± 2.3 W (C removed from the quantifica-
tion). Other elements are Ni (24.0± 2.5), Al (3.4± 0.4) and Co
(4.7± 0.5) which were detected likely due to matrix effect on
the EDS measurement. The EDS x-ray mapping from the
carbide phase is shown in Fig. 4. Despite the presence of
borides from the JMatPro simulations results, they were not
detected in the samples, possibly because of their very small
amount and size.

3.3 Microstructural Characterization of Solutioned Samples

It was possible to note an important effect of a 50 �C
temperature increase in the microstructures, by comparing the
samples solutioned at 1200 �C (Fig. 5a) and 1250 �C for
330 min (Fig. 5b). Evidences of segregation in the dendritic
regions are still observed after heat-treated at 1200 �C for
330 min. However, the microstructure of the alloy heat-treated
at 1250 �C for 330 min shows a homogeneous c matrix with
carbides.

Figure 6(a) and (b) shows the c¢ phase in the samples heat-
treated at 1200 and 1250 �C for 330 min, respectively, noting
some c¢ clusters in the sample heat-treated at 1200 �C. At the
solution heat-treatment temperature, the primary c¢ particles
from the as-cast microstructure begins to break down and
dissolves gradually into smaller particles known as secondary
c¢ up to the complete c¢ phase dissolution, leading to an

Table 1 Amount of phases for MAR-M246 nickel-based superalloy at 780, 880 and 980 �C

Temperature, �C

Phases, wt.%

c¢ M3B2 M23C6 M6C l

980 50.8 0.2 1.0 5.0 0
880 59.1 0.2 1.7 3.5 0
780 64.6 0.2 3.0 0 4.2

Fig. 3 (a) Microstructure of the as-cast MAR-M246 nickel-based
superalloy. (b) Microstructure of the as-cast MAR-M246 nickel-
based superalloy

Journal of Materials Engineering and Performance Volume 26(2) February 2017—467



Fig. 4 EDS mapping of the carbide on the microstructure of the as-cast MAR-M246 superalloy

Fig. 5 Microstructure of the samples heat-treated at: (a) 1200 �C
for 330 min (b) 1250 �C for 330 min

Fig. 6 c¢ phase in the microstructure of the sample heat-treated at:
(a) 1200 �C for 330 min (b) 1250 �C for 330 min

468—Volume 26(2) February 2017 Journal of Materials Engineering and Performance



homogeneous material composed by c (matrix), MC carbides
and possibly borides (according to JMatPro simulations). The
c¢ precipitates observed in Fig. 6 were formed via solid-state

reaction during air cooling of the samples. When the sample is
air-cooled from the solutioning temperature, the solubility of
aluminum and titanium in nickel decreases and causes precip-

Journal of Materials Engineering and Performance Volume 26(2) February 2017—469



itation of fine c¢ particles. The sample solutioned at 1250 �C for
330 min presented cubic c¢ precipitates uniformly distributed
with approximately 110± 20 nm in size.

3.4 Microstructural Characterization of Aged Samples

The equilibrium volume fraction of c¢ precipitated during
aging is dictated by thermodynamics. Figure 7(a) shows higher
magnification c¢ precipitates in the sample solutioned at
1250 �C for 330 min while Fig. 7 (b-j) shows micrographs

from samples aged at different conditions after solution at
1250 �C for 330 min. The c¢ precipitates present in all aged
samples are uniformly distributed in the c matrix.

The morphology of the precipitates depends on the value of
the c/c¢ mismatch at the aging temperature. Table 2 shows the
c/c¢ mismatch calculated with JMatPro software for MAR-
M246 superalloy at the aging temperatures. A near zero
mismatch will induce a more round shape and a mismatch
(negative or positive) different of zero will lead to a more
cuboidal shape. Higher is the mismatch amplitude, more
cuboidal is the precipitate shape.

It can be seen that the morphology and size of the
precipitates from the aging at 780 �C are almost comparable
with the as-solutioned sample, whatever the duration of the
aging treatment. The aging treatments at 780 �C have no effect
on the morphology and size of the precipitates as the
temperature is too low to induce significant evolution of the
microstructure. However, the aging treatments at 880 �C and
980 �C, in spite of higher lattice mismatch, clearly induce
changes in size and/or morphology of the precipitates, as these
temperatures are sufficiently high to activate significant coars-
ening (Ref 14).

Figure 8 shows both simulated (JMatPro software) and
minimum/maximum measured values of c¢ mean diameter as a
function of aging conditions (time and temperature). The initial
c¢ particles mean diameter for simulations was 110± 20 nm,

bFig. 7 (a) Micrograph of the MAR-M246 superalloy solutioned at
1250 �C for 330 min. (b) Micrograph of the MAR-M246 superalloy
solutioned at 1250 �C for 330 min and aged at 780 �C for 5 h. (c)
Micrograph of the MAR-M246 superalloy solutioned at 1250 �C for
330 min and aged at 780 �C for 20 h. (d) Micrograph of the MAR-
M246 superalloy solutioned at 1250 �C for 330 min and aged at
780 �C for 80 h. (e) Micrograph of the MAR-M246 superalloy solu-
tioned at 1250 �C for 330 min and aged at 880 �C for 5 h. (f)
Micrograph of the MAR-M246 superalloy solutioned at 1250 �C for
330 min and aged at 880 �C for 20 h. (g) Micrograph of the MAR-
M246 superalloy solutioned at 1250 �C for 330 min and aged at
880 �C for 80 h. (h) Micrograph of the MAR-M246 superalloy solu-
tioned at 1250 �C for 330 min and aged at 980 �C for 5 h. (i)
Micrograph of the MAR-M246 superalloy solutioned at 1250 �C for
330 min and aged at 980 �C for 20 h. (j) Micrograph of the MAR-
M246 superalloy solutioned at 1250 �C for 330 min and aged at
980 �C for 80 h

Fig. 8 Simulated (JMatPro software) and minimum/maximum mea-
sured values of the c¢ mean diameter as a function of aging condi-
tions (time and temperature)

Fig. 9 Comparison between c¢ mean diameter and hardness values
for the aged samples

Table 2 c/c¢ mismatch for MAR-M246 superalloy at the
aging temperatures

Aging temperature, �C Lattice Mismatch, %

780 �0.145
880 �0.258
980 �0.278

Table 3 Hardness of MAR-M246 superalloy after differ-
ent aging heat-treatments

Temperature, �C Time, h Hardness, HV

780 5 465± 17
20 488± 21
80 481±14

880 5 446± 18
20 439± 10
80 444± 14

980 5 431±10
20 400± 12
80 399± 16

470—Volume 26(2) February 2017 Journal of Materials Engineering and Performance



which corresponds to the measured mean diameter in solu-
tioned sample (Fig. 6b). For 780 and 880 �C, the simulated
values are between the minimum and the maximum measured
values. Additionally, samples heat-treated at 780 and 880 �C
for 5, 20 and 80 h have not shown significant differences in the
mean c¢ size when compared to the solutioned sample, in
agreement with the low c¢ coarsening rates calculated with the
aid of JMatPro software at these temperatures (14.2 nm/h1/3 at
780 �C and 37.7 nm/h1/3 at 880 �C). However, at 980 �C the
simulated values were slightly higher than those measured for
20 and 80 h. A significant increase in the c¢ coarsening rate is
observed at 980 �C, evidenced by the large mean size of these
particles after 80 h of aging heat-treatment (Fig. 7j). The c¢
coarsening rate at 980 �C is 94.2 8 nm/h1/3, 6.6 and 2.5 times
higher than those at 780 and 880 �C, respectively.

3.5 Hardness Measurements

Table 3 shows the hardness values of MAR-M246 superal-
loy after different aging heat-treatments. The hardness of as-
cast and solution heat-treated (1250 �C for 330 min) was
381±12 and 415± 20 HV, respectively. It shows that the
hardness did not vary significantly between the as-cast and
heat-treated samples. Figure 9 shows the comparison between
the c¢ mean diameter and the hardness values for the aged
samples. It can be seen a tendency of lower hardness values for
higher aging temperature, which should be mainly related to
higher c¢ particle sizes at higher temperature.

4. Conclusions

The results of this investigation concerning the MAR-M246
nickel-based superalloy have led to the following data:

• the c¢ solvus, solidus and liquidus temperatures calculated
with JMatPro software were 1181, 1317 and 1361 �C,
respectively;

• the microstructure of the as-cast material presented a seg-
regated dendritic microstructure and MC carbides with
high aspect ratio and composed mainly by Ti, Cr, Mo, Ta,
W;

• the microstructure of the alloy heat-treated at 1250 �C at
330 min showed a very homogeneous c matrix with car-
bides and cubic c¢ precipitates uniformly distributed with
approximately 110± 20 nm in size;

• the mean c¢ size of aged samples at 780 and 880 �C for
5, 20 and 80 h did not present significant differences

when compared to the solutioned sample. However, a sig-
nificant increasing in the c¢ particles was observed at
980 �C, evidenced by the large mean size of these parti-
cles after 80 h of aging heat-treatment.

• The hardness of as-cast and solution heat-treated (1250 �C
for 330 min) was 381±12 and 415± 20 HV, respectively.
It can be seen a tendency of lower hardness values for
higher aging temperature, which should be mainly related
to higher c¢ particle sizes at higher temperature.

References

1. Allen F. Denzine, Thomas A. Kolakowski, and J.F. Wallace, Method of
Improving Fatigue Life of Cast Nickel Based Superalloys and
Composition. Patent US4078951-A. 1978

2. J.R. Davis, Ed., ASM Specialty Handbook, Heat-Resistant Materials,
ASM International, Materials Park, 1997

3. M.H. Johnston, P.A. Curreri, R.A. Parr, and W.S. Alter, Superalloy
Microstructural Variations Induced by Gravity Level During Direc-
tional Solidification, Metall. Trans. A, 1985, 16A, p 1683–1687

4. R.C. Kramb, M.M. Antony, and S.L. Semiatin, Homogenization of a
Nickel-Base Superalloy Ingot Material, Scripta Mater., 2006, 54, p
1645–1649

5. K.L. Zeisler-Mashl, and B.J. Pletka, Segregation During Solidification
in the MAR-M247 System. International Symposion on Superalloys,
Pennslyvania, 1992, p. 175–184

6. S.R. Hegde, R.M. Kearsey, and J.C. Beddoes, Designing Homoge-
nization-Solution Heat Treatments for Single Crystal Superalloys, Mat
Sci and Eng A, 2010, 527, p 5528–5538

7. I.M. Wolff, Precipitation Accompanying Overheating in Nickel-Base
Superalloy, Mater. Charact., 1992, 29, p 55–61

8. A. Plati, Modelling of c¢ Precipitation in Superalloys. 2003. Disser-
tation (Master of Science). Department of Materials Science and
Metallurgy, University of Cambridge, Cambridge, 2003

9. I.S. Kim, B.G. Choi, S.M. Seo, D.H. Kim, and C.Y. Jo, Influence of
Heat Treatment on Microstructure and Tensile Properties of Conven-
tionally Cast and Directionally Solidified Superalloy CM247LC,
Mater. Lett., 2008, 62, p 1110–1113

10. S.A. Sajjadi, S.M. Zebarjad, R.I.L. Guthrie, and M. Isac, Microstruc-
ture Evolution of High-Performance Ni-base Superalloy GTD-111 with
Heat Treatment Parameters, J. Mater. Process. Technol., 2006, 175, p
376–381

11. C.M. Kuo, Y.T. Yang, H.Y. Bor, C.N. Wei, and C.C. Tai, Aging Effects
on the Microstructure and Creep Behavior of Inconel 718 Superalloy,
Mater. Sci. Eng. A, 2009, 510, p 289–294

12. W.F. Smith, Structure and Properties of Engineering Alloys, Mc-Graw
Hill, New York, 1981

13. M. Bauccio, Ed., ASM Metals Reference Book, 3rd ed., ASM
International, Materials Park, 1993

14. P. Caron and T. Khan, Improvement of Creep Strength in a Nickel-Base
Single-Crystal Superalloy by Heat Treatment, Mater. Sci. Eng., 1983,
61, p 173–184

Journal of Materials Engineering and Performance Volume 26(2) February 2017—471


	Solution and Aging of MAR-M246 Nickel-Based Superalloy
	Abstract
	Introduction
	Experimental Procedure
	Results and Discussions
	Thermodynamic Simulations with JMatPro
	Microstructural Characterization of the As-Cast Material
	Microstructural Characterization of Solutioned Samples
	Microstructural Characterization of Aged Samples
	Hardness Measurements

	Conclusions
	References