THERMAL CONDUCTIVITY OF GRAY IRON AND COMPACTED

GRAPHITE IRON USED FOR CYLINDER HEADS

W. L. Guesser, I. Masiero, E. Melleras, C. S. Cabezas - UDESC and Tupy Fundies
Abstract
It is presented an experimental study about thermal conductivity in gray iron and compacted graphite iron.
Thermal conductivity of the specimen is measured by a comparative method with stationary axial heat flow
according to ASTM E 1225. The investigated specimen is put in a stack in tight thermal contact with two
reference materials of same diameter. The upper reference specimen is coupled to a heat source, the lower
reference to a heat sink. A guard heater and other experimental setup minimize radial heat losses. Reference
material is electrolytic iron with certified thermal conductivity. It was tested 2 gray irons, the first alloyed
with CuSnCr and the second with CuSnCrMo. Two grades of compacted graphite iron were also tested, CGI
350 and CGI 450. The tests were conducted up to 400 C.
The results show that thermal conductivity decreases in the following sequence: CuSnCr gray iron,
CuSnCrMo gray iron, CGI 350, CGI 450. The thermal conductivity results of the gray iron samples decrease
with increasing temperatures, and are almost constant for the CGI samples. The results show the potential of
using CGI 350 for applications like engine cylinder heads.
Key words: Thermal conductivity, cast irons, compacted graphite iron, gray iron, cylinder heads.
1) Introduction
Thermal conductivity is, in some applications, the main reason for the material selection, in special in the
automotive industry, for internal combustion engine components and brake systems. Cast irons have been
used for such components (cylinder heads, pistons, brake drums and disks), combining good mechanical and
friction properties with thermal conductivity.
The recent use of compacted graphite iron (CGI) for cylinder heads has demanded the study of properties
of this material, in special the thermal conductivity. In comparison with gray iron, CGI shows lower values of
thermal conductivity, and this is the main reason of the designer objection regarding CGI for such
applications, in spite of its higher mechanical properties. However, one point is the comparison between the
CGI not with the regular gray iron, but with alloyed gray irons (Cu, Sn, Cr, Mo), nowadays used for diesel
cylinder heads. In a general way the alloying elements tend to decrease thermal conductivity, so those alloyed
gray irons must present thermal conductivity values lower than the regular gray irons.
After that, this paper presents a thermal conductivity comparison study between compacted graphite irons
and alloyed gray irons.
2) Literature Review
Thermal conductivity values of metallographic phases of cast irons are presented in Table 1. It can be seen
that ferrite has higher thermal conductivity than pearlite and also that cementite can lower the cast iron
thermal conductivity. Parallel to the graphite basal plane the thermal conductivity is high and, in this
condition, is the phase with highest thermal conductivity. So, a graphite shape that eases the thermal
conductivity along the basal plane must result in maximum thermal conductivity. This is the case of gray iron,
as can be seen in figure 1 (Hasse, 1996).
The amount of graphite also affects thermal conductivity, specially in gray irons, as shown in figure 2
(Silva Neto, 1978). In this figure theoretical models were plotted considering nodular graphite as perfect
spheres and flake graphite as discs. The model for nodular graphite showed good agreement with the
experimental results, while for flake graphite this agreement is poor.
Typical values of thermal conductivity of different gray and ductile irons grades can be seen in tables 2
and 3 for increasing temperatures. For gray irons thermal conductivity decreases with temperature. This trend
is observed in many reports (Angus-1960, BCIRA Broadsheet-1981, Konstruiren+Giessen-2000, Stefanescu2003), although there is no discussion on the cause of this behavior. The effect of the temperature in reducing

thermal conductivity is higher for gray irons with high carbon content (Angus-1960, BCIRA Broadsheet1981).
As for steels, the presence of alloying elements in cast irons decreases thermal conductivity for a given
matrix (it should always be considered that alloying elements can affect the amounts of ferrite and pearlite in
the matrix). In table IV the effects of alloying elements are presented. In this table it can be seen the
significant effect of silicon, which is always present in high amounts in cast irons.
The thermal conductivity of CGI is compared to an unalloyed gray irons in figure 3. It is shown that, in
CGI high nodularity decreases thermal conductivity, while increasing temperatures have little effect in this
property. Additional results are presented in Figure 4, where it can be seen that higher nodularity decreases
thermal conductivity (Monroe &Bates, 1982).
A practical consequence of the differences between gray and compacted graphite irons thermal
conductivity was verified by Cueva et all, establishing the temperature during the restrain of break discs
(figure 5). It is observed that gray iron break discs are better heat conductors than CGI, causing the CGI
casting to reach higher temperatures in service.
In the following experimental work thermal conductivities of two alloyed gray irons used for cylinder heads
for heavy diesel engines and two classes of CGI were determined.
3) Experimental Procedures
Table V shows the chemical composition of the tested materials. Copper and tin are used in CGI to obtain
the necessary amount of pearlite for each grade. In gray iron, copper, tin, chromium and molybdenum are
used as alloying elements to obtain the high strength grades.
Gray iron samples were obtained from bars with 30 mm diameter, and CGI samples were machined from
keel blocks of 25 mm, casted in chemically bonded sand moulds. Those are standards samples for tensile
tests.
Thermal conductivity was measured by a comparative method with stationary axial heat flow according to
ASTM E 1225-99. The investigated specimen is put in a stack in tight thermal contact with two reference
materials of the same diameter. The upper reference specimen is coupled to a heat source, the lower reference
to a heat sink. Radial heat losses are minimized by a guard heater and other experimental setup. Reference
material is electrolytic iron with certified thermal conductivity.
The temperature drop along the specimen TP and the references TR1 , TR2 as well as the distances
between the temperature sensors xP , xR1 and xR2 are measured. With known thermal conductivity as
function of temperature of the reference material R1 and R2 , following equation gives the thermal
conductivity of the specimen P

P =

1 x P
TR1
T
+ R2 R2 .
R1
2 TP
x R1
x R2

(1)

An apparatus of Dynatech Co., Cambridge, MA, USA, Type TCFCM was used for the measurements.
Temperature was measured by Ni-CrNi thermocouples, thermo-voltages were measured by a data acquisition
system Philips, Type PM 8237A, which automatically references to ice point and linearizes the signal. A
traveling microscope measured the distances of the drillings of the specimen, where the thermocouples are
placed.
The temperature difference in the specimen is approx. 10C. The average value of the two temperature
values measured along the specimen is considered as specimen temperature.
The tests were conducted at the labs of sterreichisches Giesserei-Institut, Leoben, Austria.
4) Results and Discussion:
Table VII and figure 6 show the results obtained. The numbers are the average of 2 measurements.
The results show that gray iron presents always higher thermal conductivity than CGI. The differences
decrease with increasing temperature, because for gray iron thermal conductivity results decrease with

increasing temperature, while for CGI the results do not show a significant variation with the temperature.
The highly alloyed Gray Iron Grade 300, used for cylinder heads of heavy diesel engines, presents lower
thermal conductivity than the Gray Iron Grade 250. This is caused by the lower carbon content and by the
alloying elements, reducing thermal conductivity.
Comparing with the results on Table II, the effect of temperature, reducing the thermal conductivity of
gray irons, is much higher in the results we obtained (figure 6). One possible reason for that is that the carbon
content from our samples (3,3-3,5) is higher than the usual carbon contents of gray iron (3,2-3,4), resulting in
higher amount of graphite (lamellar).
Comparing the two grades of CGI (figure 6), one can observe that the CGI Grade 350 should be consider a
candidate material for cylinder heads, because of the higher thermal conductivity compared to the CGI Grade
450. In this case, the larger amount of ferrite in the CGI Grade 350 resulted in increasing thermal
conductivity.
5) Conclusions:

Gray iron always presents higher thermal conductivity comparing with compacted graphite
iron, due to the graphite form;

Increasing the temperature and adding alloying elements, thermal conductivity decreases for
gray irons;

CGI 350 presents higher thermal conductivity comparing with CGI 450, due to the higher
ferrite content.
References:
Angus, H. T. Cast Iron: Physical and Engineering Properties. BCIRA, p. 126-134, 1960.
Cueva, G. et all. Desgaste de ferros fundidos usados em discos de freio de veculos automotores. SAE 2000,
So Paulo.
Gusseisen mit Kugelgraphit. Konstruiren + Giessen. Zentrale fr Gussvervendung. VDI-Verlag, Dsseldorf,
1988.
Gusseisen mit Lamellengraphit Eigenschaften und Anwendung. Konstruiren + Giessen. Zentrale fr
Gussvervendung. VDI-Verlag, Dsseldorf, 2000.
Gundlach, R. B. The effects of alloying elements on the elevated temperature properties of gray irons. AFS
Transactions, 1983, p. 389.
Hasse, S. Duktiles Gusseisen. Schiele & Schn, Berlin, 1996.
Monroe, R. W. & Bates, C. E. Some thermal and mechanical properties of compacted graphite iron. AFS
Transactions, 1982, p. 615.
Shao, S. et all. The Mechanical and Physical Properties of Compacted Graphite Iron. Sintercast, 1997.
Silva Neto, E. Relaes entre propriedades e a microestrutura de materiais bifsicos caracterizao
especfica para os ferros fundidos ferrticos nodular e cinzento. Dissertao de mestrado, UFSC, 1978.
Stefanescu, D. Physical properties of cast iron. In: Goodrich, G.M. Iron Castings Engineering Handbook .
AFS, 2003.
Thermal conductivity of solids by means of the guarded-comparative-longitudinal heat flow technique.
ASTM E 1225 -99.
Thermal conductivity of unalloyed cast iron. BCIRA Broadsheet 203, 1981.

Table I Thermal conductivity of main metallographic phases in cast irons at room temperature (Stefanescu,
2003).
Metallographic constituent
Ferrite
Pearlite
Cementite
Graphite
Parallel to basal plane
Perpendicular to basal plane

Thermal conductivity, W m-1 C-1

0 100 C
500 C
1000 C
71 80
42
29
50
44
40
78
293 - 419
84 126
42 63
84
-

Figure 1 The thermal conductivity of graphite parallel to basal plane is higher than perpendicular (Hasse,
1996).

Figure 2 Effect of graphite amount on the thermal conductivity, for gray iron ( O ) and ductile iron ( ),
with ferritic matrix (Silva Neto, 1978)

Table II - Results of thermal conductivity for different grades of gray Iron (Konstruiren + Giessen-2000).
Temperature
(C)
100
200
300
400
500

Thermal conductivity (W/K.m)

GJL 150
52,5
51,5
50,5
49,5
48,5

GJL 200
50,8
49,8
48,8
47,8
46,8

GJL 250
48,8
47,8
46,8
45,8
44,8

GJL 300
47,4
46,4
45,4
44,4
43,4

GJL 350
45,7
44,7
43,7
42,7
41,7

GJL 400
44,0
43,0
42,0
41,0
40,0

Table III Results of thermal conductivity for ductile irons (Konstruiren + Giessen-1988)
100 C
200 C
300 C
400 C
500 C

GGG-35.3
40.2
43.3
41.5
38.8
36.0

GGG-40
38.5
41.5
39.8
37.4
35.0

GGG-50
36.0
38.8
37.4
35.3
33.5

GGG-60
32.9
35.4
34.2
32.8
31.6

GGG-70
29.8
32.0
31.0
30.3
29.8

4 Si-Mo
25.1
27.2
28.1
28.6
28.9

Table IV Change in Thermal Conductivity of Gray Iron Upon Addition of 1%Alloying Element
(Stefanescu, 2003).
Element

Experimental range %
16
0.65 4.15 (ductile iron)
0 1.5
?
0 0.39
0 0.5
0 1.58
0 0.74
0 0.58
0 0.475
0 0.12

Silicon
Manganese
Phosphorus
Chromium
Copper
Nickel
Molybdenum
Tungsten
Vanadium

Change in k, %
-6
-14.7
-2.2
-6
+21
-30
-4.7
-14.5
-12
-5.2
0

oC)
Thermal
Conductivity
Therm
al Conduc
tiv ity (W/m( W/m -C)

50
Gre y Iro n, 3.25 % C
45

3.6%C, 80% P, 5% N

40

3.7%C, 70% P, 10 %N

3.7%C, 98% P, 3% N

35

3.5%C, 95% P, 2% N
30
0

100

200

300

400

500

Te
m perature(o(C)
Temperature
C)

Figure 3 Thermal conductivity results of compacted graphite iron, compared to a gray iron (Shao-1997).

Temperatura (oC)Temperature

(C)

Figure 4 Effect of nodularity on the thermal conductivity of cast irons (Monroe & Bates, 1982)

280

Fe 250

240

Fe AC
Fe Ti

200

Vermic

160
120
80

40 of the break disk during the breaking cycles (Cueva, 2000)

Figure 5 Temperature
0

Tempo de ensaio (min)

Time (min)
Table V - Chemical composition of CGI and Gray Iron samples.
Elements
C (%)
Si (%)
Mn (%)
Cu (%)
Sn (%)

CGI 350
3,65
2,45
0,37
0,41
0,031

CGI 450
3,62
2,41
0,37
1,17
0,064

Gray Iron 250

3,43
2,07
0,55
1,00
0,10

Gray Iron 300

3,30
2,05
0,56
1,20
0,11

Cr (%)

0,029

0,029

0,20

0,24

Mo (%)

0,30

Table VI - Microstructure and mechanical properties of the CGI samples.

CGI 350
48
4% nodulary
371
292
2,5

Ferrite (%)
Graphite shape
UTS (MPa)
YS (MPa)
E (%)

CGI 450
2
7% nodularity
498
443
1,4

Gray Iron 250

0
A, size 4
270
-

Table VII - Thermal conductivity results

Thermal conductivity (W/K.m)

Temperature
(C)
100
200
300
400

CGI 350
37,0
37,4
37,2
36,5

CGI 450
33,6
34,2
34,3
33,9

Gray Iron 250

50,0
46,6
43,6
40,9

Gray Iron 300

45,5
43,15
41,2
39,7

55
GI 250
CuSnCr

thermal conductivity (W/K.m)

50

GI 300
CuSnCrMo

45
40

CGI 350
35
CGI 450
30
25
20
0

50

100

150

200

250

300

temperature (C)

Figure 6 - Thermal conductivity results of CGI and Gray Iron.

350

400

450

Gray Iron 300

0
A, size 4
317
-