Download Relation between Surface Tension and Graphite Shape in Cast Iron

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Self-assembled monolayer wikipedia , lookup

Tunable metamaterial wikipedia , lookup

Low-energy electron diffraction wikipedia , lookup

Nanochemistry wikipedia , lookup

Ultrahydrophobicity wikipedia , lookup

Adhesion wikipedia , lookup

Wetting wikipedia , lookup

Sessile drop technique wikipedia , lookup

Centrifugal micro-fluidic biochip wikipedia , lookup

Surface tension wikipedia , lookup

Transcript
Materials Transactions, Vol. 49, No. 9 (2008) pp. 2163 to 2165
#2008 The Japan Institute of Metals
RAPID PUBLICATION
Relation between Surface Tension and Graphite Shape in Cast Iron
D. Shi* , D. Li, G. Gao and L. Wang
Department of Material Science and Engineering, Harbin University of Science & Technology, Harbin, 150040, P.R. China
A relationship between the surface tension of cast iron and the graphite shape has been proven by experimental analysis and thus the
graphite shape can be forecast by the surface tension. A specific relation between the surface tension and the graphite shape has been established
by experiments. When the surface tension is above 1385 mN/m, nodular graphite occurs. When the surface tension is between 1283 mN/m and
1385 mN/m, vermicular and nodular graphite occurs. When the surface tension is between 1108 mN/m and 1283 mN/m, vermicular graphite
occurs. When the surface tension is between 990 mN/m and 1108 mN/m, flake and vermicular graphite occurs. When the surface tension is
below 990 mN/m, flake graphite occurs. [doi:10.2320/matertrans.MRP2008165]
(Received May 19, 2008; Accepted July 4, 2008; Published August 13, 2008)
Keywords: surface tension, graphite shape, specific relation, cast iron
1.
Introduction
It is well known that graphite shape has an important
influence on mechanical properties of cast iron.1,2) To
effectively improve the quality of the castings and reduce
the amount of rejected materials, rapidly monitoring the
graphite shape of cast iron is becoming very important.3) As a
consequence, graphite shape monitoring is strongly recommended. Although many methods have been invented or
developed,4–7) the most widely used ones are metallographic
observation and thermal analysis. As accurate they are, their
measuring time is too long and their operation is too
complicated. Therefore, neither metallographic observation
nor thermal analysis can meet the requirement upon the
testing speed and the operative convenience. Hence, to
effectively monitor the graphite shape and thus improve
mechanical properties of cast iron, there is a need for a new
and better method. Surface tension is one of important
physicochemical properties of cast iron.8) It is the foundation
of studying the interface dynamics.9) In solidification phenomena, surface tension plays important roles in optimizing
and simulating the crystal growth parameters,10) forecasting
the degree of the modification,11) porosity formation,12)
mould filling capacity and surface roughness.13,14)
In the present study, experimental analysis has shown that
a relationship between the surface tension of cast iron and the
graphite shape does occur. A specific relation has been also
established by experiments. It will provide an effective
method and convenience for rapidly monitoring the graphite
shape of cast iron.
2.
Relationship Between Surface Tension and Graphite
Shape
Early in 1951, Buttner had concluded that high interfacial
tension of the liquid cast iron could promote the formation
of nodular graphite, which has been approved by many
researches.15,16) However, the measurement of the interfacial
tension is very difficult, and so the method for forecasting
the graphite shape by the interfacial tension is unfeasible.
According to the interfacial tension theory and the crystal
*Corresponding
author. E-mail: [email protected]
growth theory,17,18) one of conditions for the graphite
growing into nodular shape is that the interfacial tension
between the graphite and the liquid cast iron must be
higher than between others, and it is the interfacial tension
that decides final shape of the graphite. For the graphite’s
aeolotropy, however, the interfacial tension between various
crystal faces and the liquid cast iron differs. After surface
active elements (such as oxygen and sulfur) are adsorbed to
the edge plane between the graphite and the liquid cast iron,
the interfacial energy and the contact angle will reduce, and
the growth speed of the graphite will increase. In the end, the
graphite will grow into the flake along the edge plane.
Contrarily, because the modifying agents have very strong
affinities for oxygen and sulfur, surface active elements on
the edge plane will combine with the modifying agents and
thus their quantity will reduce. Accordingly, the interfacial
energy of the edge plane is much bigger than that of the base
plane, and the interfacial tension and the surface tension
increase simultaneously. As a consequence, the graphite will
change from flake shape into vermicular or nodular shape.
Whenever graphite particles occur in the liquid cast iron,
there are three tensions among the graphite, the liquid cast
iron and the air. Their relationships can be expressed by the
eq. (1).19)
LS ¼ SG þ LG cosð180 Þ
ð1Þ
where is the contact angle between the graphite and the
liquid cast iron.
Because the interfacial tension SG between the graphite
and the air retains the same, the change of the interfacial
tension LS between the graphite and the liquid cast iron
can be indicated by LG cosð180 Þ, and the bigger is the
interfacial tension LS , the greater is the surface tension LG
between the liquid cast iron and the air. Furthermore, the
graphite shape has close relation with the change of the
interfacial tension LS . Hence, the graphite shape could be
forecast instantly if the surface tension of the liquid cast iron
were measured.
3.
Experimental and Establishment of Specific Relation
A modified maximum bubble pressure method20) is used
to fast measure the surface tension of the liquid cast iron
2164
D. Shi, D. Li, G. Gao and L. Wang
and the basic principle is as follows. A capillary is
vertically dipped in the liquid cast iron and the dipping
depth can change at the range of 10–20 mm. Pressurized
inert gas is fast blown in the liquid cast iron through the
capillary and bubbles will generate and vanish continuously
at the speed of 2–3 bubbles per second. During the course
of blowing gas, the pressure in the bubble is monitored by
computer and the pressure curve is proved to be a sinusoid.
According to the Laplace’s formula,21) the amplitude and
the period of the sinusoid reflect the additional pressure
caused by the surface tension and the speed of generating
bubbles respectively. To simplify the calculation, the period
of the sinusoid is replaced by the number of bubbles in five
seconds. In addition, the temperature of the liquid cast
iron and the capillary diameter will affect the calculation of
the surface tension. Therefore, if the amplitude P, the
number N of bubbles in five seconds, the capillary diameter
x , and the temperature T of the liquid cast iron are fast
measured, the surface tension can be calculated instantly
by the following equation.
(a) σ =782mN/m, flake
(b) σ =894mN/m, flake
(c) σ =1082mN/m, flake +vermicular
(d) σ =1203mN/m, vermicular
(e) σ =1360mN/m, vermicular+nodular
(f) σ =1465mN/m, nodular
¼ 1:6641P þ 19:5022 N
þ 0:2038 ðx 600Þ 0:3581 T þ 292:9447
ð2Þ
According to this method, a device has been constructed to
fast measure the surface tension of the liquid cast iron in five
seconds. Many experiments have shown there are small
deviations between the surface tension measured by this
device and the reference values.
Experimental materials for studying the relation between
the surface tension and the graphite shape are pig iron Z14,
modifying agent 1RETiMg10-5 and inoculant 75% FeSi.
The composition of the liquid cast iron after modification
is as follows: C 3.7–4.2%, Si 2.2–3.0%, Mn 0.3–0.5%,
P < 0:06%, S < 0:02%, Re > 0:035%, Ca < 0:02%. To
obtain the liquid cast iron with different surface tension and
different graphite shape, the following melting process is
adopted. Pig iron Z14 is melted in a high frequency induction
furnace. Over the 1500–1520 celsius degree temperature
range, the liquid cast iron is modified by the modifying
agent 1RETiMg10-5, and the mass fraction of the modifying
agent is varied from 0.0% to 3.0% in 0.025% steps. Then the
liquid cast iron is inoculated by 75% FeSi. After inoculation
treatment, the surface tension of the liquid cast iron is fast
measured by the self-developed device over the 1340–1410
celsius degree temperature range. At the same time, the
metallographic samples are poured and standard graphite
shape can be got by microstructural analysis.
When the temperature of the liquid cast iron changes over
small temperature range, its surface tension will change
slightly.22) Moreover, as shown in eq. (2), the surface tension
has been amended by the temperature when the surface
tension is fast measured by the self-developed device. So,
the temperature can be neglected when the relation between
the surface tension of cast iron and the graphite shape is
established.
Many experiments have been performed. When the
quantity of the modifying agent differs, the microstructure
and the surface tension are shown in Fig. 1. The modifying
agent has a great effect on the microstructure and the surface
tension of cast iron. With the increase of the modifying agent,
Fig. 1 Microstructure and corresponding surface tension at different
quantity of the modifying agent.
the surface tension will change from the minimum to the
maximum, and the graphite will change from flake to nodular
shape correspondingly. Figure 1(a) shows the microstructure
of unmodified cast iron and its surface tension. The graphite
is flake shape and the surface tension is lower than the others,
and it is 782 mN/m. When a small quantity of the modifying
agent is added, although the surface tension increases and
it is 894 mN/m, the graphite shape is still flake, as shown
in Fig. 1(b). Then, from Fig. 1(c) to Fig. 1(e), the surface
tension increases continuously with the increase of the
modifying agent, and the graphite will change from flake
shape to vermicular shape. When the modifying agent
continues to be added, as shown in Fig. 1(f), the graphite
will be nodular shape and the surface tension will reach the
maximum, which is 1465 mN/m.
Through detailed comparison and statistical analysis between the microstructure and the surface tension, a specific
relation between the surface tension and the graphite shape
of cast iron has been obtained, which is shown in Table 1.
When the surface tension is below 990 mN/m, the flake
graphite will occur. When the surface tension is between
990 mN/m and 1108 mN/m, the flake and vermicular
graphite will occur. When the surface tension is between
1108 mN/m and 1283 mN/m, the vermicular graphite will
occur. When the surface tension is between 1283 mN/m
and 1385 mN/m, the vermicular and nodular graphite will
occur. When the surface tension is above 1385 mN/m, the
nodular graphite will occur. According to the specific
relation, if the surface tension of cast iron were measured,
Relation between Surface Tension and Graphite Shape in Cast Iron
Table 1 Specific relation between the surface tension and the graphite
shape.
Surface tension (mN/m)
Graphite shape
990
flake
990 < 1108
1108 < 1283
flake + vermicular
vermicular
1283 < < 1385
vermicular + nodular
1385
nodular
the graphite shape could be got quickly. Therefore, it will
provide an effective means for rapidly monitoring the
graphite shape and thus scientifically guiding the production
of cast iron.
4.
Conclusions
(1) Experimental analysis has proved that a relationship
between the surface tension of cast iron and the graphite
shape does occur, so the graphite shape of cast iron can
be forecast quickly by the surface tension.
(2) Through detailed comparison and microstructural analysis, a specific relation between the surface tension of
cast iron and the graphite shape has been established.
Acknowledgements
The authors would like to acknowledge Z. J. Liu, Q. Sun,
G. W. Gao, and G. Du for useful discussions and experimental support. This work was supported by National
Natural Science Foundation of China.
2165
REFERENCES
1) Y. H. Song, T. S. Lui, L. H. Chen and J. M. Song: Mater. Trans. 44
(2003) 1461–1468.
2) T. Sjogren, P. Vomacka and I. L. Svensson: Int. J. Cast Met. Res. 17
(2004) 65–71.
3) A. De Santis, O. Di Bartolomeo, D. Iacoviello and F. Iacoviello:
J. Mater. Process. Technol. 196 (2008) 292–302.
4) Z. Li and Y. Li: Metall. Mater. Trans. A 36 (2005) 2455–2460.
5) M. M. Liliac, K. Yamamoto and K. Ogi: Int. J. Cast Met. Res. 14 (2001)
157–164.
6) Y. Li and Q. Wang: J. Mater. Process. Technol. 161 (2005) 430–434.
7) C. A. S. Ribeiro, D. Santos and W. Baumgart: Mater. Sci. Forum
455–456 (2004) 699–702.
8) Z. Li, M. Zeze and K. Mukai: Mater. Trans. 44 (2003) 2108–2113.
9) P. Lallemand, L. S. Luo and Y. Peng: J. Comput. Phys. 226 (2007)
1367–1384.
10) M. Yildiz and S. Dost: Cryst. Res. Technol. 42 (2007) 914–919.
11) K. Nogita, S. D. McDonald and A. K. Dahle: Mater. Trans. 45 (2004)
323–326.
12) D. Emadi, J. E. Gruzleski and J. M. Toguri: Metall. Trans. B 24B
(1993) 1055–1063.
13) N. Hirata and K. Anzai: J. Japan Foundry Eng. 78 (2006) 636–640.
14) L. Suresh and J. Y. Walz: J. Colloid. Interface Sci. 183 (1996) 199–213.
15) A. N. Roviglione and J. D. Hermida: Metall. Mater. Trans. B 35 (2004)
313–330.
16) M. Takita and Y. Ueda: Trans. Jpn. Inst. Met. 20 (1979) 569–576.
17) K. K. Mon and D. Stroud: Phys. Rev. B 25 (1982) 6478–6481.
18) P. Bennema, H. Meekes, S. X. M. Boerrigter, H. M. Cuppen, M. A.
Deij, J. van Eupen, P. Verwer and E. Vlieg: Cryst. Growth Des. 4
(2004) 905–913.
19) K. Jagannadham and M. J. Marcinkowski: J. Mater. Sci. 14 (1979)
1717–1732.
20) D. Shi, D. Li, G. Gao and L. Wang: Sci. China Ser. E 49 (2006)
569–575.
21) A. G. Meyra, G. J. Zarragoicoechea and V. A. Kuz: Fluid Phase
Equilib. 235 (2005) 191–195.
22) N. Takiuchi, N. Shinozaki, T. Taniguchi, K. Mukai and Y. Tanaka:
J. Jpn. Inst. Met. 55 (1991) 180–185.