This article was downloaded by: [University of Alberta]

On: 14 June 2015, At: 00:13

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Welding International
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/twld20

Features of the brittle fracture of weld metal

deposited using electrodes with basic type coatings
a

I K Pokhodnya , A O Korsun , Yu Ya Meshkov , G A Pakharenko & A V Shevchenko

a

E O Paton Welding Institute, Ukraine SSR Academy of Sciences ,

Ukr. SSR Academy of Sciences Institute of Metal Physics ,

Published online: 04 Jan 2010.

To cite this article: I K Pokhodnya , A O Korsun , Yu Ya Meshkov , G A Pakharenko & A V Shevchenko (1987) Features of the
brittle fracture of weld metal deposited using electrodes with basic type coatings, Welding International, 1:7, 606-609, DOI:
10.1080/09507118709453005
To link to this article: http://dx.doi.org/10.1080/09507118709453005

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the Content) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

A,fornuticheskup Svurku 1986 39 ( 10) 1-5

Features of the brittle fracture of weld metal deposited

using electrodes with basic type coatings
I K Pokhodnya and A 0 Korsun
E 0 Paton WeldingInstitute, Ukraine SSR Academy of Sciences

Yu Ya Meshkov, G A Pakharenko and A V Shevchenko

Ukr. SSR Academy of Sciences Institute of Metal Physics

Downloaded by [University of Alberta] at 00:13 14 June 2015

Selected from Anomorirhrsskuyu S w k o 1986 39 (10); Reference AS/86/10/1;Translation 175

The difficulties of making welded joints which will be

reliable in northern areas are in many cases due to the
absence of reliable criteria for evaluating the conditions
under which brittle fracture takes place. Modem
fracture theory, based on the microscopic cleavage
model,' makes it possible to link quantitatively
particular structural components made of a particular
metal with minimum brittle fracture stresses in the
ductile-brittle transition temperature range (Fig.1) by
simple equations:
for carbon steels and armco iron (for which the stress
intensity factor at the moment of fracture K, = 6MPa
x m'/2)

R,,

...[ 11

Kf,d-'j2

and for steel with globular shaped second phase

particles

R,,

...[2]

Kfrd-r

where K, = 0.8MPa X rn'l2;

is the microscopic
cleavage stress, MPa; d is the grain size, nun; d, is the
diameter of a globular second phase particle, mm. Since
under conditions of overall macroscopic flow (when the
applied stress has reached the yield stress) submicroscopic cracks appear at the grain boundaries in
metal, and since cleavage (shearing) of the second phase
particles takes place, the level of resistance to brittle
fracture is governed by the structural parameter
initiating the larger new sub-microscopic crack.'
The ratio of the microscopic cleavage stress Kcto
the nominal yield stress
called the coefficient of

1 Relationship of mechanical properties of weld metal

- true
containing I% of nickel to temperature: S,,
breaking stress;
RZf
- experimentally found
microcleavage stress.

toughness K,in Ref.1, represents the potential capacity

of a metal for resisting brittle fracture: the greater the
difference between the values of uo.2and K,, the greater
the degree of over-stressing (caused by the deformation
rate, test temperature or type of stress state) it can
withstand without risking microscopic cleavages turning
into serious fracture.
Our work consisted of making a study of whether a
quantitative relationship can be established between the
structural state of a weld metal and the particulars of its
brittle strength and toughness.

Table 1 Chemical composition of the deposited metals, O h

Electrode

Proportion by,
mass of nickel in
coating, yo

c .

Mn

Si

Ni

NO
N1
N2

0
2
4

0.095
0.090
0.090

0.95
1.05
1.03

0.36
0.28
0.30

0.019
0.014
0.016

0.024
0.022
0.021

606

1.o
1.8-2.0

WELDING INTERNATIONAL 1987 N0.7

2 Structures of weld metals (X2000, scale reduction 2/31 containing different amounts of nickel: (a) 0%; (b) 1%; (c)
2%.

Table 2 Mechanical properties of the metals investigated

Electrode

,,,t

"C

d,,mm

NO

+ 20

Downloaded by [University of Alberta] at 00:13 14 June 2015

R$r,MPa,
calculated
.usinall1

1,%

K,

KCV, J/cm2

d,, mm

Reap
mc >

M Pa

390
400
410

1000
1000
1000

1800
1800
1800

0.20
0.20
0.20

2.56
2.50
2.44

195-210
45-47
12-16

0.005
0.005
0.005

1100
1100
1100

+ 20
- 40

0.008

370
390
410

950
950
950

2010
2010
2010

0.19
0.19
0.19

2.57
2.44
2.32

165-170
85-95
48-58

0.004
0.004
0.004

1120
1120
1120

+ 20

0.006

430
490
520

1000
1000
1000

2250
2250
2250

0.17
0.17
0.17

2.33
2.04
1.92

160-180
78-94
64-82

0.004
0.004
0.004

1140
1140
1140

- 60
N2

~ ~ , ~ , m
R','!,MPa
m

0.010

- 40
- 60

N1

Heat treated

Not heat treated

- 40
- 60

Comment. In every cased,"== 0.0008-0.001mm;R*:F. calculated using[3], is 111OMPa.

3 Distribution of non-metallic inclusions in the structure

of a weld (X5400, scale reduction 314).

4 Fracture nuclei caused by

1x2000, scale reduction 213).

non-metallic inclusion

The experiments were made on testpieces cut from

the metal in the normalised zone of a multipass welded
joint made using 4mm diameter ANO-26 electrodes
with a basic type coating. Grade PNE-1 nickel powder
(GOST 9722-79), its proportion by mass 0-4%, was
added to the coating. The welds were made with AC
from an STSh-500/80 transformer. The welding
conditions were: I, = 180A, V,,, = 23-24V. The parent
WELDING INTERNATIONAL 1987 No.7

metal was 14mm thick 09G2S steel (GOST 19282-73).

The edges were prepared by variant B from GOST
9466-75.
The chemical composition of the metals deposited
with the experimental electrodes is given in Table 1. The
structures and fractures of the testpieces were studied
with a JSM-35CF scanning electron microscope (JEOL,
Japan). The composition of the non-metallic inclusions
was determined with a Link-860 energy dispersion
spectrometer (Link, Great Britain). The tensile tests
were made at between +20 and - 196C with a UMM5 machine.
In addition to this the quality of the weld metal and
its susceptibility to embrittlement were evaluated from
the results of impact tests on testpieces with an acute
angled notch (type IX,GOST 6996-66) at temperatures
between + 20 and - 60C. The proportion f by volume
and dimensions d,, of the non-metallic inclusions were
detennined with a Quantimet-720 quantitative television
microscope (Metals Research, Great Britain). The nonmetallic inclusions smaller than 1pm were also analysed
with the JSM-35CF scanning electron microscope.
The results of the analysis with the electron
microscope confirmed that the secondary structures of
the testpieces investigated consisted of equiaxed ferrite
grains with a fringe of bainite precipitates. A large
number of non-metallic inclusions arranged both within
the grain bodies and around their boundaries were also
found. Special etching (using heated picric acid), camed
out to reveal the grain boundaries (Fig.2), made it
possible to determine the dimensions d, of the actual
structural elements at whose boundaries sub607

Downloaded by [University of Alberta] at 00:13 14 June 2015

microcracks initiating microscopic cleavage form under

general yield conditions.
Alloying weld metal with nickel reduced the grain
size in it (Fig.2). As would be expected the mechanical
properties of the metal changed. For instance, when the
weld metal contained between 0 and 2% of nickel the
impact strength of testpieces with an acute angled notch
at -60C increased by about 400%, and the strength
also increased slightly (Table 2).
The results of determining brittle strength were
however unexpected. It was established that, in spite of
the decrease in grain size d,, the microcleavage stresses
kcremained practically unchanged, while according to
calculations made using (11 there should have been a
substantial increase in brittle strength :
:*R
(Table 2).
The only reason why the calculated and experimental
data did not agree can be that, in the testpieces
investigated, microcleavages were caused not by the
'grain' source of sub-microcracks but by a different
source forming large sub-microcrack nuclei which
consequently developed under lower stresses. The nonmetallic inclusions invariably present in the weld metal
(Fig.3) might be this additional source of microcleavage
in the testpieces studied. The decisive part played by the
non-metallic inclusions in the fracturing of the weld
metal was confirmed by the results obtained in Ref.2.
Fractographic analysis of the surfaces of testpieces
which had failed by brittle fracture revealed the

42 0,s

1,O

1,4

48

52

dine ~m

5 Histogram of distribution of non-metallic inclusions by

size (weld metal containing 1% of nickel).

6 Structure of weld metal containing no nickel after heat

treatment (X2400, scale reduction 314).
608

existence of fracture nuclei caused by non-metallic

inclusions (Fig.4).
It was proved in Ref.3 and 4 that, under certain
conditions, the level of brittle strength depends on the
dispersion and distribution of the globular second phase
particles. At the same time, with a proportion by volume
of less than 4.5%, the particles within the bodies of
grains cannot be fracture sources. In this case fracture is
initiated by second phase particles at the grain
boundaries, and the microcleavage stress is found using
the equation
R,,

K,d-r

...[3]

where K , = 1.lMPa X m1'2. The most likely diameter

dincof these particles in the testpieces investigated (Table
2) was found using the histogram (Fig.5) of distribution
of non-metallic inclusions by dimensions.
Table 2 also contains the results of calculating the
microcleavage stress R'i:r using (31. Comparison of the
calculated' and experimental (RZZ) data leads to the
conclusion that the non-metallic inclusions play the
decisive part in the brittle fracture of the steel
investigated.
In order to confirm the law which had been
established an additional experiment was carried out.
Testpieces were quenched (austenitisation temperature
9SOOC) and then tempered at 380C. This heat
treatment caused a different structure to form (Figd),
different as to both phase composition and grain size,
but had no effect on the dimensions of the non-metallic
inclusions, since these are complex alloyed oxides
(Fig.7), whose temperature of solution is, according to
Ref.5, - 160O"C, well above the heat treatment
temperature. It can be seen from Table 2 that the heat
treatment also scarcely affected the microcleavage stress
R~iY
obtained by calculation using [3].
It can thus be argued that the level of brittle
strength of the testpieces considered is governed not by
the grain size but by the dispersion of the non-metallic
inclusions. Moreover the coefficient of toughness for the
steel investigated, K, = ( R J U ~ , ~ )also depends to a
great extent on the size of the non-metallic inclusions
(Table 2). Decrease in grain size increases the nominal
yield stress u ~ ,but
~ , with a fixed and constant level of
microcleavage stress kcthis is inevitably accompanied
by a drop in toughness margin. The greater the rate at
which the yield stress increases the greater will be this
drop.
It should therefore be argued that the beneficial
effect associated with the presence of nickel in weld
metal has not essentially been realised (only the impact
strength has been increased). Weld metal containing 2%
of nickel has a lower coefficient of toughness (at
temperatures between 20 and - 60C) than weld metal
containing no nickel. Using the model of microcleavage
for analysing the relationship of the resistance to brittle
fracture of weld metal to the structure has made it
possible to establish the effective structural parameter
controlling the process of brittle fracture. This
WELDING INTERNATIONAL 1987 No.7

.$

Downloaded by [University of Alberta] at 00:13 14 June 2015

7 Alloying element contents of an

- lpm diameter non-metallic inclusion.

parameter proved, in the case of the steel investigated,

to be the size of the non-metallic inclusions. This is why
alloying with nickel, which does not affect the dispersion
of these inclusions, does not, unless the system of
deoxidation is correctly selected, increase the brittle
strength of weld metal. The increase in impact strength
accompanying increase in nickel content does not
represent the true toughness margin of the metal, and is
due to increase in the yield stress. Reference 6 also
found that the beneficial effect of nickel (with
manganese contents of over 1%) on the mechanical
properties of welds is reduced.
Establishment of the relationship between the
strength and ductility of weld metal and the structural
parameters makes it possible to find ways of further
improving the mechanical properties, consisting of
reducing the size of the structure elements capable of
initiating sub-microcracks. The research described in
Ref.4 proved that the combination of brittle strength and
toughness margin is optimal with a ratio of 25:l between
the grain size in the steel and diameter of a globular
particle of the second phase, with the maximum
dispersion. It was for instance proved in Ref.7 and 8
that, when the Mn/Si ratio is increased, the size of the
non-metallic inclusions is reduced.
Therefore, in order to provide welded joints in steel
with high service particulars at subzero temperatures,
the systems by which the weld metal is alloyed and
deoxidised must be selected in conjunction.
CONCLUSIONS
1 Globular shaped non-metallic inclusions, which are

WELDING INTERNATIONAL 1987 No.7

E, keV

products of the deoxidation of weld metal with

manganese, silicon and titanium, may be sources of
brittle fracture, thus limiting the possibility of increasing
the toughness margin of a weld.
2 The existence of these non-metallic inclusions
prevents advantage from being taken of the beneficial
effect of nickel on brittle strength (in spite of decrease in
stnictural grain size and increase in impact strength of
weld metal).
3 One way of improving the brittle strength and
toughness of weld metal is to match the reduction in size
of the grains and non-metallic inclusions.
References
1
Meshkov Yu Ya: T h e physical fundamentals of the fracturing of
steel structures.' Pub1 Naukova duma, Kiev, 1981.
2
bkhodnya I K el a,! T h e failure of single pass meld metal at
temperatures within the brittleductile transition range.' Art. Svurku
1979(3) 1-4.
3
Meshkov Yu Ya er oL 'Breakdown of steel containing granular
cementite.' Aferollofuiko 1983 5 (3) 94-97.
4
hleshkov Yu Ya er ut T h e brittle fracture of carbon steels mith
different granular cementite distributions.' Ukr. SSR Academy of
Sciences Institute of Metal Physics 1984 13. hl/s deposited at VlNITI
on 21.12.84 No.8217-84.

5
Podgaetskii V V: "on-metallic inclusions in welds.' Pub1
hlashgiz. Moscow, 1962.
6
Taylor D S: 'Development of MMA electrodes for offshore
fabrication.' IVeId. 1.1983 15 f8) 438-443.
7
bkhodnya I K ef ot T h e non-metallic inclusions in welds made
using electrodes with m i l e and ilmenite coatings.' A,% Svorko 1976 (9)
8-11.

8
Widgety D J: 'Deoxidation practice for mild steel weld metal.'
\\kldJ. 1976 55 (3) 125-137.

609