Journal of Manufacturing Processes 30 (2017) 313–319

Contents lists available at ScienceDirect

Journal of Manufacturing Processes

journal homepage: www.elsevier.com/locate/manpro

Research paper

Performance of resin bonded sand for magnesium alloy casting

Fuchu Liu ∗ , Li Yang, Ying Huang, Peng Jiang, Guang Li, Wenming Jiang, Xinwang Liu,
Zitian Fan ∗
State Key Laboratory of Materials Processing and Die and Mould Technology, Huazhong University of Science and Technology, Wuhan 430074, China

a r t i c l e i n f o a b s t r a c t

Article history: The factors which affect performance of no-bake resin bonded sand (NBRBS) were investigated, including
Received 16 November 2016 furan resin, curing agent and boric acid content and kinds of collapsibility agents and base sand. The results
Received in revised form show that when the furan resin addition is 1.6 wt% of base sand content and the curing agent addition is
30 September 2017
50 wt% of furan resin content, the comprehensive performance of NBRBS achieves optimization. The use
Accepted 2 October 2017
of boric acid increases the curing speed, retained strength and gas evolution. The use of KI reduces the
retained strength with a good tensile strength. When ceramacore being used as base sand, furan resin
KeyWords:
addition decreases greatly, and the tensile strength increases, gas evolution and retained strength both
Sand casting
Furan resin
decrease, which can improve collapsibility of NBRBS. The SEM analysis demonstrates that the ceramacore
Magnesium alloy casting particle is round, appeares standard sphere, and the ceramacore can form complete bonding film and
Collapsibility bridge.
Tensile strength © 2017 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
Gas evolution

1. Introduction production requirement of magnesium alloy castings and collapsi-

No-bake furan resin bonded sand (NBRBS) is widely used in
magnesium alloy casting for its high dimensional precision, high
strength, convenient operation and so on [1–4]. Due to the low
pouring temperature ranges from 700 ◦ C to 750 ◦ C of magnesium
alloy castings and low thermal melting amount, there is not enough
heat for burning bonding bridges of the resin bonded sand sam-
ples, and so it leads to poor collapsibility for mold sand or core
of magnesium alloy castings, especially the large complex cast-
ing molds or cores for less heat of liquid metal and its difficulty
of deeping into the internal core, which could bring great difficul-
ties for cleaning after casting and reclamation treatment of used
sand [5–8]. Therefore, it has the important practical significance
for studying optimized formula of no-bake furan resin bonded sand
that satisfies production requirement of magnesium alloy castings
and collapsibility. The recent research indicates that there are many
factors which affect the performance of resin bonded sand, includ-
ing resin content, curing agent content, types of base sand and other
additives [9–12].
The effects of furan resin and curing agent addition on tensile
strength, retained strength and gas evolution of NBRBS have been
studied in this paper, the optimized basic formula which satisfies
∗ Corresponding authors.
E-mail addresses: lfchust@163.com (F. Liu), fanzt@hust.edu.cn (Z. Fan).
bility has been defined. The effects of boric acid content, kinds of
collapsibility agents and base sand on the performances of NBRBS
have been studied based on the above optimized formula. Scanning
electron microscope (SEM) has been used to analyse surface mor-
phology of NBRBS which were prepared by different kinds of base
sand.
2. Materials and methods
2.1. Materials and equipment
There were three kinds of base sand used in the experiment,
including silica sand from Dalin sand and Duchang sand, cerama-
core with 40/70 meshes, the chemical composition of different base
sand is listed in Table 1. Furan resin (Shangdong Jinan Shengquan
Group Co.,Ltd., SQG700) was used as a binder, and curing agent
(Shangdong Jinan Shengquan Group Co., Ltd., GC-09) was selected
as its hardener. The characteristics of SQG700 furan resin and GC09
curing agent are listed in Tables 2, respectively. The other addi-
tives were as follow: boric acid powder, potassium nitrate powder,
sodium hydrogen carbonate powder, potassium iodide powder and
potassium permanganate powder.
Sand mixing was used by the type SHY sand mixer. NBRBS were
fabricated manually using a standard “8” font mold with the dimen-
sion of 22.36 mm × 22.36 mm × 66 mm. The strength tests were
measured by the type SWG lever-type universal strength testing

https://doi.org/10.1016/j.jmapro.2017.10.002
1526-6125/© 2017 The Society of Manufacturing Engineers. Published by Elsevier Ltd. All rights reserved.
314 F. Liu et al. / Journal of Manufacturing Processes 30 (2017) 313–319

Table 1
The chemical composition of different molding sand.

Base sand Chemical composition(wt%)

SiO2 Al2 O3 Fe2 O3 Na2 O + K2 O TiO2

Duchang sand ≥92.13 <4.50 <0.60 <2.50 <0.60

Dalin sand ≥91 <4.50 <0.30 <3.00 –
Ceramacore 5–20 75–85 ≤5 – ≤1.5

Table 2
The characteristics of SQG700 furan resin and GC09 curing agent.

Viscosity (25 ◦ C, mPa s) Density (25 ◦ C g/cm3 ) Nitrogen content (%) Free formaldehyde (F.F.%) Total acid (%) Free acid (%)

SQG700 ≤ 75 1.15–1.2 ≤7 ≤0.05 – –

GC09 60–80 1.2–1.4 – – 24.5–27.5 2.5–4.5

Fig. 2. The effect of furan resin content and curing time on tensile strength of NFBRS.

Fig. 1. Simplified flowchart of process.

machine, and the weight was tested by the type JA5003N electronic
balance with the accurancy of 0.001 g. The gas evolution tests were
measured by the type GET-III intelligent gas evolution tester. The
muffle furnace was used to heat NBRBS. SEM analysis was used by
Quanta 200 environmental scanning electron microscope (ESEM).

2.2. Methods

Fig. 1 shows the simplified flowchart of process. Firstly, base

sand and curing agent were mixed uniformly with a certain ratio,
and then additives and furan resin were added to the mixed sand
successively using mechanical mixing to form uniform mould-
ing sand. Afterwards, the mixed sand poured into the mold and
allowed to cure at room temperature. After a certain time, NBRBS
was demolded and stored in the air for several hours. Finally, the
performance of NBRBS were measured in this paper.
The tensile strength of NBRBS being cured in the air after sev-
eral hours was tested. Retained strength was the tensile strength
of NBRBS after being put into a muffle furnace for a half hour at
different temperature. Gas evolution (G) was tested by the type
GET-III intelligent gas evolution tester. All the testing results were
the average value of five measurements.
Fig. 3. The effect of furan resin content and temperature on retained strength of
NFBRS.
3. Results
3.1. The effects of furan resin content
When the curing agent addition was 50 wt% of furan resin con-
tent, the effects of furan resin content on tensile strength, retained
strength and gas evolution of NBRBS have been studied, and listed
in Figs. 2 and 3, Table 3, respectively. Duchang sand was used as
base sand.
 F. Liu et al. / Journal of Manufacturing Processes 30 (2017) 313–319
Table 3
The effect of furan resin, curing agent and boric acid content on gas evolution of NFBRS.
Types Resin content (%)
1.2 1.6 2.0
Gas evolution (ml/g) 10.1 13.1 18.7
Fig. 4. The effect of curing agent content on tensile strength of NFBRS.
As shown in Fig. 2, the tensile strength of NBRBS increases with
the increased resin content, and the 48 h tensile strength is less than
the 24 h tensile strength for its moisture absorption [1,2]. When the
resin content is 1.6 wt% of base sand, the 6 h and 24 h strength are
above 0.3 MPa that satisfies production requirement of magnesium
alloy castings, and while the content is 1.2 wt% with low strength
and 2.0 wt% with high strength.
As shown in Fig. 3, at the same temperature, the retained
strength increases with the increased resin content. When the resin
content is constant, the retained strength firstly increases then
decreases with the rised temperature, and the retained strength
is maximum at 200 ◦ C.The reason is that the low roasting temper-
ature can make NBRBS secondary hardening which increases the
retained strength [1,2], and the high roasting temperature can burn
the bonding bridge of NBRBS which reduces the retained strength.
As shown in Table 3, the gas evolution of NBRBS increases with
the furan resin content increases. The reason is pyrolysis of furan
resin.
Therefore, the furan resin content has important influence on
the room tensile, retained strength and gas evolution of NBRBS.
During actual production process, it is better for the lower
retained strength and gas evolution, but the tensile strength should
be above 0.3 MPa which can withstand the impact of liquid metal.
Therefore, 1.6 wt% of sand should be selected as the optimized resin
content. And so the addition of furan resin is selected 1.6 wt% of base
sand for the subsequent experiments in this paper.
3.2. The effects of curing agent content
The effects of curing agent addition on tensile strength, retained
strength and gas evolution of NBRBS have been researched, and
listed in Figs. 4 and 5, Table 3, respectively. Duchang sand was used
as base sand.
As can be seen, the tensile strength of NBRBS increases firstly
and then becomes constant with the increase of curing agent con-
tent and time, while the content of curing agent is 40 wt%, the
final strength is lower than that with the curing agent content of
50 wt% and 60 wt%. The reason is that when the resin content is
315
Curing agent content (%) Boric acid content (%)
40 50 60 0 1 2 3
11.2 13.1 16.1 13.1 14.2 16.1 17.7
Fig. 5. The effect of curing agent content on retained strength of NFBRS.
constant, the more curing agent content, the faster curing speed,
and so the strength increses quickly but the final strength has no
distinct changes. Therefore, the content of curing agent should be
suitable to obtain good strength.
At the same temperature, the retained strength has little change
with the increased curing agent content. When the curing agent
is constant, the retained strength firstly increases then decreases
with the rised temperature, and the retained strength is maximum
at 200 ◦ C, the results is similar to the effect of furan resin content.
Therefore, the curing agent content has little important influence
on the final tensile strength and retained strength of NBRBS, but
it can increase the curing speed remarkablely with higher curing
agent.
As shown in Table 3, the gas evolution of NBRBS increases with
the curing agent content increases. The reason is pyrolysis of curing
agent.
During actual production process, it is better for the lower
retained strength and gas evolution, but the tensile strength should
be above 0.3 MPa which can withstand the impact of liquid metal.
And so 50 wt% of resin content should be selected as the optimized
curing agent content.
As a result, the optimized basic formula is that furan resin con-
tent is 1.6 wt% of base sand and the curing agent addition is 50 wt%
of furan resin content.
3.3. The effects of boric acid content
During sand casting of magnesium alloy castings, flame retar-
dant should be added in the molding sand in order to prevent burn
of magnesium alloy. In this paper, boric acid was selected as flame
retardant, and so the effects of boric acid content on the perfor-
mance of NBRBS were studied based on the above optimized basic
formula, and listed in Figs. 6 and 7, Table 3, respectively. Duchang
sand was selected as base sand.
As can be seen from Fig. 6, the final tensile strength of NBRBS
decreases with the increase of boric acid content. The reason is that
curing speed is improved by adding boric acid which can accelerate
curing of furan resin, and so the final strength reduces [2,12,13].
316 F. Liu et al. / Journal of Manufacturing Processes 30 (2017) 313–319
Fig. 6. The effect of boric acid content on tensile strength of NFBRS.
Fig. 7. The effect of boric acid content on retained strength of NFBRS.
As shown in Fig. 7, the retained strength highly increases with
the increased boric acid content. The reason is that boric acid
can promote the roasting of NBRBS and so improve the retained
strength.
As shown in Table 3, the gas evolution of NBRBS increases with
the boric acid content increases. The reason is pyrolysis of boric
acid.
Therefore, the use of boric acid increases the curing speed,
retained strength and gas evolution of NBRBS. The boric acid
content should be low under the premise of ensuring the flame
retardant.
3.4. The effects of collapsibility agent
In order to solve poor collapsibility of NBRBS which makes mag-
nesium alloy castings difficult to be clean after casting, the effects of
different collapsibility agents on the strength of NBRBS were inves-
tigated based on the above optimized basic formula. The results are
listed in Table 4. Duchang sand was selected as base sand.
As can be seen from Table 4, the tensile strength firstly increases
and then decreases with the increased temperature. The use of
KNO3 or NaHCO3 has no obvious decrease on the retained strength
of NBRBS, and NaHCO3 substantially reduces the 24 h tensile
strength. The use of KMnO4 decreases the retained and tensile
Table 4
The effects of different collapsibility agents on the strength of NFBRS.
Kinds of NFBRS Tensile strength(MPa)
24h 100 ◦ C 200 ◦ C 300 ◦ C
Basic formula 0.89 1.13 1.2 0.52
Basic formula + 1% KNO3 0.81 0.89 1.02 0.46
Basic formula + 1% NaHCO3 0.17 0.81 0.63 0.48
Basic formula + 0.5% KI 0.77 0.83 1.01 0.32
Basic formula + 1% KMnO4 0.66 1.04 0.93 0.41
strength of NBRBS. The use of KI reduces the retained strength of
NBRBS with high tensile strength. The reason for decreasing the
retained strength of NBRBS caused by adding KMnO4 and KI is that
pyrolysis of KMnO4 and KI, and that can release strong oxidizing
atmosphere which accelerates thermal decomposition of resin [13].
Therefore, KI is selected as a suitable collapsibility agent for
NBRBS used in magnesium alloy casting production.
3.5. The effects of base sand
Recent research demonstrates that different base sand has an
important influence on the performance of moulding sand [14]. In
this paper, the effects of different base sand on the performance
of NBRBS were studied, and listed in Tables 5 and 6, respectively.
The kinds of base sand were Duchang sand, Dalin sand and Cerama-
core. The surface morphology of NBRBS was shown in Figs. 8 and 9,
respectively.
As shown in Tables 5 and 6, the tensile strength and retained
strength of the Dalin sand samples are much higher than that of the
Duchang sand samples, and the gas evolution is similar. When the
resin content is 1 wt% of ceramacore and the curing agent addition
is 50 wt% of furan resin content, the tensile strength is higher than
that of the Duchang sand samples prepared by the above optimized
basic formula which consists of 1.6% resin content of base sand and
50% curing agent addition weight of furan resin content, and the
retained strength is similar, the gas evolution is only 7.2 ml/g which
is much less than that of the Duchang sand samples.
Therefore, the use of ceramacore can reduce largely the furan
resin content, improve the strength of NBRBS, decrease the retained
strength and improve the collapsibility of NBRBS.
The following conclusions can be seen from Figs. 8 and 9.
(1) The Duchang sand particles shape is not round, appears half
angle with larger angle coefficient and has uneven surface with
many edges. The Dalin sand particles shape is also not round,
has some edges which is more round than that of the Duchang
sand and with less edges. The ceramacore particles shape is
round, appeares standard sphere with angle coefficient of 1,
has even surface without edges and corners.
(2) The thickness of bonding film of ceramacore samples is about
8 ␮m with the Duchang sand samples of 5 ␮m and the Dalin
sand samples of 3 ␮m. The quantity of bonding bridge of the
Duchang sand samples, Dalin sand samples and ceramacore
samples is 2, 3 and 5, respectively, which are obtained and
calculated from Fig. 9.
(3) The tensile strength of the Dalin sand samples are much higher
than that of the Duchang sand samples due to the Dalin sand
particles shape is more round than that of the Duchang sand
and with less edges.
(4) Due to roundness and small angle coefficient of the cerama-
core particles, and the ceramacore can form complete bonding
film and bonding bridge which reduces largely the resin con-
tent, improves the strength, decreases the retained strength and
improves the collapsibility of NBRBS.
 F. Liu et al. / Journal of Manufacturing Processes 30 (2017) 313–319 317

Table 5
The effect of different base sand on room properties of NFBRS.

Base sand Resin content (weight of sand,%) Curing agent content (weight of resin,%) Tensile strength (MPa)

4h 6h 24h 48h

Duchang 1.6 50 0.25 0.53 0.84 0.81

Dalin 1.6 50 0.53 1.32 1.92 1.89
Ceramacore 1 50 0.41 0.79 0.91 0.88

Table 6
The effect of base sand on high temperature properties of NFBRS.

Base sand Tensile strength (MPa) Gas evolution(ml/g)

24 h 100 ◦ C 200 ◦ C 300 ◦ C 400 ◦ C

Duchang 0.84 1.03 1.21 0.64 0.05 13.1

Dalin 1.92 1.9 2.04 1.20 0.09 13.4
Ceramacore 0.91 1.4 1.05 0.55 0.07 7.2

Fig. 8. The surface morphology of NFBRS (50 times microscope observations).

Fig. 9. The surface morphology of NFBRS (200 times microscope observations).

4. Disscussion of moulding sand is intrinsic failure of bonding film by the action

According to surface morphology of NBRBS shown in
Figs. 8 and 9, if sand particles are round and its accumulation is sim-
ple cubic, the bonding bridge diagram of adjacent moulding sand
can be shown in Fig. 10.
In Fig. 10, the radius of the sand particles is R0 , the thickness of
bonding film which coates the sand particles is d0 . If the destruction
static tensile and the binder is evenly distributed on the surface
of sand particles, the tensile strength of moulding sand () can be
listed in Eq. (1) [2,15].
 = N · n · Sm (1)
Where N (mm−2 ) is the quantity of bonding bridge on the unit cross-
sectional area of moulding sand,  n (MPa) is cohesion strength of
318 F. Liu et al. / Journal of Manufacturing Processes 30 (2017) 313–319

Fig. 10. Bonding bridge diagram of adjacent moulding sand.

bonding bridge, and Sm (mm2 ) is minimum cross-sectional area of
bonding bridge between adjacent sand particles.
When the other conditions is the same without binder content,
Eq. (2) can be used to describe relation between cohesion strength
of bonding bridge and the thickness of bonding film.
n = K1 · d0
Where K1 is relationship constant which is constant greater than
zero.
According to Eqs. (1) and (2), when the binder content increases,
the thickness of bonding film and cohesion strength of bonding
bridge both increase, and so the tensile strength of moulding sand
increases, and it agrees with the above experimental results.
When the binder content is the constant, the thickness of bond-
ing film (d0 ) and the quantity of bonding bridge (N) both increase
with specific surface area of base sand decreases, in our current
study, specific surface area of ceramacore is minimum and specific
surface area of Duchang sand is maximum, therefore the tensile
strength of ceramacore moulding sand is higher than that of the
Dalin and Duchang moulding sand, which means that ceramacore
can obtain better tensile strength with less resin content, and at the
same time reduces gas evolution. And it is also in accordance with
the above experimental results.
5. Conclusion
(1) The effects of furan resin and curing agent addition on ten-
sile strength, retained strength and gas evolution of NBRBS were
investigated, the optimized formula which satisfies production
requirement of magnesium alloy castings and collapsibility has
been determined. The optimized formula is that the amount of
furan resin is 1.6 wt% of base sand and the amount of curing agent
is 50 wt% of furan resin content.
(2) Based on the above optimized basic formula, the effects of
boric acid content on the performances of NBRBS were studied.
The results show that the room tensile strength of NBRBS increases
firstly and decreases with the boric acid content increases. At the
same temperature, the retained strength highly increases with the
increased boric acid content. The gas evolution of NBRBS increases
with the boric acid content increases. Therefore, the boric acid
content should be less under the premise of ensuring the flame
retardant.
(3) Based on the above optimized basic formula, the effects
of different collapsibility agents on the strength of NBRBS were
investigated. The results show that the use of KMnO4 decreases
the retained and tensile strength of NBRBS, the use of KI reduces
the retained strength of NBRBS with a good 24 h tensile strength.
The reason for decreasing the retained strength of NBRBS is that
pyrolysis of KMnO4 and KI, and that can release strong oxidizing
atmosphere which will accelerate thermal decomposition of resin.
Therefore, KI can be selected as a suitable collapsibility agent for
(2) NBRBS used in magnesium alloy casting production.
(4) The effects of kinds of base sand on the performance of NBRBS
were studied. The results show that when the resin and curing
agent addition are the same, the room tensile strength and retained
strength of the Dalin sand samples are much higher than that of the
Duchang sand samples, and the gas evolution is similar.
When the resin content is 1 wt% of ceramacore and the curing
agent addition is 50 wt% of furan resin content, the room tensile
strength is higher than that of the Duchang sand samples prepared
by the optimized formula, and the retained strength is similar,
the gas evolution is only 7.2 ml/g which is much less than that
of the Duchang sand samples. Therefore, the use of ceramacore
reduces largely the resin content, improves the strength, decreases
the retained strength and so improves the collapsibility of NBRBS.
(5) The surface morphology of NBRBS were investigated by SEM.
The results show the Duchang sand particles shape is not round,
larger angle coefficient and has uneven surface with many edges,
the Dalin sand particles shape is also not round with less edges, the
ceramacore particles shape is roundness, appeares standard sphere
with angle coefficient of 1, has even surface without edges and cor-
ners. Therefore, the ceramacore can form complete bonding film
and bonding bridge which could reduce largely the resin content,
improve the strength, decrease the retained strength and improve
the collapsibility of NBRBS.
Acknowledgements
The authors would like to thank the support of the
National Nature Science Foudation of China (No.51375187 and
No.51775204), the Project of China Postdoctoral Science Founda-
tion, the Research Project of State Key Laboratory of Materials
Processing and Die & Mould Technology and the Analytical and
Testing Center, HUST.
References
[1] Ding Wenjiang. Science and technology of magnesium alloys. Beijing: Science
Press; 2007 (in Chinese).
[2] Huang Tianyou. Casting manual- modeling material. Beijing: China Machine
Press; 2012 (in Chinese).
 F. Liu et al. / Journal of Manufacturing Processes 30 (2017) 313–319
[3] Khandelwal H, Ravi B. Effect of binder composition on the shrinkage of chemi-
cally bonded sand cores. Mater Manuf Process 2015;30(12):1465–70.
[4] Dobosz SM, Major-Gabryś K. Strength properties of moulding sands with cho-
sen biopolymer binders. Arch Foundry Eng 2010;10(3):17–20.
[5] Khandelwal H, Ravi B. Effect of molding parameters on chemically bonded sand
mold properties. J Manuf Process 2016;22:127–33.
[6] Stachowicz M, Granat K, Nowak D, Haimann K, et al. Effect of hardening meth-
ods of moulding sands with water glass on structure of bonding bridges. Arch
Foundry Eng 2013;10(3):123–8.
[7] Surekha B, Rao DH, Mohan Rao GK, Pandu Vundavilli R, Parappagoudar MB,
et al. Prediction of resin bonded sand core properties using fuzzy logic. J Intell
Fuzzy Syst 2013;25(3):595–604.
[8] Caylak I, Mahnken R. Thermomechanical characterisation of cold box sand
including optical measurements. Int J Cast Met Res 2010;23(3):176–84.
[9] Chen Yunnan. Resin sand performance testing technology. Modern Cast Iron
2012;1:85–8.
319
[10] Wang Xun. Research on the modification of furan no-bake foundry binders.
Shandong University of Science and Technology; 2006 (in Chinese).
[11] Surekha B, Rao DH, Rao GKM, Pandu Vundavilli R, et al. Design and development
of knowledge base scheme for chromite-based resin bonded sand core system.
Procedia Mater Sci 2014;6:919–25.
[12] Chang, Furan no-bake foundry binders and their use [P]. United States
Patent:6593397, 2002.
[13] Kang Ming, Lu Dingquan, Ma Chaoyang. Research on improving the collapsi-
bility of phenol-formaldehyde resin coated sand. Foundry 2001;50(6):337–41
(in Chinese).
[14] Li Weihua, Dong Xuanpu, Li Jianjun, Wu Xiangqing, et al. Application of cerama-
core to ester cured water glass sand. Foundry 2011;60(5):481–5 (in Chinese).
[15] Yang Jing, Li Ouqing, Cheng Jun. Mathematical model of relationship between
the resin addition levels and tensile strength of moulding sand in PEPSET pro-
cess. Foundry 1997;12:13–7 (in Chinese).