See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/47451273

Emission of organic compounds from mould and core binders used for casting
iron, aluminium and bronze in sand moulds

Article  in  Journal of Environmental Science and Health Part A Toxic/Hazardous Substances & Environmental Engineering · December 2010
DOI: 10.1080/10934529.2010.520595 · Source: PubMed

CITATIONS READS

19 522

4 authors, including:

Niels Skat Tiedje Niki Bey

Technical University of Denmark Technical University of Denmark
65 PUBLICATIONS   397 CITATIONS    49 PUBLICATIONS   381 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Continuous casting View project

AREUS - Automation and Robotics for EUropean Sustainable manufacturing View project

All content following this page was uploaded by Niels Skat Tiedje on 01 August 2016.

The user has requested enhancement of the downloaded file.

This article was downloaded by: [Technical Knowledge Center of Denmark (DTIC)]
On: 4 November 2010
Access details: Access Details: [subscription number 912941598]
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

Journal of Environmental Science and Health, Part A

Publication details, including instructions for authors and subscription information:
http://www.informaworld.com/smpp/title~content=t713597268

Emission of organic compounds from mould and core binders used for
casting iron, aluminium and bronze in sand moulds
Niels Tiedjea; Rudolf Crepazb; Torben Eggertb; Niki Beyc
a
Department of Mechanical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark b
Danish Technological Institute, Høje Taastrup, Denmark c Institute for Product Development, Kgs.
Lyngby, Denmark

Online publication date: 13 October 2010

To cite this Article Tiedje, Niels , Crepaz, Rudolf , Eggert, Torben and Bey, Niki(2010) 'Emission of organic compounds
from mould and core binders used for casting iron, aluminium and bronze in sand moulds', Journal of Environmental
Science and Health, Part A, 45: 14, 1866 — 1876
To link to this Article: DOI: 10.1080/10934529.2010.520595
URL: http://dx.doi.org/10.1080/10934529.2010.520595

PLEASE SCROLL DOWN FOR ARTICLE

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial or
systematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply or
distribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contents
will be complete or accurate or up to date. The accuracy of any instructions, formulae and drug doses
should be independently verified with primary sources. The publisher shall not be liable for any loss,
actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directly
or indirectly in connection with or arising out of the use of this material.
 Journal of Environmental Science and Health Part A (2010) 45, 1866–1876
Copyright
C Taylor & Francis Group, LLC

ISSN: 1093-4529 (Print); 1532-4117 (Online)

DOI: 10.1080/10934529.2010.520595

Emission of organic compounds from mould and core

binders used for casting iron, aluminium and bronze in sand
moulds

NIELS TIEDJE1 , RUDOLF CREPAZ2 , TORBEN EGGERT2 and NIKI BEY3

1
Department of Mechanical Engineering, Technical University of Denmark, Kgs. Lyngby, Denmark
2
Danish Technological Institute, Høje Taastrup, Denmark
3
Institute for Product Development, Kgs. Lyngby, Denmark
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010

Emissions from mould and core sand binders commonly used in the foundry industry have been investigated. Degradation of three
different types of binders was investigated: Furfuryl alcohol (FA), phenolic urethane (PU) and resol-CO2 (RC). In each group of
binders, at least two different binder compositions were tested. A test method that provides uniform test conditions is described.
The method can be used as a general test method to analyse off gasses from binders. Moulds, containing a standard size casting,
were produced and the amount and type of organic compounds, resulting from thermal degradation of binders, was monitored when
cast iron, bronze and aluminium was poured in the moulds. Binder degradation was measured by collecting off gasses in a specially
designed ventilation hood at a constant flow rate. Samples were taken from the ventilation system and analysed for hydrocarbons and
CO content. It is shown how off-gasses vary with time after pouring and shake out. Also the composition of off-gasses is analysed and
shown. It is further shown how the composition of off-gasses varies between different types of binders and with varying composition
of the binders as well as function of the thermal load on the moulding sand.
Keywords: Binder, emission, casting, degradation, mould production.

Introduction sand (green sand). When hollow castings are produced,

Production of castings in sand moulds is the most wide
spread route to production of cast products. The majority
of all castings and almost all types of alloys are cast in
sand moulds.[1] Large castings in cast iron and Cu-based
alloys, as well as many aluminium castings are commonly
produced in sand moulds where the sand is bonded us-
ing organic binders.[2] Only very few binder systems exist
that contain negligible or no organic materials, and these
systems are not suited for modern high productivity manu-
facturing of castings.[2] The most commonly used binder
systems are: the so-called furan binders (based on fur-
furyl alcohol, FA), Cold box binders (based on phenolic
urethane, PU) and resol-CO2 , RC, binders. The most com-
monly used binder system for mass production of small and
medium size cast iron and copper castings is clay-bonded
Address correspondence to Niels Skat Tiedje, Department
of Mechanical Engineering, Technical University of Denmark,
Building 425, DK-2800 Kgs. Lyngby, Denmark; E-mail: nsti@
mek.dtu.dk
Received March 26, 2010.
the cores that shape the internal geometry are produced
with one or more of the organic binder systems mentioned
above. Large castings weighing more than approximately
100 kg are commonly produced in furan moulds.
Over the past 10 years new binders have been developed
that improve the working environment on the moulding
line and in the core shop. Serious improvements have been
achieved in the later years with regard to toxicity and smell
of binders in the part of the foundry where moulds and
cores are produced.[3−7] However, it is not clear from these
investigations whether the use of new binder systems have
reduced the type and level of emissions in the pouring,
cooling or shake out areas in the foundry.
Work has been done to describe the pyrolysis products
arising from the thermal degradation of binders under lab-
oratory conditions.[8−10] Typically, sand mixed with binders
has been heated to a specific temperature in a laboratory
furnace and the off-gasses have been analysed. But these
investigations do not describe the off-gasses and concen-
tration levels of pyrolysis products found in foundries. Here
evaporation, pyrolysis and condensation of organic mate-
rial in the mould takes place at all temperatures ranging
 Organic compound emissions from sand mould metals
from room temperature to the temperature of the metal
poured into the mould. Further, very few of the investiga-
tions provide information on how pyrolysis products de-
velop over time when the heat from the casting penetrates
into the mould.
The investigations have focused on individual combina-
tions of binder systems and cast metals. It has not been
possible to find studies that compare several binder sys-
tems and metals, and where both total emissions and evo-
lution of degradation products over time is analysed under
conditions resembling those found in the foundry.
From previous investigations a few general conclusions
can be drawn: Serebro et al. finds that water evaporates
from core sand at 130 to 140◦ C and that the organic ma-
terial degenerates from 350 to 480◦ C.[9] This is confirmed
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010
by Wang et al.[11] who compares three phenolic urethane
resins and among other show how binders degrade at dif-
ferent temperatures. They describe and quantify the degra-
dation products of the binders they investigate. And they
find that emissions are very different from core shop to pro-
duction line. They also conclude that more comprehensive
studies are necessary to compare the many binder systems
available.
Kauffmann and Voigt[12] mention that emission of ben-
zene is reduced when the sand to metal ratio is large, in-
dicating that off-gasses condense in the cold areas of the
mould. But there is no detailed description of the chemical
processes that take place there.
Only very few investigations distinguish between emis-
sion of fumes and toxic gasses in the different parts of the
cooling line, such as: in the pouring area where liquid metal
is cast in the moulds, in the cooling area where the castings
sit in the moulds while they cool before they are extracted,
and at the shake out where the moulds are opened.[12−14]
In a study by Westberg et al.[14] it was found that emis-
sions are very different from core shop to production
line. They too conclude that more comprehensive studies
are necessary to compare the many binder systems avail-
able. It has generally been assumed that when the binders
are kinder to the environment in the core shop or when
moulds are made, they would also improve the environ-
ment in the casting zone but this has not been verified in
practice.
At present there are no standard methods for mea-
suring emissions from casting processes. Some methods
have been proposed in individual projects, but they have
not found widespread recognition.[15,16] In the present in-
vestigation a method for measuring emissions from sand
casting processes is proposed and used to test emissions
from a range of commonly used binder systems and al-
loys. Emissions from three types of binder systems are
analysed when cast iron, copper and aluminium alloys are
cast in the moulds. Emissions are monitored as function of
time during cooling and shakeout, and total emissions are
measured.
1867
Experimental setup
Binders
Combinations of alloys and binder systems that are com-
monly used in foundries were selected for the experiments.
It was not possible to test all binders on the market, but
some commonly used binders were selected to represent the
following types of binders:
1. types of furan binders (FA) with two different catalysts
2. types of cold-box (PU) binders
3. types of resol-CO2 (RC) binders
Table 1 shows the chemical composition of the binders
based on information from data sheets provided by the sup-
pliers. Table 2 lists the individual experiments and shows
experiment number, binder type, alloy cast, pouring tem-
perature, sand to metal ratio (kg sand per kg metal) and
the content of binder and other additives in the moulds.
The Table also shows the time during experiments where
samples were taken for chemical analysis.
Fume analysis
The key point in the analysis is to make sure that all
fumes are properly extracted from the moulds, and that
the gasses extracted are not contaminated by foreign pro-
cesses. For that purpose, a special exhaust hood was de-
signed that covers the mould completely, but allows flow
of fresh air through the ventilation system. The hood has
a door through which the freshly poured mould can be
placed under the hood; see the sketch in Figure 1. The
moulds were cast next to the exhaust hood and inserted
immediately. The hood was connected to a ventilation duct
with a ventilator (at the end) providing a constant airflow.
Along the duct, before the fan, samples were extracted for
chemical analysis.
In most of the experiments the moulds were kept un-
der the hood for 30 min, while samples were taken every
5 minutes. Though, in experiments 5 and 12 (see Table 2)
the moulds remained under the hood for 90 minutes to
investigate the extent of fume emission over longer time.
In experiments 4 and 11 the emissions during shakeout
only, was measured. In all experiments the extracted fumes
were analysed for CO and total hydrocarbons using a CO-
meter and photo- and flame ionisation detectors. Thus a
Fig. 1. Experimental setup for measurements of emissions from
sand castings.
 1868 Tiedje et al.
Table 1. Binder systems and their chemical composition according to the supplier’s safety sheets.

Binder code Binder type Composition according to supplier [wt%]

A Furan, FA Binder:
Furfuryl alcohol 50–55
Nitrogen <0.7
Free Phenol <1
Free Formaldehyde 0.2–0.5
Catalyst:
Para-Toluenesulfonic acid (PTA)
B Binder:
Furfuryl alcohol 50–55
Furan, FA Nitrogen <0.7
Free Phenol <1
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010

Free Formaldehyde 0.2–0.5

Catalyst:
Benzenesulfonic acid (BSA)
C Furan, FA Binder:
Furfuryl alcohol 75
Nitrogen <5
Free Phenol <0.2
Water <8
Catalyst:
Para-Toluenesulfonic acid (PTA)
D Cold Box, PU Part I:
Phenol 2–5
3,5,5-trimethyl-2-cyklohexene-1-on 3–10
Aromatic Carbohydrates C6 ,-C14 18–30
Formaldehyde <0.3
Part II:
Diphenylmethane-4-41 -diisocyanate >0.7
Solvent naphtha 10–30
E Part I:
Phenol 2.5
Cold Box, PU Formaldehyde <0.6
Methanol <0.25
Part II:
Diphenylmethane-4-41 -diisocyanate 70–90
Solvent naphtha 10–25
F Cold Box, PU Part I:
Phenol <5
Formaldehyde <1
Solvent naphtha 30–35
N-butanol 10
Part II:
Diphenylmethane-4-41 -diisocyanate 70–80
Aromatic Carbohydrates 15–20
Aliphatic Carbohydrates 5–10
G Sodium hydroxyde 10–20
Resole/CO2 , Phenol <1.5
RC Formaldehyde <0.2
Ethylene glycol <5
1-hydroxy-3.6-dioxane 5–10
H Sodium hydroxyde 14–18
Resole/CO2 , Butyl glycol 4–7
RC 2-fenoxiethanol <3
Phenol <0.3
 Organic compound emissions from sand mould metals 1869
Table 2. Expermental plan.

Samples taken

Experiment Binder Pouring temperature Sand – metal Binder content During cooling During
number code Alloy [◦ C] ratio [kg/kg] [wt %] [min] shake-out

1 A Fe 1400–1420 3:1 1.0/0.4 0–30 –

2 A Cu 1180–1200 2.8:1 1.0/0.4 0–30 –
3 A Al 700–720 8.1:1 1.0/0.4 0–30 –
4 A Fe 1400–1420 3:1 1.0/0.4 0–30 Yes
5 A Fe 1400–1420 3:1 1.0/0.4 30–90 –
6 B Fe 1400–1420 3:1 1.0/0.4 0–30 –
7 C Fe 1400–1420 3:1 1.0/0.4 0–30 –
8 D Fe 1400–1420 3:1 0.8/0.8 0–30 –
9 D Cu 1180–1200 2.8:1 0.8/0.8 0–30 –
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010

10 D Al 700–720 8.1:1 0.8/0.8 0–30 –

11 D Fe 1400–1420 3:1 0.8/0.8 0–30 Yes
12 D Fe 1400–1420 3:1 0.8/0.8 30–90 –
13 E Fe 1400–1420 3:1 0.8/0.8 0–30 –
14 F Fe 1400–1420 3:1 0.8/0.8 0–30 –
15 G Fe 1400–1420 3:1 2.5 0–30 –
16 H Fe 1400–1420 3:1 2.5 0–30 –
The table shows binder code, alloy used, pouring temperature, sand-metal ratio (kg sand pr. kg metal) and binder content (wt%). Alloy type codes
are: Fe – grey cast iron (EN-GJL 200), Cu – Sn bronze (EN CuSn6), Al – aluminium alloy (EN ABSi12). Where the binder content is given as two
figures (X.X / Y.Y) it relates to the ration of binder to catalyst for Furan binders or Part I and Part II for Cold box binders.

time-dependent analysis of the fume composition was ob-
tained. These data are presented in Figures 2, 3, 5, 6, 8 and
9. Pyrolysis products were also extracted on filters for the
full duration of the experiments. These were analysed in the
laboratory to give a more detailed description of the types
of hydrocarbons emitted. The data gave detailed, accumu-
lated distribution of pyrolysis products for the duration
of the experiments. In Figures 4, 7 and 11 these data are
showed in main groups: Aromatic hydrocarbons, phenols
and aldehydes. More detailed compositions are presented
in Tables 3, 4 and 5. With this combination of methods it
was possible to compare the more detailed analysis of prod-
ucts found on the filters with the on-line measurements of
CO and total hydrocarbons.
In all experiments the air velocity in the duct was
4.9 m/s (± 0.1 m/s), which corresponds to 140 m3 /h. The
air temperature was monitored; it varied between 20 and
30◦ C, though in experiments 5 and 12, it shortly rose to
40◦ C immediately after shake out.
Moulding and casting
The casting geometry used in the experiments was a block
of 125 by 125 by 50 mm fitted with a suitable gating system.
The mould was designed to give a sand-to-metal ratio for
cast iron of 3:1. For aluminium and bronze alloys the sand
to metal ratio was 8.1:1 and 2.8:1, respectively.
Binder content for all batches is given in Table 2. Cast-
ing alloys were melted in an induction furnace. The alloys
that were used for the experiments were grey cast iron (EN-
GJL 200), tin-bronze (EN CuSn6) and an aluminium alloy
(EN ABSi12). These alloys were chosen because they rep-
resent alloys that are normally used in the industry with
the binders analysed. The melt was transferred to a 1.5 L
hand-held ceramic ladle, which was used for pouring. Cast
iron was poured at 1400–1420◦ C, bronze at 1180–1200◦ C
and aluminium at 700–720◦ C. The pouring temperatures
(Table 2) are temperatures that are normally used for cast-
ing industrial components so that the thermal load of the
moulds is similar to that of an industrial environment.
Results
Emissions from Furan (FA) binders
Two different FA binders were tested. Binder A was cured
with para toluenesulfonic acid (PTA) and castings were
made with cast iron, aluminium and bronze, see Table 2
experiments 1 to 3. In these experiments emissions were
measured for 30 min.
An extra experiment was performed in which emissions
during shake-out, after the casting had been sitting in the
mould for 30 min, were measured, experiment 4 in Table 2.
In an additional experiment emissions were monitored for
90 minutes, experiment 5 in Table 2. Binder “B” is equal
to “A”, but was cured with benzenesulfonic (BSA) acid.
One experiment was made with this binder, experiment 6 in
Table 2. Cast iron was poured in the mould and emissions
were measured for 30 minutes.
 1870
Fig. 2. Emission of hydrocarbons as a function of time for FA-
bonded sand.
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010
As an alternative, FA binder “C” cured with PTA was
chosen; see experiment 7 in Table 2. Emission of hydro-
carbons and CO is shown in Figures 2 and 3, respectively.
When iron is cast in the moulds, the concentration of hy-
dro carbons in the off-gasses reaches the maximum value
(250 to 350 ppm) within 1 to 2 minutes after pouring. From
thereon, the concentration decreases continuously. After
30 minutes it has fallen to about 10% of the maximum.
Over the following hour the concentration of hydrocar-
bons falls to a level that cannot be measured, cf. experiment
5 in Figure 2.
The metals with lower pouring temperature (Cu- and
Al-alloys) have less thermal impact on the mould and it is
seen that the initial level of emissions of hydro carbons is
relatively low but it reaches a peak after 4 minutes and 6
minutes, respectively for Cu- and Al-alloys. For CO emis-
sions, the picture is similar, only the maximum concentra-
tion is higher varying from approximately 60 and 30 ppm,
respectively for the two metals. So, when the thermal load
on the mould is reduced, the emission of CO is also reduced
considerably. Total emissions of CO from experiment 6 is
less than in the other experiments where iron is cast in the
mould. It indicates that furan binders cured with BTS emit
less hydrocarbons and CO than is the case when PTA is
used.
Fig. 3. Emission of CO as a function of time for FA-bonded sand.
Tiedje et al.
Fig. 4. Pyrolysis products from FA-bonded sand as a function of
time.
The detailed results of the analysis of pyrolysis prod-
ucts collected on filters for the FA binders are shown in
Table 3, which Table shows the amount of each pyrolysis
product collected during the experiment. Data are collected
in groups of related pyrolysis products: Hydrocarbons, phe-
nols, aldehydes and CO.
It is clear that there is very little difference between emis-
sions from binder A and C. However, the composition
of the off-gasses changes considerably by changing cur-
ing agent. The PTA cured binder primarily emits toluene
and some benzene while the BTA cured binder primarily
emits benzene. In the latter, emission of phenol and cresol
is increased tenfold, but emissions are minute compared to
benzene emissions.
The data in the table shows that in all experiments there
are two components that constitute that largest part of the
total emission, namely benzene and CO. Figure 4 shows
that emissions are highest when iron is cast in the moulds,
but there is quite a difference in the emissions between the
types of binders used in the experiments. The binders used
in experiments 1 and 7 develop less benzene than the binder
used in experiment 6.
Emissions from cold-box (PU) binders
A series of experiments similar to those for the FA binders
were conducted for PU binders. In experiments 8, 9 and 10
castings were made with cast iron, bronze and aluminum,
respectively.
The results are shown in Figures 5, 6 and 7. All data from
the filter analysis are shown in Table 4. From Figures 5 and
6 we see that there is some delay in the breakdown of PU
binders. The maximum level of emissions is not reached
until about 20 min after pouring. After 30 min, where most
 Organic compound emissions from sand mould metals 1871
Table 3. Detailed analysis of off-gasses from moulds made with FA binders.

Experiments

4-A-Fe 5-A-Fe
1-A-Fe 2-A-Cu 3-A-Al Shake out 90 min 6-B-Fe 7-C-Fe
Pyrolysis products ↓ [mg/kg] [mg/kg] [mg/kg] [mg/kg] [mg/kg] [mg/kg] [mg/kg]

Aromatic Benzene 32.9 17 1.2 1.2 2.2 78 27

hydrocarbons
Toluene 63.1 66 27.8 6.5 12.1 6.1 61
Xylene + ethylene benzene 1.8 1.9 0.4 0.05 0.05 0.4 1.8
Other aromatic hydrocarbons 0.2 1.1 1.2 0.3 0 0 0
Total aromatic hydrocarbons 98 86 30.6 8.05 14.35 84.5 89.8
Phenols Phenol 0.1 3.7 0.1 0.05 0.05 1.4 0.1
Cresole, o-, m-, p- 0.2 1.4 0 0.2 0.2 1.2 0.5
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010

Xylenol 0 0.9 0 0 0 0.3 0

Total phenols 0.3 6 0.1 0.25 0.25 2.9 0.6
Aldehydes Formaldehyde 0.4 0.4 0.2 0.6 0.6 0.2 0.4
Acetaldehyde 0.4 0.3 0.2 0.3 0.4 0.3 0.4
Acrolein 0 0 0 0 0 0 0
Acetone 2.3 2.2 1.7 0.6 0.7 2.6 2.9
Propanol 0 0.05 0 0 0 0 0
2-Butanone 0.2 0.2 0.1 0 0.05 0.2 0.2
Butanal 0 0 0 0 0 0 0
Benzaldehyde 0.1 0.2 0 0 0.05 0 0.1
Pentanal 0 0 0 0 0 0 0
Hexanal 0 0 0 0 0 0 0
2-Hydroxybenzaldehyde 3.3 0 0 1.5 1.8 3.2 3.8
Total aldehydes 6.7 3.35 2.2 3 3.6 6.5 7.8
CO (integrated) 487 65 24.2 188 368 258 578
Data is grouped in aromatic hydrocarbons, phenols, aldehydes and total CO.

of the experiments were finished, there is still considerable
emission of pyrolysis products. Experiment 12, which lasted
for 90 min, shows that for iron castings, emission levels drop
quickly after 30 min. But it appears that for the lower melt-
ing alloys, bronze and in particular the Al-alloy, emissions
have not peaked yet after 30 min. Regarding the effect of
the metal poured, the general trend is similar to that found
for the furan binders: that when the thermal load on the
mould is reduced the level of emissions per mould is also
reduced.
In general, the total emissions are approximately 20%
higher for PU binders than for FA binders (Table 4). This
is a result of a much higher emission of phenols when the
PU binders degrade. In the pyrolysis products from FA
binders phenols are hardly found, but in the PU binders
emission levels range from 11.2 to 52 mg/kg mould sand.
This is to some extent, but not completely, compensated
by lower emissions of CO. CO emissions vary between the
3 binders analysed, binders E and F emit less than half as
much CO than binder D.

Fig. 5. Emission of hydrocarbons as a function of time for PU

sand. Fig. 6. Emission of CO as a function of time for PU sand.
 1872 Tiedje et al.
Table 4. Detailed analysis of off-gasses from moulds made with PU binders.

Experiments

11-D-Fe
8-D-Fe 9-D-Cu 10-D-Al Shake out 12-D-Fe 13-E-Fe 14-F-Fe
Pyrolysis products ↓ [mg/kg] [mg/kg] [mg/kg] [mg/kg] [mg/kg] [mg/kg] [mg/kg]

Aromatic Benzene 70 48 0.6 4.8 11.6 63 83

hydrocarbons
Toluene 10.4 11.8 0.6 1.1 5.3 6.5 10.7
Xylene + ethylene benzene 5.7 8.1 0.5 1.4 5.6 2.2 9.1
Other aromatic hydrocarbons 294 334 241 193 351 214 429
Total aromatic hydrocarbons 380.1 401.9 242.7 200.3 373.5 285.7 531.8
Phenols Phenol 46 47 11.9 35.7 34.2 52 32
Cresole, o-, m-, p- 9.3 4.6 2 5.9 8.2 5 5.4
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010

Xylenol 1.1 0.7 0.05 0.5 1.1 0.7 1

Total phenols 56.4 52.3 13.95 42.1 43.5 57.7 38.4
Aldehydes Formaldehyde 0 0 0.3 0.5 0.2 0.2 0.3
Acetaldehyde 0.2 0.4 0.2 0.4 0.4 0.3 0.3
Acrolein 0 0 0 0 0 0 0
Acetone 0.2 0.4 0 0.3 0.3 0.1 0.05
Propanol 0.05 0.1 0.05 0.1 0.2 0.1 0.1
2-Butanone 0.1 0.1 0.05 0.1 0.2 0.2 0.3
Butanal 0.1 0.1 0.1 0.2 0.2 0.3 1.5
Benzaldehyde 0 0 0 0 0 0 0
Pentanal 0.1 0.1 0.05 0.1 0.2 0 0
Hexanal 0.2 0.3 0.2 0.3 0.5 0.1 0
2-Hydroxybenzaldehyde 2.5 3.4 1.7 6.2 9.3 4.2 4.9
Total aldehydes 3.45 4.9 2.65 8.2 11.5 5.5 7.45
CO (integrated) 501 55 1.7 60 9.4 230 181
Data is grouped in aromatic hydrocarbons, phenols, aldehydes and total CO.

As shown in Table 4, the variation in emission of other
hydrocarbons resulting from change of binder is not large,
though binder E emits only half the amount of aromatic
hydrocarbons than D and F. But by choosing a binder in
which part I contains significantly less hydrocarbons the
emission of CO is reduced considerably.
Emissions from resole-CO2 (RC) binders
Two experiments were done with RC binders. The differ-
ence in the binders is that “G” contains 5–10% 1-hydroxy-
3,6-dioxane and “H” contains 4–7% butyl glycol.

Fig. 8. Emission of hydrocarbons as a function of time for RC

Fig. 7. Pyrolysis products from PU sand as a function of time. sand.
 Organic compound emissions from sand mould metals 1873
Table 5. Detailed analysis of off-gasses from moulds made with
RC binders.

Experiments

15-G-Fe 16-H-Fe
Pyrolysis products [mg/kg] [mg/kg]

Aromatic Benzene 45 48
hydro-
carbons
Toluene 8.4 16.5
Xylene + ethylene benzene 4.5 0.05
Other carbohydrates 11.3 13.30
Total aromatic 69.2 77.85 Fig. 9. Emission of CO as a function of time RC sand.
hydrocarbons
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010

Phenols Phenol 9.8 0.05

Cresole, o-, m-, p- 8.6 41.40 The total amount of hydrocarbons measured was similar
Xylenol 7 13.10 to that of the PU binders, (Fig. 10). It should be noted
Total phenols 25.4 54.55
that when the experiments were stopped, there was still
Aldehydes Formaldehyde 1.7 0.1
Acetaldehyde 9 0.4 a relatively high rate of hydrocarbon emission. Thus, it is
Acrolein 0 0 possible that the total amount emitted will exceed that of
Acetone 0 0.5 the PU binders.
Propanol 1.9 0.1
2-Butanone 0.6 0.1
Butanal 3.5 0.1 Comparison between analysis methods
Benzaldehyde 0 0
Pentanal 0.3 0.2 The total emission of hydrocarbons and CO from all ex-
Hexanal 0.3 0.1 periments is shown in Figures 11 and 12. In the graphs,
2-Hydroxybenzaldehyde 0 1.5 the results acquired using filters are compared with the in-
Total aldehydes 17.3 3.1 tegrated data from the time-dependent analyses. There is
CO (integrated) 487 65 some variation between the exact results obtained by the
two methods. Qualitatively, the two methods give equal re-
Data is grouped in aromatic hydrocarbons, phenols, aldehydes and total
CO. sults. Both methods show high and low emissions for the
same experiments. In the hydrocarbon analysis it appears
that filter analysis gives lower values than integration of the
time dependent curves especially when the concentration of
Cast iron was poured in both moulds and emissions were CO is low. However with 5 minutes between each sample
measured for 30 minutes. taken, the integration over time becomes relatively crude,
There was very little difference between the emission and will to some extend also explain the difference in the
of total hydrocarbons and phenols from the two types of results.
binders (Figs. 8, 9 and 10). Emission of hydrocarbons (Fig.
8) was generally lower than from FA and PU binders. Ini-
tial hydrocarbon concentration was approximately 50 ppm,
which after 5 min was reduced to 35 ppm. This level of con-
centration was constant for the remaining 25 min, and there
was no indication that emission levels were falling at that
time.
All data from the filter analysis are shown in Table 5,
which shows that less aldehydes and CO is emitted from
binder H than binder G. This is to some extend compen-
sated by a higher emission of phenols and cresol, so that
the total emission of hydrocarbons is the same for the two
binders.
Emission of CO (Fig. 9) was similar to that found when
iron was cast in FA and PU moulds. The highest level (275
– 300 ppm) was found immediately after pouring. After
30 min the level vas reduced to about 20 ppm. Fig. 10. Pyrolysis products from RC sand as a function of time.
 1874
Fig. 11. Total carbohydrate emission per kg. mould sand for each
experiment. Black bars are results from analysis of filters and grey
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010
bars are integrated values.
Discussion
Quality of the analyses
A part of the present work was aimed at developing a
method to analyse emission from degradation of core and
moulding sand binders so that they can be compared with
respect to their impact on the working environment.
Figures 11 and 12 shows the correlation between integra-
tion of the time-dependent measurements and the accumu-
lated values from the filters, this proves a good agreement
between the two types of measurements. Both methods
show that the same binders have high, medium or low lev-
els of emissions. The relative difference in the results is
largest for compounds with very low emission levels. This
is because low concentrations are more difficult to measure
correctly since fluctuations in air flow around the tube,
through which samples are extracted, has the highest influ-
ence on the results.
The chemical analysis compare well with laboratory ex-
periments found in the literature. In a laboratory study of
Fig. 12. Total CO emission per mould for each experiment. Black
bars are results from analysis of filters and grey bars are integrated
values.
Tiedje et al.
degradation products from FA and PU binders at 750◦ C
Dungan and Reeves[16] found that the most important
pyrolysis products from both binders were benzene and
phenol. In addition to that the FA binder created some
cresol and the PU binder created toluene. Their analysis is
not quantitative, but only points to predominant pyrolysis
products at the particular temperature. The results are in
good agreement with the analysis done under similar con-
ditions by Wang et al.[11] who finds that the main pyrolysis
products from degradation of a PU binder is phenol, cresol,
benzene and toluene. Though these experiments were con-
ducted at constant, high temperature in the laboratory, the
results also compare well with the findings in this work.
The experiments show that of the many compounds anal-
ysed, it is few that make up the bulk of the total fumes
emitted. Therefore, it is possible to reduce the test proce-
dure considerably by only measuring these key components
of new binders to create a simpler bench marking test of
binders. The key components are: benzene and CO for fu-
ran binders, benzene, phenol and CO for cold box binders
and benzene, total phenols and CO for Resol-CO2 binders.
Since the phenol binders were cured with PTA or BSA it is
very likely that emissions from those will also include SO2 .
It was not possible to test that in the present analysis.
Emissions while cooling in the mould
For FA binders emission of CO and hydrocarbons follow
the same mode in all cases when iron is cast in the moulds.
Almost immediately, within 1 to 2 minutes after pouring,
the maximum amount of fumes is emitted. The level of
emissions falls exponentially, and in most cases emissions
are very low after 30 minutes. The picture is the same for
emissions of CO from the other types of binders.
When alloys with lower melting point is poured in fu-
ran moulds, it takes 5 to 8 minutes before the emission of
fumes reach its maximum, after that emission levels also
fall exponentially.
Since pyrolysis of the binders is determined by the local
temperature in the mould,[8,9] it seems logical that emis-
sions are high when the pouring temperature is high and
just after the casting has been poured. As the heat from the
casting flows into the mould the casting is cooled. The max-
imum temperature falls as the thermal field spreads into the
sand. As a result the binder in the sand close to the casting
degenerates quickly. Further from the casting temperatures
are lower, and degeneration of binders is slow. Fe and Cu
based castings are poured at temperatures above 1000◦ C
and will as such have a high impact on the sand binders.
Al alloys are poured at much lower temperature, typically
650 to 750◦ C. In the literature it has been found that the
organic components in binders degrade in the temperature
range from 200 to 600◦ C, with the largest amount of fumes
emitted from 350 to 480◦ C.[9] The sand around aluminium
castings is quickly cooled to temperatures within that range.
As a result binder degradation is reduced too.
 Organic compound emissions from sand mould metals
Emission of hydrocarbons from PU binders is postponed
almost 20 min. As a result, emission levels are still high after
30 min. It takes approximately 70 minutes before the levels
are reduced to below 10% of the maximum. This means that
in foundries where cold box moulds are opened manually,
it will be advisable to wait more than an hour after pouring,
before moulds are opened.
The reason why emissions from PU binders take rel-
atively long to reach the peak is not immediately clear.
But a study by Dungan and Reeves[16] showed that PU
binders emit a large amount of volatile compounds (ben-
zenes) which are produced early in the process. It is possible
that light components evaporate from the hot parts of the
mould and then condense again in the outer, cold layers. As
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010
heat travels into the mould the condensed material evapo-
rate again and is emitted together with the heavier pyrolysis
products.
Emission of hydrocarbons from RC moulds is different
from the other binders. Initially emissions are high, though
considerably lower than for the other binders. After 7 to
10 minutes, the level has fallen to approximately 60% of
the maximum. This level of emission remains constant for
at least 30 minutes. There is nothing that indicates that
emission levels will fall shortly after 30 minutes. This means
that also in this case it will be advantageous to postpone
shake-out for some time.
Effect of binder composition on emissions
Generally, there is very little difference between the com-
pounds emitted from binders within the same binder sys-
tem. It is only the FA binders where curing with BSA in-
stead of PTA reduce the total amount of toluene signifi-
cantly. This is at the expense of an increase in the level of
phenol and cresol. Though the latter compounds are still
only emitted at low levels. In the PU binders it seems that
changing the binder composition and the curing agent can
lead to a reduction in CO emission whereas binder compo-
sition has little effect on the hydrocarbons.
In the literature ithas in one case been shown that the
composition of PU binders affect emissions. Wang et al.[11]
studied degradation of three PU binders in the labora-
tory: Conventional, biodiesel and collagen based binders.
They found that the collagen based binder emitted lower
amounts of CO and hydrocarbons, particularly benzene,
phenol and cresol over a wide range of temperatures.
Toluene, however, was emitted in large amounts over a very
small temperature range compared to the other binders.
This indicates that it, to some degree, is possible to op-
timise emissions from organic binders through a correct
choice of chemical components. Between the three differ-
ent types of binders studied here, there are clear differences
in the organic compounds emitted. Where emission of hy-
drocarbons in FU binders is immediate and fades relatively
quick the other two binders are more sluggish in their degra-
dation and tend not to fade in the same way.
1875
As it is shown in Figure 11, PU binders emit the highest
total amount of hydrocarbons. This trend is emphasised
by the fact that the present investigation is most likely to
underestimate the total level of hydrocarbons emitted from
PU and RC binders. In the working environment it is in-
teresting to note that the amount of off-gasses from FU
binders is reduced to a very low level after approximately
30 minutes. Emissions from the two other binders continue
for a much longer time. Thus the need for ventilation in
the work place is very different depending on the type of
binder used.
With regard to CO emission the three types of binders
seem to behave in the same way: immediate high emission
that gradually fades within 30 min. It is interesting to note
that some of the RC type binders have been marketed as
“greener” than other binders, primarily because they are
better for the environment in the core shop than PU type
binders. But the present analysis indicates that the improve-
ment in working environment does not necessarily extend
to the casthouse. Emissions immediately after pouring are
low, but the total emitted compounds are the same.
Emissions during shake-out
When moulds are opened, trapped gasses are released to the
environment. The concentration of hydrocarbons increases
immediately to a level equal to the maximum level emitted
after pouring. After 7 to 8 min it is reduced to half the max-
imum value but emissions continue for at least 30 min. Cold
box binders emit almost 3 times as many hydrocarbons as
furan-bonded moulds.
CO emissions also increase rapidly at shake out but max-
imum concentration only reaches half the maximum level
after pouring. Emissions decrease very rapidly and after
20 min almost no CO is emitted.
The result is that in the shake out, the level of emissions
is very high. In foundries where there is continuous opening
of moulds, the level of emissions will permanently be very
high, especially when cold box binders are used.
Effect of casting wall thickness and sand-metal ratio
The gasses emitted are formed inside the mould, very close
to the surface of the casting. Then they travel through the
mould to the surface. Mass transport in the mould is a
complex process driven by thermal conditions, concentra-
tion gradients, casting geometry, etc. As the fumes flow
through the mould, they are cooled. Some components of
the gas will condense on the sand grains depending on lo-
cal temperature and dew point of the gas. Looking at the
results of the time-dependent analyses, we see that the emis-
sion is high immediately after the mould is filled. Emissions
fall approximately exponentially with time, which is to be
expected.
In some cases, especially where the thermal load on
the mould is small, the maximum emission of pyrolysis
 1876
products is found after some min. This is also understand-
able since it will take some time for the fumes to travel
through the mould and because some of the products emit-
ted will condense in the sand as long as it is cold. When the
mould heats up, the gases evaporate again and are emitted
to the atmosphere.
If the sand-metal ratio is increased total emissions will
be reduced, but the time for when the maximum level of
emissions are reached will be delayed. With increasing sand
to metal ratio it is to be expected that the amount of organic
compounds that condense and is absorbed in the sand will
increase. Some of this will be emitted to the air in the
shake-out while other compounds will remain dissolved
in the sand. Compounds that remain in the sand are later
Downloaded By: [Technical Knowledge Center of Denmark (DTIC)] At: 09:52 4 November 2010
removed as the sand is regenerated.
If the wall thickness of the casting is decreased, total
emissions will be reduced, but if the sand-metal ratio is
kept constant, the time to reach maximum emission should
not be delayed because the temperature field in the sand
is unchanged. Many cold box cores using PU binders are
relatively small because they are used for intricate internal
geometries in the castings. They will experience a high ther-
mal load, and generally higher temperatures than the other
binders.
Conclusion
I. A method has been proposed to analyse and bench-
mark mould and core binders with regard to emissions
during production of castings.
II. When high melting point alloys (iron and bronze) are
cast, emission levels are initially very high and fall
gradually to less than 10% of the maximum emission
within 1 h. Two exemptions are: emission of hydrocar-
bons from PU binders and emission of hydrocarbons
from RC binders. In the first case, the emission starts
at a medium level. It reaches a maximum level after 20
min, after which it falls to below 10% of the maximum
within 70 min after pouring. In the second case emis-
sion is generally low compared to the other binders,
but the emission continues for longer time.
III. Total emission and melt pouring temperature is re-
lated so that higher melting point results in higher
total emission levels.
IV. PU binders emit far more hydrocarbons than the
other binders and emissions take place over longer
time.
V. The initial binder composition of FA binders does
to some degree influence the composition of the off-
gasses. For other binders the picture is unclear.
VI. At shake-out there is a drastic increase of compounds
emitted from the moulds, even in the cases where the
emission from unopened moulds have fallen to a very
low level prior to shake-out.
View publication stats
Tiedje et al.
VII. The most important compounds emitted from all the
binders are benzene and CO. In addition total phenols
play an important role for furan binders.
Acknowledgments
The project was initiated and funded by the Danish Min-
istry for the Environment, and run by the Danish Foundry
Association. Experiments have been carried out at the Tech-
nical University of Denmark, Department of Mechanical
Engineering in collaboration with the Danish Technologi-
cal Institute. Chemical analysis was carried out at Danish
Technological Institute.
References
[1] CAEF - European Foundry Association, Casting Production Statis-
tics, http://www.caef.org, 2010.
[2] Campbell, J. Castings, Edition 2; Butterworth Heinemann: Oxford,
2003.
[3] Ladegourdie, G.; Loechte, K.; Schuh, W. Cold Box 96 — the new
environmentally friendly cold box binder system. Foundry Trade J.
1997, October, 434–436.
[4] Löchte, K.; Clarke, C. Don’t Cold Shoulder Cold Box Process.
Foundry Trade J. 1999, January, 46–49.
[5] Berthelet, M; Vargas, M. Nouvelles possibilités pour les fonderies,
grâce l’emploi du système liant polysialate pour l’agglormération
de matériaux de moulage. Hommes et fonderie 1999, 297, 22–25.
[6] Siak, J.-S.; Whited, W.T.; Schreck, R.M.; Ferrero, A.; Badiali, M.
A green core sand binder facilitating easy core removal from com-
plex aluminium castings. In CIATF Technical Forum 1999; Verein
Deutsche Geissereifachleute, Düsseldorf, June 10–11, 1999.
[7] Hendershot, G.; Werner, A. Eine neue Cold-Box-Technologie. In
CIATF Technical Forum 1999; Verein Deutsche Geissereifachleute,
Düsseldorf, June 10–11, 1999.
[8] Lemon, P. H. R. B.; Morgan, A. D.; Terron, C.; Youren, J. High-
Temperature Properties of Furane Cold-set Resins. Foundry Trade
J. 1973, 134, 423–437.
[9] Serebro, V. S.; Rits, B. A.; Sergeeva, T. E. Curing and Thermal
Degradation of Phenol-Formaldehyde Binders. Russ. Cast. Prod.
1974, 2, 55–56.
[10] Southern Research Institute, Binder Decomposition During Pour-
ing and Solidification of Foundry Castings - Particulate Emissions
From Foundry Molds. Cast Metals J. 1979, 4, 14–15.
[11] Wang, Y.; Cannon, F. S.; Salama, M.; Goudzwaard, J.; Furness,
A. J. Characterization of hydrocarbon emissions from green sand
foundry core binders by analytical pyrolysis. Environ. Sci. Technol.
2007, 41, 7922–7927.
[12] Kauffmann, P.; Voigt, R. Modeling foundry air emissions. Mod.
Cast. 1997, 87, 53–55.
[13] Rarick, T.R.; Crandell, G.R. Grasping the impact of emission fac-
tors. Mod. Cast. 2003, 93, 35–37.
[14] Westberg, H.; Lofstedt, H.; Selden, A.; Lilja, B.G.; Naystrom, P.
Exposure to low molecular weight isocyanates and formaldehyde in
foundries using hot box core binders. Ann. Occup. Hyg. 2005, 49,
719–725.
[15] Scarber, J. Gas defects in aluminum castings from cores and mold
washes. Foundry Manag. and Technol. 2004,132, 11–12.
[16] Dungan, R. S.; Reeves III, J. B. Pyrolysis of foundry sand resins: A
determination of organic products by mass spectrometry. J. Environ.
Sci. Health to Part A 2005, 40, 1557–1567.