Recent developments in the application of the

Polyurethane Cold Box process

The reason for its popularity is

based upon several factors such
as high out of box strength, high
productivity levels, excellent
breakdown properties and cost.

Polyurethane Cold Box process

Developed over 30 years ago, the

Polyurethane Cold Box Process
(PUCB) is still the most popular
organic cold box technique for
the automated production of
high volume cores in both
ferrous and non-ferrous
foundries.

PSA Peugeot-Citron Sept-Fons site is one such foundry that is driving

binder suppliers to take up the challenge.

Type of Casting

Tonnes Produced

Cylinder Blocks

61,163

Brake Discs

44,192

Brake Drums

11,721

Total

117,076
Table 1 Foundry output 2003

To produce all these castings, the foundry uses a tremendous amount of

sand. Due to the stringent legislation appertaining to the dumping of
waste sand and the ever increasing costs of new sand, the Sept-Fons
site has introduced a dry attrition sand reclamation process.
Mechanical and magnetic treatment of the sand:
The sand coming from the shake out (mix of moulding sand and core
sand, burnt or unburnt) is treated:

Introduction
FOSECO has recently introduced new
technologies with the emphasis on
adding value for the customer by
improving key areas such as productivity,
lower VOC (volatile organic compounds)
emissions and by reducing waste.
The following case studies are good
examples of these new technologies in
action.

Case study 1
PSA Peugeot-Citron Sept-Fons Site. The
development of a new PUCB binder for
use on mechanically reclaimed sand
based on low VOC solvent technology.
Environmental regulations and laws for
the protection of the external environment
and the internal work place are having a
major impact on foundries using chemical
binders . Nowadays the foundry industry
demands new developments in chemical
binders which will offer excellent
performance but have little or no
detrimental effect on the environment.
The present challenge for all major binder
suppliers is to develop and supply binder
systems that offer lower monomer
contents and improved environmental
properties such as the use of solvents and
other additives that conform to ever more
stringent health and safety requirements.

Magnetically separate green sand

Mechanically to eliminate the resin residues
The reclaimed sand is used at 100% in the coreshop.
The challenge for the binder supplier was to develop a system that
would give the same strength characteristics as for new sand as well as
lowering the emissions of aromatic hydrocarbon solvents, without
compromising casting quality.
Using a conventional PUCB resin system based on aromatic solvents,
revealed that on reclaimed sand the core strengths were 20% lower than
with new sand and in addition the bench life of the mixed sand was
much shorter. Obviously a novel approach to the chemistry of the binder
system was needed.

New Sand

Reclaimed Sand

Loss on Ignition

0.15

2.29

Average Grain Size

280

291

Fines Content

0.150

0.25

Dust Content

0.00

0.00

Table 2 Sand properties (base sand Silfraco LA32)

New Sand

Reclaimed Sand

As Gassed

350

310

1 Hour

495

410

24 Hour

550

515

Humidity Resistance

320

260

Table 3 Strength development comparison

Values are transverse strength N/cm2 based on a total binder content

o
of 1.2%. Humidity Resistance = 24hours at 20 C / 95% Humidity.

01

New Sand

Reclaimed Sand

As Gassed

365

325

1 Hour

310

275

2 Hours

280

200

3 Hours

245

150

Table 4 Bench life properties

Values are transverse strength N/cm2 based on a total binder content

of 1.2%.
The simple solution to increase strength on reclaimed sand would be
to increase the resin addition, however, there are clear disadvantages
associated with this:

This new development resin system,

Politec E2000 / E9000 has now been
established at PSA Peugeot-Citron SeptFons Site for use throughout the core shop
using reclaimed sand (figure 1). This
system has met the technical requirements
in terms of strength performance and
environmental improvements with
additional process benefits such as good
bench life, humidity resistance and casting
quality being achieved. A further benefit
has been a reduction in resin wipe off
which in turn reduces box cleaning thus
improving productivity.

Reduced sand flowability requiring higher blowing pressures

Higher blow pressures would cause resin wipe off and sticking of
the cores
Higher binder levels would result in poor core breakdown and core
shakeout
Higher binder levels would also increase the risk of casting defects
associated with lustrous carbon and gas evolution
Higher process costs

Clearly an increase in resin content was not a satisfactory solution and

as such, alternative solutions would need to be identified.
A new binder was selected based on two novel manufacturing
processes. Firstly, a route was identified whereby water could be
removed in process more efficiently. This resulted in a significant
change to the molecular structure of the resin resulting in a much more
reactive resin that enabled higher out of box strengths to be achieved.
The second development route was based on the introduction of new
low aromatic solvents. These had a significant impact on the levels of
VOC emissions particularly during the casting stage.

Aromatic Solvent System

Politec* E2000 / E9000

Benzene

13 ppm

3 ppm

Toluene

10 ppm

2 ppm

Xylene

15 ppm

1 ppm

Table 5 Comparison of VOC emissions during the casting process

Aromatic Solvent Politec E2000 / E9000

System
As Gassed

310

350

1 Hour

420

480

24 Hour

550

600

Humidity Resistance

250

350

Table 6 Strength development comparison: Politec E2000 / E9000 on reclaimed sand

Values are transverse strength N/cm2 based on a total binder content of

1.2 %.

02

Figure 1 Core assembly at PSA Peugeot-Citron

Sept-Fons Site

TMA: The application of (TMA)

Tri-methylamine using new amine
generator technology
The choice of amine for use as a trigger in
the PUCB process is based upon the
following factors; reactivity, odour, cost
and the ease at which the amine liquid
can be vapourised and injected into a
suitable inert carrier gas (usually hot air).
This carrier gas penetrates the mould or
core mass promoting a reaction between
the resin and MDI (Methly Di-Isocyanate)
components.
Several amines have been used
successfully, however, they all have
process limitations, for example TEA
(Tri-Ethyl Amine) is not very reactive
resulting in higher amine demand, it also
has the highest boiling point making it
difficult to keep in a vapour form, leading
to condensate forming in the gas supply
system. DMEA (Di-Methyl Ethyl Amine),
is more reactive and has a lower boiling
point but has a greater odour and is
traditionally more expensive.
FOSECO recently introduced a new amine
for use with the PUCB process based on
TMA. This amine is significantly more
reactive than DMEA resulting in extremely
low application rates, which if applied
using specifically designed amine
generators gives rise to low levels of
odour in use.
Furthermore the increase in reactivity and
low application rates means that higher
productivity gains are achieved by using
shorter gassing and purge cycle times,
making the overall process costs lower.
TMA exists as a gas at room temperature,
compared to other traditional amines,
which are liquid. Based on this difference
a new approach had to be found in terms
of injecting TMA into the carrier gas

Polyurethane Cold Box process

Case study 2

stream to ensure that process reproducibility could be achieved. Initially

a TMA generator was developed which involved heating the TMA gas
cylinder above 65oC to achieve a high pressure enabling TMA gas to be
drawn off and injected into the carrier gas. Whilst several generators of
this type have been built and commissioned successfully in several
foundries a number of technical concerns needed to be resolved. The
major concern was that as TMA was drawn out of the gas cylinder the
pressure dropped. To maintain pressure, and thus ensuring consistent
flow application rates, more heat had to be put into the cylinder, this
proved difficult especially at low volumes of TMA within the cylinder.
In addition, when the TMA was exhausted, a new cylinder had to be
heated to the required temperature to achieve working pressure
resulting in production down time. A final obvious concern was the
health and safety issue concerning the basic principle of heating gas
cylinders containing TMA having a low flash point.
Clearly a better design based on the use of cold TMA had to be found
if the benefits of this new amine technology were to be exploited.
Reviewing a combination of existing gas application techniques and
looking at new developments in low level metering devices a prototype
TMA generator was built with the co-operation of Omega Foundry
Machinery Limited.
TMA Generator
TMA is now supplied in a dual port syphon cylinder which when
pressurised with a constant nitrogen pressure at ambient temperature
enables liquid TMA to be drawn from the cylinder. As the TMA is at a
fixed pressure the flow rate from the cylinder remains constant.
The liquid TMA is then atomised at the point of entry into the carrier gas
stream using a low level atomiser capable of atomising 1cc of amine per
cycle. To achieve this level of accuracy a digital pressure control system
prior to the atomiser removes any pressure spikes.
The above process is built into a generator package which supplies the
carrier gas at the required temperature, controls gas and purge cycle
times through an on board PLC system and allows variations to be made
in terms of carrier gas application pressure. As this system allows the
use of TMA as received, no down time is necessary when the cylinder is
exhausted, bottle change can be made in less than 1 minute. A further
benefit can be achieved by using a multiple gas cylinder manifold or
bulk TMA supply, in that the operator does not have to handle the amine
component resulting in the elimination of a hazard risk.
This new concept generator (figure 2), has undergone field application
studies and the results observed show that this process has a lot to offer
particularly in terms of improved productivity and the safe handling and
application of a very reactive amine.

03

TEA
DMIA
DMEA
TMA
Chemical
Formula
(C2H5)3N (CH3)CHN(CH3)2 C2H5N(CH3)2 (CH3)3N
Molecular Wt
g/mol
101.2
87.1
73.1
59.1
0
Boiling Point C
89
65
35
3
Vapour Pressure
mbar (200C)
69
170
580
2200
DMIA: Di Methyl Iso Propyl Amine
Table 7 Comparison of Physical Properties of Commercial Amines

Figure 2 TMA Amine generator

Cope/Drag Assembly

Water Jacket

Inlet Port Core

Machine Cycle Time : DMEA

56 seconds

43 seconds

42 seconds

Machine Cycle Time : TMA

43 seconds

37 seconds

35 seconds

Cores per Hour : DMEA

64

84

86

Cores per Hour : TMA

84

97

103

Water Jacket

Inlet Port Core

Table 8 Comparative productivity rates

Cope/Drag Assembly
Total Binder Content (BOS)

1.3 %

1.4 %

1.1 %

DMEA Addition Rate (BOR)

16.7 %

32.2 %

31.46 %

TMA Addition Rate (BOR)

3.85 %

3.98 %

5.45 %

Cost/Tonne of Sand : DMEA

21.95

26.68

20.85

Cost/Tonne of Sand : TMA

20.42

22.04

17.80

6.9%

17.4%

16.6%

Amine cost saving

Table 9 Foundry case study: Cost in use comparison, TMA vs. DMEA

04

Reclaimed Shell Sand for the production of PUCB cores,

the reduction of a waste sand stream.
RSM Castings is a repetition iron foundry producing
automotive, hydraulic and general engineering
components in grey and SG Iron using the shell mould
and core process.
Four years ago, RSM Castings invested in 2 corebelters
(figure 3), to enable them to produce PUCB cores so
that they could eliminate the sub contract purchase of
cores and convert some shell production to PUCB to
reduce costs.

Figure 3 Corebelter

It soon became apparent, however, that production costs

were increasing significantly due to rising costs in
respect to new sand purchase and also the everincreasing cost of sand disposal

Polyurethane Cold Box process

Case study 3

Initial chemical examination of this waste stream showed

that the level of binder residue was very low, typically
0.4% due to the low sand to metal ratio associated with
the shell process resulting in excellent thermal
combustion of the organic component. The main
concern, however, was the increase in alkalinity of the
sand, caused by oxide residues melting and forming an
alkali coating on the sand grain. It is known in the PUCB
process that high alkali sands can give technical
problems such as low out of box strengths and an
increase in amine demand which in turn results in
reduced productivity.
Trials were conducted using a pilot dry attrition plant to
see if this increase in alkalinity would be detrimental to
the process prior to embarking on any capital outlay on
new equipment.
Several tonnes of shell sand were processed ready for
evaluation on one of the production corebelters. A sample
of this sand was tested and the results indicated that the
alkalinity was even higher than the initial waste stream
sample, this high alkalinity being due to the very high
dust content of the sand, the pilot plant had no extraction
system at this time. It was decided, however, that this
sand would be evaluated on a trial production basis with
binder levels being increased to compensate for the
increase in AFS and surface area of the sand caused by
the high dust content. The resultant cores produced
(figure 4), were surprisingly of excellent quality in terms
of edge retention and strength, it was also noted that the
amine dosing and purge times did not need adjustment.

With assistance from FOSECO, RSM Castings began a

study to investigate the feasibility of using the waste
sand stream from its shell moulding process as a
replacement for new silica sand in its PUCB process. This
PUCB process being based upon a 50 AFS grade new
silica sand, bonded with FOSECO Politec XP1000 /
XP1080 PUCB binder system in conjunction with TEA
amine. RSM Castings produce approximately 250 tonnes
of waste stream shell sand per month.

Figure 4 Cores (reclaim)

Castings from these cores were produced with

comparable results in surface finish to castings made in
new sand and with the added benefit of a reduction in
the amount of veining defects .

05

From these positive results, further quantities of waste

stream sand were treated on the pilot plant, with a dust
extraction system fitted, and the production exercise
repeated with standard binder addition levels. Once
again good quality cores and castings were produced
(figure 5).

Figure 5 Castings

RSM Castings were pleased with the observed results

and a commercial study confirmed that the purchase of a
3 tonnes per hour reclamation plant would give a return
on investment within 12 months.
Based on this a reclamation plant was purchased and
commissioned in September 2003.
RSM Castings are now able to satisfy totally their
requirement for PUCB sand from their waste shell
stream. Furthermore, RSM Castings supplier of shell
sand showed an interest in the waste shell stream and
are now successfully recoating 100 tonnes per month
with a moulding grade shell resin.

Sample

AFS

Loss On Ignition

pH

Fines%

Dust %

New Silica Sand

50

0.27%

7.40

0.45%

0.01%

Waste Stream Pilot Plant

No Extraction

70

0.68%

8.10

10.96%

1.41%

Waste Stream
Production Plant

64

0.32%

7.38

0.88%

0.32%

Table 10 Sand properties

Sand

Total Binder

1 Hour

2 Hour

4 Hour

24 Hour

New Silica Sand

1.2 %

290

310

360

375

Waste Stream

1.2 %

220

265

280

295

Waste Stream

1.4 %

285

325

355

39

Strength values: Transverse N/cm2

Table 11 Strength development comparison

06

PUCB Waste Stream Sand

New Silica Sand

24.00

Sand Disposal

10.00

Waste Stream Sand *

Total Binder 1.2 %

0.20

17.59

20.31

51 .59

20.51

Total Binder1.4 %
Total Cost

Polyurethane Cold Box process

PUCB New Sand

* Energy cost required to process 1 tonne of waste stream sand .

Table 12 Commercial comparison

Conclusion
The above case studies clearly demonstrate that the introduction of new technology to
an existing core process can help foundries to improve their process costs, whether it is
by improved productivity, the reclamation of sand or indeed by looking at alternative
methods for eliminating waste streams. Above all, it is important to continue this
process of continuous improvement to ensure that foundries remain competitive and
maintain a working environment that meets current and future Health and Safety
requirements.

References/acknowledgments
Peugeot-Citron Sept-Fons Foundry
Omega Foundry Machinery Limited
RSM Castings Limited
Dynea UK Limited
Moulay Bouhanani : Dynea UK Limited
Christoph Genzler : FOSECO
Barrie Jones : RSM Castings Limited

07