International Journal of Hygiene and Environmental Health 213 (2010) 475488

Contents lists available at ScienceDirect

International Journal of Hygiene and

Environmental Health
journal homepage: www.elsevier.de/ijheh

Isocyanate and total inhalable particulate air measurements in the European

wood panel industry
E. Vangronsveld 1,2, S. Berckmans, K. Verbinnen, C. Van Leeuw, C. Bormans ,1
Industrial Hygiene Centre of Excellence, Huntsman, Everslaan 45, B-3078 Everberg, Belgium

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

Article history: Introduction: It is well known that the use of MDI (methylene diphenyldiisocyanate) as an alternative for
Received 20 April 2010 formaldehyde-based resins is seen as a responsible option to reduce formaldehyde emissions for CWP
Received in revised form 13 August 2010 (Composite Wood Products) in buildings. However, there are concerns raised regarding the exposure risk
Accepted 19 August 2010
of workers. The purpose of this article is to provide the reader with factual information to demonstrate
that the use of MDI compared to other agents used in CWP production processes does not pose increased
Keywords:
inhalation exposure risks for workers. Personal and area air measurements were carried out at nine
Isocyanate
Composite Wood Panel plants throughout Europe to assess potential inhalation exposures to MDI and
Total inhalable particulates
Composite wood products
wood dust as Total Inhalable Particulates (TIP). In total, 446 pairs of samples were collected for MDI and
Industrial hygiene TIP of which 283 pairs were personal samples and the remaining 163 pairs were area samples collected at
Air monitoring key locations along the production line. This data together with published formaldehyde exposure data
European wood panel industry has been used to evaluate the exposure safety margin opposite what are considered relevant occupational
exposure limits.
Material and methods: The methods used for sampling and analysing MDI and TIP are based on interna-
tionally accepted methods, i.e. MDHS 25/3 (or ISO 16702) for MDI, and MDHS 14/3 for TIP.
Results: The job functions with an increased exposure prole for TIP were the cleaners, drying operators
and quality control staff, and for MDI, the cleaners and quality control staff. The areas with an inc reased
exposure prole for TIP are the conveyor area from OSB blender to former area and the OSB press infeed,
and for MDI the OSB weigh belt and OSB former bin area.
Conclusions: The exposure safety margin opposite the selected exposure limits can be ranked as
MDI > TIP > formaldehyde (high margin of safety to low margin of safety), indicating that the use of MDI
also reduces the exposure risks to workers during production of CWP compared to formaldehyde.
By reducing the airborne TIP concentrations, a respiratory sensitiser, MDI workplace concentrations
in general can be reduced further. This can be achieved by improving design and/or maintenance and
testing programmes of existing control measures, which should be in place already to effectively control
exposure to TIP and formaldehyde. The airborne concentration of MDI at workstations situated after
pressing (curing) is regarded as extremely low and likely mainly constituted by workplace emissions
from elsewhere in the plant.
2010 Elsevier GmbH. All rights reserved.

Introduction
MDI (methylene diphenyldiisocyanate) and its derivatised
products, such as polymeric methylene diphenyldiisocyanate
(pMDI), are used by the industry for the production of Polyurethane
(PUR) products, e.g., rigid foams, exible foams, elastomers, adhe-
sives, binders, coatings, etc. (Randall and Lee, 2002).
Corresponding author. Tel.: +32 0 2 758 93 36; fax: +32 0 2 758 73 36.
E-mail address: Christof bormans@huntsman.com (C. Bormans).
1
These authors contributed equally to this work.
2
Industrial Hygienist.
Currently, there are four main different Composite Wood Prod-
ucts (CWP) manufactured where pMDI is used as a wood binder, i.e.
Oriented Strand Boards (OSB) (Fig. 1, United Kingdom Environment
Agency, 2006), Medium Density Fibre Boards (MDF) (Fig. 2, United
Kingdom Environment Agency, 2006), Particle Board (PB) and
Wood Fibre Insulation Boards (WFI). The percentage of pMDI used
for these applications can vary from 0.25% (where it is used as
hybrid system in combination with formaldehyde-based resins) to
7% for high quality CWP. The pMDI for OSB can be used in the
core layer alone or in all layers depending on the quality and/or
formaldehyde emissions are required.
Polymeric MDI (pMDI) (the primary form of pMDI is typically a
mixture that contains 50% monomeric MDI as well oligomers con-
taining 36 rings and other isomers) is introduced as a spray into an

1438-4639/$ see front matter 2010 Elsevier GmbH. All rights reserved.
doi:10.1016/j.ijheh.2010.08.011
476 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488

Fig. 1. OSB line.

enclosed rotating blender system (OSB and PB) or into a so called
dry blending system (MDF and WFI) or blown into an enclosed
wood transfer piping system (MDF and WFI). Wood strands or par-
ticles (OSB and PB) or bres (MDF and WFI) are coated with pMDI
in these systems, then transported by the conveyors (OSB and PB)
or piping (MDF and WFI) to the former bins and former laydown
area. At this laydown area, the wood strands, particles or bres
are dropped onto a conveyor, which transports the formed mat
to the press. It is no surprise that the dropping operation can
result in signicant generation of airborne TIP (wood dust) and
pMDI coated wood particles, and if engineering controls like con-
tainment and/or ventilation are not in place or are not operating
efciently, this leads to increased inhalation and dermal exposure
risks in this area. Normally there are 14 former bins and former
areas on a production line, each creating a layer to build up the nal
thickness of the mat. After the former area, the mat is conveyed
to the pre-press area, if present, to slightly press (cold press) the
mat to the desired thickness before being pressed in a discontinu-
ous or continuous hot press to produce the nal composite wood
product.

Fig. 2. MDF line.

 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488
After the press, the wood panels are transported to the
initial saw station where the boards are cut to the required
lengths. The boards are then conveyed to the boardcooler before
being transported to the saw nishing area where they are
cut to the nal product sizes (with or without tongue and
groove).
It is well known that the use of pMDI as an alternative binder
for formaldehyde-based resins is seen as a responsible option to
reduce formaldehyde emissions in buildings (Allport et al., 2003a;
US LEED). However, concerns are raised regarding the exposure
risk of workers. The purpose of this article is to provide the reader
with factual information to demonstrate that the use of pMDI com-
pared to other agents used in these production processes does not
pose increased inhalation exposure risks for workers. Personal and
area air measurements have been carried out at nine CWP plants
throughout Europe to assess potential inhalation exposures to MDI
and TIP, the latter assumed in general to be constituted of 100%
wood dust. Of the 446 pairs of samples collected for MDI and TIP,
283 pairs were personal samples measured over a ve-year period
(20042009) during 37 Industrial Hygiene (IH) surveys and the
remaining 163 pairs were background area samples collected at key
locations along the production line, measured in 2008/2009 during
15 IH surveys. Data has been collected for OSB, MDF, WFI and for PB.
However, no results for the PB application are presented since cur-
rently only hybrid systems are used whereby only 0.252.5% MDI
is added to other resin systems and exposure risks are not expected
to be substantially different compared to the results presented for
the other applications.
Table 1
Chemical structure and CAS registry numbers MDI isomers, pMDI and formaldehyde.
477
In addition, some public (Canada Monitoring, 2006; Lavou et
al., 2006; Discussion Paper, 2007; Lavou et al., 2005, IRSST and
Acton, 2009) available formaldehyde exposure data has been used
to evaluate the exposure safety margin opposite what is considered
relevant occupational exposure limits (2 mg/m3 for TIP, 0.05 mg/m3
for MDI, and 0.37 mg/m3 for formaldehyde).
Physical properties
Table 1: (Allport et al., 2003c).
Table 2: (Allport et al., 2003d; Occupational Safety & Health
Administration).
Potential health effects
MDI
Respiratory health effects (from mild irritation of the airways to
more severe effects), dermal effects (irritation, sensitisation) and
eye effects (irritation) (Allport et al., 2003e) due to exposure to
vapour (heating) or MDI (Figs. 3 and 4) coated on wood dust, could
occur. People with a history of other allergies may be especially vul-
nerable and a medical evaluation prior to the use of MDI, 6 weeks
after use and at regular intervals thereafter, is useful to identify
these people, since health effects usually become noticeable during
rst two years after using MDI. Affected people require a job where
they do not come into contact with MDI. However, MDI has a low
478 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488

Table 2
Typical physical properties MDI isomers, pMDI and formaldehyde.

References Parameter Unit 4,4 MDI pMDI HCHO

(a) Salthammer et al. (2010) Physical state NA Solid Liquid Gas

(b) Occupational Safety & Health Colour @RT NA White to yellow Brown Colourless
Administration, formaldehyde Molecular weight [g/mol] 250.3 (250.3)n 30.03
exposure evaluation Boiling point [C] @ 101.3 kPa >300 >300 96 (37%solution)
(c) Allport et al. (2003b) Melting point [C] 40 5 15 (37%solution)
Vapour pressure @20 C 6.2 104 3.1 104 1.23

Fig. 3. MDI structure (see Table 1).

vapour pressure at room temperature leading to low emissions and

with appropriate control measures in place, airborne concentra-
tions in the workplace are well below the recommended exposure
limits. Fig. 4. pMDI structure (see Table 1).

Total inhalable particulates (assumed in general to be constituted

of 100% wood dust)

A dusty environment can cause respiratory health effects (ne

particles penetrate deep into the lungs, more remain in nose
and throat) (Government of Alberta, 2009). Wood dust (HSE
Information Sheet, 1997) may cause irritation and allergic skin
and respiratory symptoms (sensitisation) and some wood species
are known to cause cancer, i.e. hardwood such as red western
cedar (Government of Alberta, 2009). In Europe, typically soft wood
species are used to produce CWP.
Determining TIP hazards and health risks is complex and
depends, in addition to exposure duration and airborne concentra-
tions, heavily on the dust properties (size, weight, shape, moisture
content and the wood species used) (Government of Alberta,
2009).
Fig. 5. Formaldehyde structure (see Table 1).
Under IARC (International Agency for Research on Cancer),
formaldehyde is designated as a suspected human carcinogen.
Studies from the IARC and recent studies have identied formalde-
hyde as potentially causing leukemia.
However, it must be noted that different authorities in different
countries and organisations as well as other epidemiologists have
different views on the conclusions made by IARC.
Exposure limits

Formaldehyde Inhalation exposure risks are often assessed by using air-

Formaldehyde (Fig. 5) is a strong irritant and sensi-
tizer in humans and animals (U.S. Environmental Protection
Agency; NIOSH, 2005; EHSO, 2009). Low-level exposure
to formaldehyde has been known to cause irritation of the
eyes, nose and throat (Salthammer et al., 2010). Prolonged expo-
sure may cause various symptoms, including prolonged eye, nose,
and throat irritation, coughing, wheezing, diaorrhea, nausea,
vomiting, headaches, dizziness, lethargy, irritability, disturbed
sleep, olfactory fatigue and skin irritation.
monitoring techniques whereby the results are compared to
international/national occupational exposure limits (OEL) (Table 3).
The exposure limits for formaldehyde vary quite a lot within
Europe. In almost all EU countries, limits for occupational expo-
sure to formaldehyde already exist and are between 0.37 and
0.62 mg/m3. The 3rd EU Scientic Committee on Occupational
Exposure Limits (SCOEL) list proposes 0.25 mg/m3. This limit is
considered unjustiable low by other stakeholders and a limit of
0.37 mg/m3 seems to be the most appropriate solution for this
industry sector.

Table 3
Typical exposure limits for MDI, TIP and formaldehyde.

References Country MDI TIP Formaldehyde

mg/m3

8-h TWA STEL 8-h TWA STEL 8-h TWA STEL

UK, EH40/2005, expressed as NCO (2007) (UK MDI exposure UK 0.02 0.07 5 NA 2.5 2.5
limits expressed as total isocyanate)
www.ispesl.it, based on ACGIH (American Conference of Italy 0.051 NA 5 NA 0.6 NA
governmental Hygienists)
Arbeidplatzgrenzwerte TRGS 900 (2006) Germany 0.05 0.05 2 8 0.6 NA
KB 06 juni 2009 Belgium 0.052 NA 3 NA NA 0.38
American Conference of governmental Hygienists ACGIH 0.051 NA 5 NA 0.6 NA
Scientic Committee on Occupational Exposure Limits SCOEL NA NA NA NA 0.25 NA
Recommendations (2008)
 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488 479

For this evaluation 0.37 mg/m3 has been used and not the newly
proposed limit of SCOEL of 0.25 mg/m3. This limit is also rec-
ommended by The Deutsche Forschungemeinschaft (DFG) and
German MAK Commission. In addition, it was also considered
more appropriate as this exposure limit or lower limit values
are already applicable in some countries. The exposure limit for
formaldehyde is 7 times higher than the exposure limit for MDI
(Table 3).
Air sampling methods and equipment
MDI
A known volume of the workplace air is drawn through
the sampling device (IOM sampler and 25 mm 1-2MP (1- (2-
methoxyphenyl) piperazine) impregnated lter) at a xed
sampling rate (2 l/min) and duration (According to ISO 16702 and
UK HSE method MDHS 25/3), typically over a half or full work-shift.
Any airborne MDI in the air reacts with the derivatising reagent
(1-2MP) on the lter to form a stable non-volatile urea derivative.
After sampling, lters are desorbed in a glass vial with 2 ml 1-2MP-
toluene-solution and transported to the laboratory for analysis.
The analysis of the samples is carried out in a dedicated Indus-
trial Hygiene (IH) laboratory (The Huntsman dedicated laboratory,
located at Everberg, Belgium, has a record of excellent performance
since 1994 in the prociency tests scheme (Workplace Analysis
Scheme for Prociency, i.e. WASP) following the ISO 17025 guide-
lines, using a liquid chromatograph (LC) equipped with a ultraviolet
(UV)/mass spectrometer (MS) detection system. Samples requiring
target analysis are screened for the m/z target ions in each time
segment as dened by the UV retention times. The [M+H]+ ion for
derivatised PI (Phenyl Isocyanate) (m/z 312), the [M+2H]2+ ion for
MDI (m/z 318), and the[M+3H]3+ ion for TRI (Triisocyanate) (m/z
320) are selected as precursor ions for subsequent MS analysis.
Quantication of the 1-2MP derivatives is performed by summing
the intensities of the fragment ion m/z 193 [1-2MP+H]+ and of the
complementary fragment ion corresponding to the residual iso-
cyanate moiety, i.e. the ion due to expulsion of 1-2MP, [M192+H]+.
(Vangronsveld and Mandel, 2003).
Total inhalable particulates
A known volume of workplace air is drawn through the sampling
device (IOM sampler equipped with preconditioned preweighed
25 mm 0.8 m membrane lter) at a xed sampling rate (2 l/min)
and duration (According to UK HSE method MDHS 14/3), typically,
i.e. over a half or full work-shift. After sampling, the lters are stored
in a sealed container for transport to the laboratory. At the labora-
tory, the lters are reconditioned and reweighed. The difference
in mass of the lters, between preweighing and after sampling,
divided by the sample volumes provides the mass/volumetric con-
centration of TIP and allows comparison with the occupational
exposure limits set by the authorities.
Formaldehyde
NIOSH Manual of Analytical Methods (NIOSH, 2003), suggests
that the results of vapour-phase and formaldehyde-containing par-
ticulates should be reported separately until sufcient data has
been collected to allow appropriate epidemiological interpreta-
tion of formaldehyde-containing particulate exposures. This type
of monitoring is normally not carried out routinely and it remains
uncertain if formaldehyde at the CWP workplaces is only present
as vapour or as an aerosol form.
The data has been obtained from various sources (Canada
Monitoring, 2006; Lavou et al., 2006; Discussion Paper, 2007;
Lavou et al., 2005, IRSST and Acton, 2009) where formaldehyde
has been sampled and analysed using a widely used method based
on DNPH (2,4-dinitrophenyl hydrazine) derivatisation and GC or LC
for identication/quantication.
DATA formatting
Personal measurements are associated with the work/tasks per-
formed by the personnel being sampled, area/zone measurements
represent the air quality at key locations in the workplace and
potential exposure risks in these areas. The functions and areas
in Table 4 have been studied. Where results are reported as below
limit of detection (LOD) or quantication (LOQ), 50% of the LOD or
LOQ value has been used in line with NIOSH guideline (Hornung
and Reed, 1990; Ogden, 2010).
Statistical interpretation
The overall Geometric Mean (GM), Geometric Standard Devi-
ation (GSD), probability of exposure (%P) and Risk Factor (RF) is
calculated for the complete dataset, containing all CWP applica-
tions and for all separate CWP application (OSB, MDF, WFI and
PB).
GM, GSD and %P are calculated using IHSTAT (IHSTAT from AIHA
EASC book, author 3M). The RF is calculated by dividing the GM by
the relevant OEL and is an indication of exposure safety.

Table 4
Functions and areas in the reconstituted wood panel industry, predened by the IH CoE group.

Predened functions Description Predened areas Description

Band saw operator Operates at the saw Weigh belt Weighing the wood
Cleaner Worker operating the mechanical sweeper throughout Blender Blending wood and mixing with MDI
the facility, cleaning machines, etc.
Dryer operator Operates at the dryer machine [usually located prior to Conveyor Transporting the wood particles from
introduction of MDI] blender to former area
Electrician Electrician working inside the facility Former bin area Temporary storage bin MDI with Wood,
above former line
Line operator Operates at the production line Former area Area at the former line
Packaging operator Works at the packaging area Mat area Mat area between former area
Press operator Operates the press Press infeed Press entrance (mat)
QC operator Operating the QC samples Press area Pressing the mat (curing/reacting)
Rener operator Operates at the rene area Press out feed Press exit (compressed boards)
Saw operator Operates at the band saw
Supervisor Team leader (during a shift or in an area) Saw area Sawing the wood panels
Tongue and groove operator Operates at the tongue and groove Boardcoolers Cooling of the fresh pressed wood panels
General All none predened jobs, doing different jobs in the Tongue and groove Sawing activity (click-t system)
workplace
480 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488

Table 5
Air-monitoring programme recommendations based on P-value.
If P 0.1% The exposure is considered below the OEL, no further
monitoring is necessary, unless there are changes to
procedures and/or processes.
If 0.1 < P < 5% The exposure is considered apparently below the OEL;
this must still be conrmed by periodic measurements.
If P 5% The probability that the OEL will be exceeded is high.
New or improved control measures must be
implemented or veried to demonstrated their
efciency to reduce the exposure; new measurements
must be carried out to verify the effectiveness of the
new/improved control measures.
A Risk Factor >1, i.e. no exposure safety margin, is considered as
exceeding the OEL, whereas Risk Factor 1 is considered as compli-
ant with the OEL. It should be noted that for the substances being
assessed it is always recommended that exposure safety margins
are as high as possible, i.e. the RF should be as low as possible and
preferably 0.1.
For each application a global GM has been calculated for func-
tions and areas (Lavou et al., 2005). The global GM for all functions
or all areas is calculated by multiplying the GM with the sample size,
divided by the total sample size (to weigh out the effect of sample
size).
According to the IH statistical programme used (IHSTAT from
AIHA EASC book, author 3M), a Geometric Standard Deviation
GSD3.5 can be regarded as acceptable. A GSD > 3.5 can occur
when:
(1) there are not sufcient sample data points,
(2) there are poorly dened homogeneous groups (=people with
same exposure prole),
(3) poor representative sampling/analysis, e.g. changes in pro-
cess, activities, control measures, or use of different sam-
pling/analysis techniques.
The P-value calculations are based on a 95th percentile. An
exposure probability value, i.e. a P-value, is calculated for every
function/area (IHSTAT from AIHA EASC book, author 3M), which
can be interpreted in different ways:
(1) indication of a % probability of results exceeding the substance
specic OEL;

Table 6
Personal data samples for TIP and MDI, for all apllications (OSB, MDF, WFI and PB).

Function Product Sample size GMa TIP (mg/m3) GM MDI (mg/m3) GSDb TIP GSD MDI %Pc TIP %P MDI RFd TIP RF MDI

Band Saw Operator OSB 10 0.34 0.0002 2.15 1.87 1.1 0.1 0.17 0.004

Total
OSB 51 0.69 0.0011 2.83 5.58 15.2 1.3 0.34 0.022
Cleaner 40 0.88 0.0019 3.26 5.90 24.5 3.3 0.44 0.038
MDF 10 0.35 0.0002 1.77 4.77 0.1 0.1 0.17 0.004
WFI 1 0.82 0.0004 NAe NAe NAe NAe 0.41 0.007

Total 5 0.71 0.0002 2.07 4.21 7.8 0.1 0.36 0.004

Dryer operator OSB 4 0.93 0.0002 NAe NAe NAe NAe 0.46 0.004
e e e e
MDF 1 0.25 0.0001 NA NA NA NA 0.12 0.002

Total 21 0.49 0.0001 2.70 5.82 7.4 0.1 0.25 0.002

Electrician OSB 20 0.50 0.0001 2.77 6.09 8.7 0.1 0.25 0.002
MDF 1 0.38 0.0001 NAe NAe NAe NAe 0.19 0.002

Total 22 0.66 0.0002 2.59 7.16 12.3 0.2 0.33 0.004

General operator OSB 20 0.69 0.0001 2.68 5.89 14.2 0.1 0.35 0.002
MDF 2 0.41 0.0009 NAe NAe NAe NAe 0.21 0.018

Total 36 0.48 0.0003 1.94 4.67 1.6 0.1 0.24 0.006

OSB 27 0.48 0.0003 2.02 4.63 2.2 0.1 0.24 0.006
Line operator MDF 3 0.41 0.0001 NAe NAe NAe NAe 0.21 0.002
WFI 2 0.40 0.0001 NAe NAe NAe NAe 0.20 0.002
PB 4 0.66 0.0010 NAe NAe NAe NAe 0.33 0.02

Packaging operator OSB 5 0.44 <0.0001 1.90 4.29 0.9 0.1 0.22 <0.001

Total 44 0.27 0.0006 NL7 4.15 11.8 0.1 0.14 0.012

OSB
Press operator 33 0.27 0.0006 6.82 4.08 15.1 0.1 0.14 0.012
MDF 10 0.25 0.0007 1.88 4.34 0.1 0.2 0.12 0.014
WFI 1 0.49 0.0001 NAe NAe NAe NAe 0.24 0.002

QC operator OSB 5 0.89 0.0015 4.01 3.46 27.9 0.2 0.44 0.03

Rener operator MDF 3 0.36 0.0002 NAe NAe NAe NAe 0.18 0.004

Total 16 0.64 0.0003 NL7 3.89 3.7 0.1 0.32 0.006

Saw operator OSB 11 0.55 0.0002 1.51 3.38 0.1 0.1 0.28 0.004
WFI 5 0.88 0.0009 2.60 3.65 19.4 0.1 0.44 0.018

Total 10 0.27 0.0006 2.37 4.29 0.9 0.1 0.13 0.012

OSB
Supervisor 3 0.57 0.0005 NAe NAe NAe NAe 0.28 0.010
MDF 4 0.15 0.0009 NAe NAe NAe NAe 0.07 0.018
WFI 3 0.28 0.0006 NAe NAe NAe NAe 0.14 0.012

Total 8 0.46 0.0002 2.14 5.11 2.7 0.1 0.23 0.004

T&G operator OSB 4 0.42 0.0001 NAe NAe NAe NAe 0.21 0.002
WFI 4 0.50 0.0004 NAe NAe NA7 NAe 0.25 0.008
a
Geometric mean.
b
Geometric standard deviation.
c
% Probability of exposure.
d
Exposure risk factor.
e
NA: not enough samples for statistical analysis; NL: not Lognormal distribution and value not taken into consideration.
 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488 481

(2) indication of % of time that somebody is probably exposed to a Discussion

concentration above the OEL during a day or
(3) indication of % of personnel that are probably exposed above the
OEL continuously at a specic location or carrying out a specic
activity.
In addition using the %P to assess the probability of exposure,
it can also be used to develop an air-monitoring programme, as dis-
played in Table 5. The OEL is used to assess compliance with typical
OELs, i.e. 2 mg/m3 for TIP and 0.05 mg/m3 for MDI. Although there is
a lot of variance between the formaldehyde OELs of different Euro-
pean countries, as explained above 0.37 mg/m3 has been selected
as the most appropriate limit value.
Results
Personal air-monitoring MDI and TIP (Table 6).
Zone monitoring MDI and TIP (Table 7).
Formaldehyde functions (Table 8).
Formaldehyde zones (Table 9).
Factors inuencing the reliability of data
Having limited number of samples per function or area can
result in a higher GSD as a consequence of poorly dened homoge-
nous group. Other factors which can have an impact on the data are
chemical factors (e.g. amount and/or % pMDI used, physical state
of MDI during sampling), operational conditions (e.g. press pro-
les, conveyor speed, temperature variability, thickness of boards
pressed, wood species used, level of enclosure of the line, the
ventilation effectiveness, wood moisture etc.), and, not to be under-
estimated, the adminstrative management systems determining
the health and safety culture (behaviour) in every plant etc.
It is clear that the level of containment in combination with
the ventilation at the different stages of the production pro-
cess, which can differ substantially in wood panel manufacturing
plants, is a determining factor on the level of exposure and there-
fore on the reliability of the TIP, MDI/formaldehyde on wood,
MDI/formaldehyde vapour concentrations measured for each prod-

Table 7
Area sample data for TIP and MDI, for all applications (OSB, MDF, WFI and PB).

Zone Product Sample GM TIP GM MDI GSD TIP GSD MDI %P TIP %P MDI RF TIP RF MDI
size (mg/m3) (mg/m3)

Weigh belt Total 4 0.46 0.0040 NAa NAa NAa NAa 0.23 0.08
OSB 3 0.52 0.0050 NAa NAa NAa NAa 0.26 0.10
MDF 1 0.32 0.0020 NAa NAa NAa NAa 0.16 0.04

Blender area Total 18 0.55 0.0020 2.80 3.31 10.4 0.5 0.27 0.04
OSB 14 0.71 0.0020 2.55 2.81 13.4 0.1 0.36 0.04
WFIB 4 0.22 0.0070 NAa NAa NAa NAa 0.11b 0.14b

Conveyor (blender to former area) Total 13 0.97 0.0020 4.29 2.06 30.8 0.1 0.48 0.04
OSB 11 1.34 0.0020 3.76 2.04 38.2 0.1 0.67 0.04
WFIB 2 0.58 0.0040 NAa NAa NAa NAa 0.29 0.08

Former bin area Total 35 0.69 0.0040 2.55 3.66 12.6 2.4 0.34 0.08
OSB 30 0.86 0.0040 2.23 3.80 14.6 3.1 0.43 0.08
MDF 1 0.24 0.0020 NAa NAa NAa NA 0.12 0.04
WFIB 4 0.16 0.0030 NAa NAa NAa NAa 0.08 0.06

Former area Total 38 0.57 0.0020 1.76 3.44 1.3 0.8 0.29 0.04
OSB 35 0.62 0.0020 1.67 3.37 1.1 0.6 0.31 0.04
MDF 1 0.26 0.0090 NAa NAa NAa NAa 0.12 0.18
WFIB 2 0.22 0.0020 NAa NAa NAa NAa 0.11 0.04

Mat area (between former and Total 3 0.17 0.0010 NAa NAa NAa NAa 0.09 0.02
press) WFIB 3 0.17 0.0010 NAa NAa NAa NAa 0.09 0.02
Press area Total 17 0.39 0.0020 2.84 4.48 5.9 1.0 0.20c 0.04
OSB 15 0.43 0.0010 2.93 4.71 7.7 1.0 0.22c 0.02
WFIB 2 0.19 0.0040 NAa NAa NAa NAa 0.09c 0.08

Press infeed Total 6 0.72 0.0020 2.15 1.80 9.0 0.1 0.36c 0.04
OSB 5 0.90 0.0020 NLa 1.91 8.7 0.1 0.45c 0.04
MDF 1 0.23 0.0020 NAa NAa NAa NAa 0.12c 0.04

Press outfeed Total 11 0.56 0.0010 2.27 4.38 6.0 0.5 0.28c 0.02
OSB 8 0.77 0.0020 1.92 2.97 7.2 0.1 0.39c 0.04
MDF 2 0.26 0.0004 NAa NAa NAa NAa 0.13c 0.008
WFIB 1 0.19 0.0003 NAa NAa NAa NAa 0.09c 0.006
a
Saw area Total 9 0.18 0.0008 NL 8.73 0.1 2.7 0.09 0.016
OSB 4 0.15 0.0010 NAa NAa NAa NAa 0.08 0.02
MDF 1 0.15 0.0001 NAa NAa NAa NAa 0.08 0.002
WFIB 4 0.21 0.0021 NAa NAa NAa NAa 0.10 0.04

Boardcoolers Total 3 0.13 0.0002 NAa NAa NAa NAa 0.07 0.004
OSB 2 0.10 0.0002 NAa NAa NAa NAa 0.05 0.004
MDF 1 0.22 0.0002 NAa NAa NAa NAa 0.11 0.004

T&G area Total 7 0.22 0.0001 1.67 4.87 0.1 0.1 0.11 0.003
OSB 5 0.26 0.0001 1.69 5.64 0.1 0.1 0.13 0.002
WFIB 2 0.15 0.0003 NAa NAa NAa NAa 0.08 0.006
a
NA: not enough samples for statistical analysis; NL: not lognormal distribution and value not taken into consideration.
b
Mainly caused by 1 sample where no ventilation is present, without this sample the RF would be 0.07 for TIP and 0.08 for MDI.
c
TIP (instead of wood dust) caused by the presence of other organic compounds, i.e. hydraulic oil particles, etc.
 482
Table 8
Overview functions formaldehyde results, compared to MDI results (IH CoE measurements).

E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488
References Function Product Formaldehyde MDI Formaldehyde MDI

[a] [b] [c] [d] [e] [a] [b] [c] [d] [e]
TWA-8 h (mg/m3) RF

[a] = Canada Monitoring (2006) 0.047 0.049-0.0

86 0.0006 0.13 0.130.23 0.01
[b] = Lavou et al. (2006) OSB 0.049 0.13
[c] = Discussion Paper (2007) 0.086 0.23
Press operator assistant
[d] = Lavou et al. (2005) 0.37 0.110.33 0.0007 1 0.30.9 0.01
[e] = Acton (2009) MDF 0.11 0.3
0.33 0.9

QC operator OSB 0.051 0.0015 0.14 0.03

QC technician OSB 0.046 0.0015 0.12 0.03

0.016 0.0370.098 0.0006 0.04 0.100.27 0.01

OSB 0.037 0.1
0.098 0.27
Press operator
MDF 0.0007 0.01
0.086 0.23
Global 0.066 0.18

Dreyer operator Global 0.062 0.0002 0.17 0.004

Bander operator OSB 0.0381 0.0002 0.1 0.004

Giben + Bander OSB 0.0157 0.0002 0.04 0.004

operator

Tongue and groove OSB 0.0082 0.0001 0.02 0.0028

operator

0.0257 0.0740.098 0.0019 0.07 0.200.27 0.04

OSB 0.061 0.17
Cleaner 0.66 1.8
0.23 0.140.15 0.0002 0.63 0.370.40 0.004
MDF
0.098 0.27
Global 0.479 0.0011 1.29 0.022
 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488 483

uct sector. This is apparent where only a limited number of plants

for a particular product sector have been surveyed, for example,

0.004c

0.004c

0.004

0.02c
only two for WFI, the effect of the availability and/or efciency of

0.04

0.04
0.02

0.08

0.08

0.04
MDI

control measures will have a higher impact on these results.

There are factors inuencing the reliability of the formaldehyde
results used in this article (Canada Monitoring, 2006; Lavou et al.,
3.25

0.96

0.30
2006; Discussion Paper, 2007; Lavou et al., 2005, Acton, 2009).
[e]

Some of the reports merely mention a range of formaldehyde con-
0.11 centration measured, which can be regarded as an indication only.
0.32

2.05
In such circumstances, the mean value has been taken.
0.7
[d]

0.731.11

Inhalation exposure risk per type of CWP

0.3
[c]

Based on the global GMs and RFs calculated and displayed in
Table 10, the MDI concentrations across the applications for func-
Formaldehyde

tions and areas can be regarded as similar, with exception of the

[b]

WFI saw operator, OSB Cleaner and OSB QC Operator, which have
a slightly lower exposure safety margin (higher risk) compared to
b
0.99a

1.62

0.13

0.11

0.17

the other functions.

0.2
[a]
RF

This interpretation should be regarded as an indication only,

since the number of samples taken for OSB is in general consider-
ably higher compared to the other applications. The global GM/GSD
0.0002c

0.0002c

0.0002

0.001c
0.002

0.002
0.001

0.004

0.004

0.002
MDI

is calculated by multiplying the GM/GSD with the sample size (to

weigh out the effect of the number of samples), divided by the total
sample size per application.
In general it can also concluded that the global exposure RF for
1.20

0.36

0.11
[e]

TIP and MDI is the lowest for MDF for both functions and areas, and

more or less equal for WFI and OSB (Table 10).

0.04
0.12

0.26
0.76
[d]

Inhalation exposure risk for workers

0.270.41

Cleaners and QC Operators are the functions with an overall

0.11
[c]

increased exposure risk to TIP and MDI, for most of the wood panel

applications. However, the exposure risk to MDI for all functions is

TWA-8 h (mg/m3 )

well below the OEL (Occupational Exposure Limit). The TIP expo-
Formaldehyde

[b]

sure risk is higher than the exposure risk to MDI since the RF is

higher (Table 10). Therefore, since MDI coated on wood particles is

Overview area formaldehyde results, compared with the MDI results (IH CoE measurements) for areas.

b
0.3672a

0.5985

0.0478

0.0407

0.0736

0.0615

one of the major contributing exposure routes, the exposure risk for
[a]

both TIP and MDI can be reduced by reducing the TIP concentrations

in the workplace air.

Cleaners and QC Operators have the highest RF for MDI (see
Application

Table 6), irrespective of which application, while for TIP the cleaner,
dryer, QC and saw operator have the highest RF.
Global

Global

Global
MDFd
MDFe
OSBd
OSBe
MDF

It is well known that exposure risks increases if air pressure or

OSB

OSB

OSB

OSB

OSB

OSB

brushing are used and not vacuum cleaning techniques for clean-
ing; therefore it is recommended that vacuum techniques be used
wherever possible.
Bottom former bin

Some plants have introduced a very effective piping system at

Top former bin

the forming line, which can be connected at the bottom with a

Boardcoolers

Former line
Press begin

mobile vacuum unit at the ground oor and at the top with long
Saw area

exible tubing to clean oors and the installation.

Press
Area

end

The QC operator has probably a higher exposure risk because

they collect samples at the production line, which will cause the
generation of airborne dust during this handling.
The WFI saw operator has a higher exposure risk to TIP and MDI,
this is probably because WFIB are light density type of boards, i.e.,
not pressed as dense as OSB, MDF boards or PB.
[a] = Canada Monitoring (2006)

Average of 6 measurements.
Average of 2 measurements.
[c] = Discussion Paper (2007)

It must also be noted that after the press most, if not all, of the
MDI should be reacted away (cured), consequently the functions
[b] = Lavou et al. (2006)

[d] = Lavou et al. (2005)

Governmental project.
Less than 5 samples.

involved at the stage following pressing should have a very low RF

Research project.

(Risk Factor). The job function rst in contact with freshly pressed
[e] = Acton (2009)

boards is the saw operator, followed by the tongue and groove oper-
ator and the packaging operator. The RFs for these functions for OSB
References

are 0.004, 0.002 and <0.001. Since the saw and tongue and groove
Table 9

operations are normally carried out in the vicinity of the produc-

e
b

d
a

tion line, these can be regarded as background RFs. Packaging is

484 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488

Table 10
Global GM per application for MDI and TIP.

Substance Application GSD % Global geometric mean (mg/m3) /risk factor

Functions Areas Functions Areas

GM RF GM RF

TIP OSB 2.68 2.29 0.55 0.27 0.70 0.35

MDF 1.86 NLa 0.31 0.15 0.24 0.12
WFI 2.60 NLa 0.58 0.29 0.22 0.11

MDI OSB 4.70 3.46 0.0007 0.014 0.0023 0.046

MDF 4.76 NLa 0.0004 0.008 0.0020 0.040
WFI 3.65 NLa 0.0005 0.001 0.0030 0.060
a
NL: not lognormal.

normally carried out in a separate vicinity and as can be seen the
RF can be regarded as zero.
Areas with an increased inhalation exposure risk
Based on the results in Table 7 the conveyor and press infeed
have the highest RFs for TIP (0.48 for conveyor and 0.36 for press
infeed), while for MDI, the former area has the higher potential for
exposure with a risk factor of 0.18 (MDF, 1 sample size) followed
by the former bin area and weigh belt with a RF of 0.08.
The fact that after the press most of the MDI should be reacted
away, as is stated above for the personal samples, is supported by
the area samples. The RF gradually reduces from the saw area (RF of
0.016), to the board cooling area (0.004) to the tongue and groove
area (0.003).
When MDI cures, it predominantly reacts with water to pro-
duce a non-volatile urea. There is also a substantial reaction with
this urea to produce biurets. These two reaction products dominate
the reaction product mixture by far. It could be possible that there
still are ultralow (ppb levels) of residual NCO groups present. The
residual NCO groups are predominantly not extractable and so it is
assumed that the NCO bearing molecule is anchored somehow to
the wood.
Another benet of using pMDI as a binder is the VOC emis-
sions during pressing are likely to be lower compared to the press
emissions when using ureaformaldehyde-based resins, especially
methanol and of course formaldehyde (Jian et al., 2002). This can be
explained by the fact that pMDI binds chemically to other reactive
species present in wood whereas ureaformaldehyde resins only
have a mechanical bonding mechanism.
Relationship TIP and MDI
It is clear from the results in Tables 6 and 7 that all the GM values
compared to the relevant OEL result in RFs for personal and area
samples below 0.5 for TIP (1 mg/m3 TIP), below 0.2 for MDI area
samples (0.01 mg/m3) and below 0.04 (0.002 mg/m3) for personal
MDI samples.
This can be interpreted that the exposure safety margin for per-
sonal exposure to MDI is more than an order of magnitude higher
(i.e. less exposure risk) than for TIP and that the exposure safety
margin at areas investigated is a factor of circa 2 higher for MDI
compared to TIP.
Fig. 6 combines the TIP and MDI concentrations for the functions
(Fig. 6A regardless of which composite wood products are being
manufactured and Fig. 6BD per composite wood product being

Fig. 6. MDITIP relationship for functions.

 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488 485

Fig. 7. MDITIP relationship for areas.

manufactured). Fig. 7 combines the TIP and MDI concentrations
for the areas (Fig. 7A regardless of which composite wood products
are being manufactured and Fig. 7BD per composite wood product
being manufactured).
Figs. 6 and 7 demonstrate the level of concentrations of wood
dust and MDI from the moment MDI is introduced in the production
process through to where the MDI is expected to be reacted away
in most cases if not all, regardless of the product being produced.
The highest concentrations for TIP can be found in general at the
conveyor and press infeed, for MDI this is the weigh belt and former
bins. The higher values found at the weigh belt are because this was
a relatively uncontained area (nearby the blenders) and the higher
values at the former bins is most probably due to the turbulence
(observed during the survey) caused by dropping MDI coated wood
akes on the mat. At the press, MDI reacts due to high temperatures
and pressure.
In general, as can clearly be derived from Fig. 7A (areas), the
high concentrations of TIP in the workplace air are associated with
a higher exposure risk to MDI, which appears logical since the MDI
is present on wood bres or particles from the addition of the MDI
onto the wood bres up to the hot pressing of the boards. This
effect is more pronounced for OSB, slightly less for WFI and less
pronounced for MDF; however, the dataset for MDF is relatively
small.
Effect for full-MDI and core-MDI for areas in the OSB industry
TIP
The TIP concentrations (Table 11 and Fig. 8A and B) for full-
MDI (MDI in all layers) and core-MDI (MDI only in the core
layer, not in the face layers) show a signicant statistical differ-
ence in TIP concentrations between full- and core-MDI (one-tailed
t-test: P = 0.009 see Fig. 9, Boxplot A). The area responsible
for the signicant difference cannot be determined due to the
too limited datasize for full-MDI and consequently the statisti-
cal analysis should be regarded as an indication only until the
datasize is representable enough. Based on Fig. 8B, one area has
higher GM when using full-MDI, i.e. the former bin area with no

Table 11
TIP and MDI differences between full-MDI and core-MDI for the OSB application.

Zone Sample size Full-MDI (mg/m3) Core-MDI (mg/m3) Full/core ratio RF - full-MDI RF - core-MDI

GM TIP GM MDI GM TIP GM MDI TIP MDI TIP MDI TIP MDI

Weigh belt 3 NAa NAa 0.5170 0.0046 NAa NAa NAa NAa 0.26 0.10
Conveyer area 11 0.8534 0.0021 0.8742 0.0022 0.9 1.0 0.38 0.04 0.44 0.04
Former bin area 30 1.2227 0.0065 0.7860 0.0037 1.6 1.8 0.61 0.13 0.40 0.07
Former area 35 0.6110 0.0044 0.6213 0.0022 1.0 2.0 0.31 0.09 0.31 0.04
Press area 16 NAa NAa 0.3882 0.0013 NAa NAa NAa NAa 0.19 0.03
Press infeed 5 0.8457 0.0012 0.9170 0.0024 0.9 0.5 0.42 0.02 0.46 0.05
Press outfeed 8 0.9137 0.0020 0.7323 0.0015 1.2 1.3 0.46 0.04 0.37 0.03
Saw area 4 NAa NAa 0.1532 0.0010 NAa NAa NAa NAa 0.08 0.02
Boardcoolers 2 NAa NAa 0.1011 0.0002 NAa NAa NAa NAa 0.05 0.004
T&G area 5 NAa NAa 0.2600 0.0001 NAa NAa NAa NAa 0.13 0.002
Sample size 119 19 100

NA: not enough samples.

486 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488

Fig. 8. Full/core differences in MDI and TIP for OSB.

signicant statistical difference in TIP concentrations (one-tailed Fig. 8A and B, where an increased concentration of TIP demon-
t-test: P = 0.23 see Fig. 9, Boxplot B). The higher GM for TIP strates that because the wood is coated with MDI there is an
for full-MDI could be explained by the fact that considerably less increased airborne concentration of MDI at this location (see
MDI binder is used compared to the level of ureaformaldehyde- below).
based resin used for the face layers of the boards leading to
less tacky or moist wood strands/particles and have therefore an
increased probability to become airborne. This is conrmed in

Fig. 9. Boxplot on t-test detemrinations (Xa: full-MDI, Xb: core-MDI, Y: LogNorm result, Boxplot A: total TIP, Boxplot B: former bin area for TIP, Boxplot C: total MDI, Boxplot
D: former bin area for MDI, Boxplot E: former area for MDI).
 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488
Table 12
Global GM all applications pooled for MDI, TIP and formaldehyde.
Substance Reference Application
TIP Global
MDI Global
Formaldehyde Lavou et al. (2005) Global
MDI
18 MDI samples have taken during production where MDI has
been used in all the layers of the OSB product and 100 samples
where MDI has been used in only the core layer.
The MDI (Table 11 and Fig. 8A and B) concentrations for full-MDI
and core-MDI show no signicant statistical difference in MDI con-
centrations between full- and core-MDI (one-tailed t-test: P = 0.21
see Fig. 9, Boxplot C), even not at the former bin area (one-tailed
t-test: P = 0.36 see Fig. 9, Boxplot D) and the former area (one-
tailed t-test: P = 0.25 see Fig. 9, Boxplot E), although, based on
Fig. 8A, the GMs of these areas indicate higher exposure risks to
MDI when using full-MDI. It could indicate that full-MDI OSB pro-
cesses require increased engineering controls, i.e. enclosed former
bins and enclosing former areas. However, the RF for MDI is still
acceptable and is not a determining factor for increasing exposure
controls, whereas this is considered to be more appropriate to lower
the TIP concentrations (see above).
General
19 TIP samples were taken during production where MDI has
been used in all the layers of the OSB product and 100 samples
where MDI has been used in the latter core layer only. It must be
noted that the small datasize of full-MDI do not allow for a reliable
statistical interpretation. However, the statistical results above can
be considered as an indication.
It is recommended that between the different former bins, the
former areas are covered or more contained, preferably connected
with existing ventilation systems, which are usually designed to
be efcient over the complete open former area. This will result
in reducing the airborne concentration of the agent with the high-
est exposure risk, i.e., TIP and as a consequence, may also reduce
the exposure risk to MDI and formaldehyde. This is normally a
low cost improvement with a high effect, and if designed well it
should enable easy access for maintenance or other interventions.
Initial experiments at some customers have indicated that cover-
ing the former area with low cost material signicantly reduces
the airborne concentrations of TIP and hence the MDI in workplace
air.
Exposure safety margin MDI versus formaldehyde
Based on data received from CWP manufacturers and on liter-
ature research, reported formaldehyde results are used to obtain
an indication of the differences of exposure between MDI and
formaldehyde opposite the selected reference OELs. It should be
noted, that due to the limited amount of formaldehyde data and
information, e.g., the number of samples for formaldehyde, type of
ureaformaldehyde resins used, this evaluation is only an indica-
tion and a more scientic based comparison is needed.
Probably the most important factors affecting the exposure
safety margin are the difference of volatility between formaldehyde
and pMDI and the difference of % resins required for produc-
tion of wood panels ( 7% pMDI and typically between 10% and
15% ureaformaldehyde-based resins). This could potentially also
487
GSD % Global geometric mean (mg/m3)/risk factor
Functions Areas Functions Areas
GM RF GM RF
2.49 2.48 0.41 0.21 0.56 0.28
4.70 3.85 0.0004 0.009 0.002 0.04
2.3 3.1 0.21 0.57 0.35 0.95
results in less tack of the wood since pMDI binders do not
contribute to the moisture content of the wood compared to
formaldehyde which could result in higher airborne TIP concen-
trations; however, the results above in general do not support this
hypostesis.
The level of free formaldehyde in the resin itself is usually very
low, typically <0.2%. The main emissions of formaldehyde into the
workplace occur at the press stage where, due to condensation
reactions and emissions from the wood itself, substantial amounts
of formaldehyde are released.
The evaluation is only carried for those functions and areas for
which results were reported above detection limit for both com-
pounds. The functions are compared in Table 8 and areas in Table 9
(Canada Monitoring, 2006; Lavou et al., 2005, 2006; Discussion
Paper, 2007, IRSST and Acton, 2009). It can be concluded that the
RF of formaldehyde is higher than for MDI for both functions and
areas or otherwise stated, there is a higher probability that work-
ers may be exposed to formaldehyde levels close to or above the
formaldehyde OEL than personnel are being exposed to levels of
MDI close to the OEL.
Lavou et al. (2005), have calculated the global GM for
formaldehyde-based on 590 samples (mixed area and personal
samples) on 12 plants manufacturing OSB, MDF and PB. These val-
ues where used to compare (Table 12) the global TIP and MDI GM
values based on results of all functions and areas across all appli-
cations reported in Tables 6 and 7.
Conclusion
Based on the data represented, personal exposure risks to MDI
during production of CWP can be regarded as low to very low and
considerably less than the exposure risks to TIP and formaldehyde.
There are some areas where concentrations of MDI are more ele-
vated, mainly caused by increased wood dust levels in these areas.
The MDI concentrations in these areas can be further reduced by
minimising the airborne wood dust generation by containment of
the conveyors and former areas and/or investigating if the design
and/or efciency of the engineering controls can be improved (i.e.
by establishing regular maintenance and testing programmes). In
general, this is not required for MDI but more to lower the RF related
to wood dust and formaldehyde. This conrms previous studies
carried out in CWP production facilities (Maddison, 1998; Booth et
al., 2009). The exposure risks opposite the selected exposure limits
can be ranked as formaldehyde > TIP > MDI (highest margin of expo-
sure safety for MDI), indicating that the use of MDI also reduces
the exposure risks to workers during production of CWP compared
to the use of ureaformaldehyde type resins. The ranking for per-
sonal exposure risks remains equal even if an exposure limit of
0.6 mg/m3 and not the proposed SCOEL limit of 0.37 mg/m3 would
have been used. Exposure risks at the areas would result in a rank-
ing of TIP > formaldehyde > MDI when an OEL of 0.6 mg/m3 would
have been used.
To further reduce potential exposure risks to the binder/resins
and TIP, key task assessments should be carried out to investigate
488 E. Vangronsveld et al. / International Journal of Hygiene and Environmental Health 213 (2010) 475488
which tasks could cause peak and dermal exposures and appropri-
ate measures introduced to eliminate or reduce the risk, which will
also effectively reduce the long-term exposure potential. In addi-
tion, behaviour related exposures should be addressed by ensuring
personnel receive appropriate function related training in hazards
and risks for the tasks they carry out and to implement a quanti-
able behaviour inspection system.
Based on the limited data and information available, there is
an indication that the potential exposure risk to formaldehyde can
be regarded as high compared to MDI. Therefore, a production pro-
cesses that can operate at or below the OEL for formaldehyde should
have no problems meeting the same criteria for MDI, even for the
production of full-MDI OSB.
Exposure risk to MDI at workstations situated after pressing
(curing) can be considered low to extremely low and likely mainly
constituted by workplace emissions from elsewhere in the plant. In
addition, there are indications that VOC emissions in the exhaust air
at the press stage are lower for pMDI bonded boards and that there
are no relevant MDI emissions during use in buildings. Therefore
we can conclude that the use of MDI for CWP leads to an increased
exposure safety margin for workers, community nearby the plant
and building occupants.
Funding
This article is funded by the IH CoE laboratory.
Afftliation of authors with the industry
The authors are working for the Industrial Hygiene Cen-
ter of Excellence (IH CoE) laboratory, committed to provide
extensive ongoing support to ensure safe operating proce-
dures in the workplace to prevent dermal and/or inhalable
exposures, including carrying out workplace industrial hygiene
surveys to measure the concentrations of MDI and TIP/Wood
Dust.
The ongoing support to prevent exposure risks includes:
Assessing inhalation exposure risks by air measurements;
Assessing dermal exposure risks by surface sampling;
Assistance in obtaining local usage permits or authority audits;
and
Management and operator training to address workplace safety
and behaviour.
One of the authors (Erik Vangronsveld) is a recognised certied
Industrial Hygienist.
Conict of interest
The Industrial Hygiene Center of Excellence (IH CoE) labo-
ratory is an independent laboratory (requirement for ISO 17025)
from Huntsman, an MDI manufacturer. However, in accordance to
the ISO 17025 requirements, the IH CoE laboratory ensures that
under no circumstances any team member is inuenced by any
inducement by Huntsman personnel or from outside, which may
adversely affect his/her judgment or the validity of the results of
his/her work.
Acknowledgments
The authors would like to thank the Jan Vandenbrouck, and the
Huntsman CWP I-Bond technical service team for the additional
information about the CWP operations.
References
Acton B.C., 2009. European Formaldehyde Measurements Campaign at Reduction
of Formaldehyde Exposure in the Woodworking Industries (Feb 2010). CEI-Bois
assblAlderly Consulting Group, Project Number 29154, CEI-Bois & EPF, URL:
http://www.cei-bois.org/./REDUCTION OF FORMALDEHYDE EN.pdf.
Allport, D.C., Gilbert, D.S., Outterside, S.M., 2003a. MDI and TDI: Safety, Health and
the Environment. A Source Book and Practical Guide. J. Wilewt & Sons, New York,
pp. 229272 (Chapter 4).
Allport, D.C., Gilbert, D.S., Outterside, S.M., 2003b. MDI and TDI: Safety, Health and
the Environment. A Source Book and Practical Guide. J. Wilewt & Sons, New York,
p. 232 (Chapter 4).
Allport, D.C., Gilbert, D.S., Outterside, S.M., 2003c. MDI and TDI: Safety, Health and
the Environment. A Source Book and Practical Guide. J. Wilewt & Sons, New York,
p. 287 (Chapter 5).
Allport, D.C., Gilbert, D.S., Outterside, S.M., 2003d. MDI and TDI: Safety, Health and
the Environment. A Source Book and Practical Guide. J. Wilewt & Sons, New York,
p. 30 (Chapter 2), p. 302 (Chapter 5).
Allport, D.C., Gilbert, D.S., Outterside, S.M., 2003e. MDI and TDI: Safety, Health and
the Environment. A Source Book and Practical Guide. J. Wilewt & Sons, New York,
pp. 155157 (Chapter 3).
Karroll Booth, et al., 2009. Measurements of airborne methylene diisocyanate
(MDI) concentration in the U.S. workplace. J. Occup. Environ. Hyg. 4 (April (6)),
228238.
Canada Monitoring, 2006. Formaldehyde & MDI, xxx Inc. (condential data).
Discussion Paper, 2007. Changes to Occupational Exposure Limits for Formalde-
hyde, WorkSafe BC, URL: http://worksafebc.com/regulation and policy/
policy decission/board decission/2009/assets/fo.
Formaldehyde: Environmental, Health and Safety Information, 2009. www.
ehso.com.
Government of Alberta, 20092010. CH045 Chemical Hazards.
Hornung, R.W., Reed, L.D., 1990. Estimation of average concentration in the presence
of nondetectable values. Appl. Occup. Environ. Hyg. 5 (1), 4151.
HSE Information Sheet: Toxic Woods. Woodworking Sheet No 30, First Published
10/97, HSE.
Jian, T., et al., 2002. Volatile organic compound emissions arising from the hot-
pressing of mixed hardwood particleboard (composites and manufactured
products). Forest Prod. J. 52 (11-12), 6677.
Lavou, J., Goyer, N., Perrailt, G., Beaudry, C., Grin, M., 2005. Investigation of deter-
minants of past and current exposures to formaldehyde in the reconstituted
wood panel industry in Quebec. Ann. Occup. Hyg. 49 (7), 587602.
Lavou, J., Bgin, D., Beaudry, C., Grin, M., 2006. Monte carlo simulation to recon-
struct formaldehyde exposure levels summary parameters reported in the
literature. Ann. Occup. Hyg. 51 (2), 161172.
Maddison, P., 1998. Workplace air measurements in the wood panel industry: iso-
cyanates and total inhalable particulates. In: Second European Panel Products
Symposium, Llandudno, Wales, UK, 2122 October 1998.
NIOSH, 2003. Manual of Analytical Methods: Formaldehyde (Method Number 2541),
http://www.cdc.gov/niosh/docs/2003-154/pdfs/2541.pdf.
NIOSH, 2005. Pocket Guide to Chemical Hazards. NIOSH Publication, Formaldehyde.
Ogden, T.L., 2010. Handling results below the level of detection. Ann. Occup. Hyg. 54
(3), 255256.
Randall, D., Lee, S., 2002. The Polyurethanes Book. J. Wiley & Sons, New York, pp. 68
(Chapter 1).
Salthammer, T., et al., 2010. Formaldehyde in the indoor environment. Chem. Rev.
110 (4), 25362572, URL: http://pubs.acs.org/doi/full/10.1021/cr800399g.
United Kingdom Environment Agency, 2006. Secretary Guidance Note IPPC SG1
Integrated Pollution Prevention and Control (IPPC). Secretary Guidance for A2
Particle Board, Oriented Strand Boards and Dry Process Fibreboard Sector.
US LEED. Credit EQ-4.4: Low-Emitting Materials Composite Wood, Section
2: Individual LEED Credit Reviews, URL: http://www.tpub.com/content/
gsacriteria/gsaleed/gsaleed0416.htm.
U.S. Environmental Protection Agency, EPA Toxic Substance Control Act (TSCA). An
Introduction to Air Quality: Formaldehyde, http://epa.gov/iaq/formalde.html.
Vangronsveld, Erik, Mandel, Friedrich, 2003. Workplace monitoring of isocyanates
using ion trap liquid chromatography/tandem mass spectrometry. Rapid Com-
mun. Mass Spectrom. 17, 16851690.