Identifying thermal breakdown products of thermoplastics

HAL Id: hal-01666609https://hal.archives-ouvertes.fr/hal-01666609

Submitted on 18 Dec 2017

HAL is a multi-disciplinary open accessarchive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come fromteaching and research institutions in France orabroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, estdestinée au dépôt et à la diffusion de documentsscientifiques de niveau recherche, publiés ou non,émanant des établissements d’enseignement et derecherche français ou étrangers, des laboratoirespublics ou privés.

Identifying thermal breakdown products ofthermoplastics

Marianne Guillemot, Benoît Oury, Sandrine Melin

To cite this version:Marianne Guillemot, Benoît Oury, Sandrine Melin. Identifying thermal breakdown products of ther-moplastics. Journal of Occupational and Environmental Hygiene, Taylor & Francis, 2017, 14 (7),pp.551-561. �10.1080/15459624.2017.1302586�. �hal-01666609�

https://hal.archives-ouvertes.fr/hal-01666609

https://hal.archives-ouvertes.fr

1

Title: Identifying thermal breakdown products of thermoplastics

Journal: Journal of Occupational and Environmental Hygiene

Corresponding Author:

Marianne Guillemot, INRS, Pollutant Metrology Dept., 1 rue du Morvan, CS60027,

54519 VANDOEUVRE Cedex, France, [email protected]

Co-authors:

Benoît Oury, INRS, Pollutant Metrology Dept., 1 rue du Morvan, CS60027, 54519

VANDOEUVRE Cedex, France, [email protected]

Sandrine Melin, INRS, Pollutant Metrology Dept., 1 rue du Morvan, CS60027, 54519

VANDOEUVRE Cedex, France, [email protected]

Keywords: Plastic thermal degradation, analytical protocol, contaminant in plastic

molding shops

Word count: 3308

2

Identifying thermal breakdown products of thermoplastics

Marianne Guillemot, Benoît Oury, Sandrine Melin

INRS, Pollutant Metrology Dept., 1 rue du Morvan, CS60027, 54519 VANDOEUVRE

Cedex, FRANCE

ABSTRACT

Polymers processed to produce plastic articles are subjected to temperatures between

150°C and 450°C or more during overheated processing and breakdowns. Heat-based

processing of this nature can lead to emission of volatile organic compounds (VOCs) into the

thermoplastic processing shop. In this study, laboratory experiments, qualitative and

quantitative emissions measurement in thermoplastic factories were carried out. The first step

was to identify the compounds released depending on the thermoplastic nature, the

temperature and the type of process. Then a thermal degradation protocol that can extrapolate

the laboratory results to industry scenarios was developed. The influence of three parameters

on released thermal breakdown products was studied: the sample preparation methods –

manual cutting, ambient or cold grinding – the heating rate during thermal degradation – 5, 10

20 and 50°C/min – and the decomposition method – thermogravimetric analysis and

pyrolysis. Laboratory results were compared to atmospheric measurements taken at 13

companies to validate the protocol and thereby ensure its representativeness of industrial

thermal processing. This protocol was applied to most commonly used thermoplastics to

determine their thermal breakdown products and their thermal behaviour. Emissions data

collected by personal exposure monitoring and sampling at the process emission area show

airborne concentrations of detected compounds to be in the range of 0-3 mg/m3 under normal

operating conditions. Laser cutting or purging operations generate higher pollution levels in

particular formaldehyde which was found in some cases at a concentration above the

workplace exposure limit.

3

INTRODUCTION

Plastics form a major part of our living environment because of their application potential

and adaptability. They are the basic materials of many products in multiple areas such as

agriculture, healthcare, energy generation, aerospace, automotive, maritime, construction,

electronics, packaging or textile. In Europe, this activity is undertaken by 1.45 million people

working at more than 60,000 companies (1)

.

Main chemical exposure routes for employees in the workplace include dermal contact and

via the respiratory system by inhalation of dust and volatile compounds released by polymers

during thermal processing (2)

. The plastic material choice and the thermal process

implemented depend on the type of article, its size and the physical properties that are

required for its use. During production of plastic articles, polymers are subjected to

temperatures between 150 °C and 450 °C. These thermal processes can lead to emission of

toxic VOCs in plastic molding shops (3)

. In addition to exposure during normal production

operations, workers can be exposed to toxic compounds during operations that may cause

serious overheating of the material such as drilling, abrading, cutting or welding operations as

well as during process malfunctions and breakdowns. Plastics overheating can lead to the

release of thermal breakdown products in much higher concentrations or of different type than

during normal production.

In general, polymer heating causes release of free monomers and decomposition of

polymers or their additives; this has been reported in several studies (4)

.Our knowledge of

occupational exposure to toxic compounds during thermal processing of plastics is limited as

most of the studies are focused on a thermoplastics or pollutants family (5-7)

or conducted to

determine plastics thermal behaviour (8)

. Numerous studies have been carried out on

thermoplastic decomposition in an inert atmosphere (nitrogen or helium) (9)

in order to

determine thermoplastic composition or assess thermoplastic ageing pathways. However,

4

little is known of the degradation products released under real thermal processing conditions

and in the presence of air, which can lead to the formation of oxygenated decomposition by-

products such as aldehydes, ketones or alcohols (10)

. Besides, K. Yamashita et al. have

observed that during polymer degradation, smaller amounts of VOCs were emitted under

nitrogen atmosphere than under air atmosphere (11)

.

The difficulty with such an assessment resides in the wide variety of plastics, additives

and processes used. Worker exposure assessment requires knowledge of the type of thermal

degradation products, which depend on polymer type and temperature implementation.

Moreover, there is no recognized, standardized method of thermal degradation of plastics

for representatively determining the thermal degradation products, to which workers may be

exposed in plastic production shops.

The purpose of this study was therefore twofold. The first objective was to develop a

thermal degradation protocol representative of industrial thermal processing and applicable to

many thermoplastics. To achieve this goal, several degradation conditions were studied and

the defined protocol was validated by atmospheric measurements during industrial thermal

processing. The second objective was to define, from the developed protocol, thermal

decomposition products for the most commonly used polymers and temperatures in order to

develop a diagnostic tool for improving worker exposure assessment.

METHODS AND MATERIALS

We chose to study the most widely used thermoplastics. Table I shows the polymer

families studied as well as the corresponding polymer consumption in Europe and the

industrial processes used during the field survey. 15 of the 25 thermoplastics were studied

both in the laboratory and the factory. The polymers with no process type reported were only

studied in the laboratory.

Laboratory analysis

5

Conditions for thermoplastic thermal decomposition were chosen as close as possible to

those under which real industrial processes are performed.

Not all industrial processing conditions can be reproduced in the laboratory. In real life,

thermoplastics are heated in an endless screw under high pressure before injection in a mould

or extrusion threw the die of the extruder. Melt thermoplastics are maintained at high

temperature under air atmosphere for several minutes. Two methods were selected thanks to

their ability to subject polymers to heating conditions close to those encountered in many real

situations (12-14)

: thermogravimetric analysis (TGA) and pyrolysis under airflow. Both these

methods allow thermal breakdown in air, controlled temperature with slow or fast heating rate

and steps at constant temperature.

Thermogravimetric analyses were performed on a TA Instruments Q50 analyser under

90 ml/min airflow. A piece of thermoplastic (with a mass of several micrograms) was placed

on a platinum pan in the analyser oven. The heating rate could be varied from 5 °C/min to

50 °C/min. Weight loss was monitoring during heating so that the onset degradation

temperature (Ti), the temperature at maximum weight loss (Tmax) and the final weight loss

could be determined (15-17)

. Released compounds were sampled at ambient temperature using

convenient samplers connected to the outlet of the thermogravimetric analyser oven. Multi-

bed sorbent tubes (Carbotrap 300) were used for qualitative analysis purposes as described in

the National Institute for Occupational Safety and Health (NIOSH) method 2549 (18)

. Multi-

bed tubes were then analysed off-line by thermal desorption (Perkin Elmer ATD50). Inlet

split and outlet flow was respectively of 40 ml/min and 50 ml/min. The desorption flow was

of 10 ml/min. The multi-bed sorbent tubes was desorb at 320 °C during 20 min. The cold trap

was an air toxic trap, maintained at -30 °C during desorption and then heated at 330 °C during

injection in a GC/MS analyser (Perkin Elmer).

6

Pyrolysis-in-air analyses were performed using a CDS Analytical Pyroprobe 5200

connected on-line to a GC/MS analyser. During pyrolysis, released products were collected

on a cold trap filled with Tenax and maintained at -30 °C under liquid nitrogen flow. At the

end of decomposition, the trap was flushed with helium and heated very quickly to 280 °C to

inject the adsorbed compounds into the split/splitless injector of the GC maintained at 270 °C

with a split ratio of 100. This method does not allow quantitative analysis, but does allow

degradation involving several temperature steps. Table II summarises all the features of each

technology.

Thermal breakdown products were analysed using a semi-polar capillary column (vf-

624ms, 0.32mm*1.4µm*60m) with a column flow of 1 ml/min. The oven programming was

35 °C for 5 minutes, 250 °C for 15 minutes with a heating rate of 5 °C/min.

To determine influencing factors on released thermal breakdown products, three

parameters were studied: the sample preparation methods – manual cutting, ambient or cold

grinding – the heating rate during thermal degradation – 5, 10 20 and 50°C/min – and the

decomposition method – thermogravimetric analysis and pyrolysis. All these experiments

were performed under airflow of 90 mL/min. To develop the protocol, the influence of the

three parameters was evaluated comparing the thermal breakdown compounds obtained for

each experiment.

8 different samples of PP (named PP-a to PP-h), and 2 different samples of ABS (named

ABS-a and ABS-b) and PS (named PS-a and PS-b) were studied. They were supplied by

different manufacturers and they differ by their additives type or content and their rate or

nature of mineral loading.

For each reference of thermoplastics, laboratory experiments were replicated between 4

and 13 times.

Industrial processing

7

Data obtained in the laboratory were compared to those collected in 13 thermoplastic

processing shops selected to provide a wide range of polymers and processes.

Thus, 15 of the 25 polymers were studied in both the laboratory and the factory (Table I).

Various processes were encountered at the 13 industrial sites: 16 extrusion lines for

continuous production of tubing, cables and films on 4 sites; 50 injection-molding presses for

production of car components, connections and packing on 8 sites, hot wire film cutting on 2

sites, laser cutting on 2 sites, and automatic and manual welding on 3 sites.

Both sampling at the processes emission area and individual monitoring were performed.

In the first case, sampling devices were placed as close as possible to the emission location

(about 30 cm) to identify and quantify compounds released in air during the heating process.

Personal monitoring was carried out on operators during normal operating conditions, during

purging operations and laser cutting operations. For screening identification by NIOSH

method 2549 (16)

, the collection device used was a thermal desorption tube filled with three

adsorbent media that allowed collection of C3 to C30 molecules. For quantifying airborne

concentrations, the sampling and analytical methods listed in Table III were used.

Polymer pellet samples were simultaneously collected at the process feed side to compare

the atmospheric measurements with the results obtained by thermal decomposition in the

laboratory for the relevant polymer.

RESULTS AND DISCUSSION

The onset degradation temperature (Ti), the temperature at maximum weight loss (Tmax)

and the final weight loss were determined as illustrated in Figure 1.

Laboratory analysis – Parameter optimisation

This survey studied the influence of three parameters on thermal decomposition product

characteristics.

8

Thermogravimetric analysis-based thermal degradation was undertaken on three

thermoplastics (polyethylene (PE), polyacrylonitrile-butadiene-styrene (ABS) and

poly(styrene/acrylonitrile) (SAN)) subjected to an airflow of 90 ml/min and a heating rate of

20°C/min up to 450 °C in order to compare different sample preparation methods.

The main products proved to be the same whatever the preparation method (manual

cutting, ambient or cold grinding), specifically C1-C5 aldehydes for PE, styrene,

benzaldehyde, acrylonitrile and alpha-methylstyrene for ABS and styrene, benzaldehyde,

alpha-methylstyrene and acrylonitrile for SAN.

Among the preparation methods tested, cold grinding gave more homogenous samples than

ambient grinding and the former preparation method is safer for the analyst.

Heating rate influence was assessed by varying the temperature gradient between 5 and

50 °C/min. Polyethylene (PE) and polypropylene (PP) decomposition was performed by

thermogravimetric analysis under an airflow of 90 ml/min. The final temperature was 450 °C.

PE thermal breakdown at 10 °C/min, 20 °C/min and 50 °C/min caused emission of C1-C5

aldehydes, propane and methylvinylketone. At 5 °C/min, the detected products were C1-C3

aldehydes, C3-C4 alkanes, methylvinylketone and acetic acid.

In the case of PP, acetaldehyde, C3, C5, C8, C9 ketones and acetic acid were observed at

10 °C/min, 20 °C/min and 50 °C/min. No C5 ketone was detected at 5 °C/min. We therefore

conclude that the same compounds are obtained with a heating rate of 10, 20 and 50 °C/min.

Only the products obtained at 5 °C/min differ, probably due to the lower energy supply. The

20 °C/min heating rate offers the best compromise between the technical properties of the

analyser oven and a short-time analysis.

The third parameter evaluated was the thermal degradation method. Compounds obtained

from decomposing 21 thermoplastics by pyrolysis and thermogravimetric analysis in air were

compared. The thermoplastic samples were decomposed with a heating rate of 20 °C/min up

9

to 450 °C. The released compounds was analysed by GC/MS. Only the method of sampling

the released compounds differed in the two methods. Figure 2 shows typical results. The main

decomposition compounds are identical; only the response observed for each compound

differs according to the degradation method used.

The data obtained allowed us to select the best protocol. TGA and pyrolysis in air both

enable qualitative determination of thermal decomposition products, either by off-line

analysis of screening samplers or by on-line analysis. However, only thermogravimetric

analysis enables determination of the initial decomposition temperature and weight loss at a

determined temperature.

The selected protocol therefore takes advantage of each method:

- Cold grinding, thermal breakdown by thermogravimetric analysis in air up to

450 °C with a 20 °C/min heating rate: monitoring of the polymer weight loss with

temperature and quantitative analyses of toxics compounds such as aldehydes,

aromatic compounds and acrylates on the appropriate samplers

- Cold grinding, thermal breakdown by pyrolysis in air with a 20 °C/min heating

rate: qualitative determination of decomposition compounds released by the polymer

at the process temperature and at 450 °C.

Laboratory analysis – Thermal degradation products

The products of thermal decomposition in air and weight loss were determined using the

selected protocol for each of the 25 thermoplastics studied. The results obtained from

pyrolysis in air of all the polymers were classified by chemical compound family as illustrated

in Figure III. These results highlight the fact that the main decomposition products are free

monomers, monomers from polymer cracking, oxygenated compounds from polymer groups

and products of oxidative reactions (19)

. Observed degradation compounds can vary from a

polymer reference to another from the same family, depending on the additive type and

10

content. In parallel, the weight loss of the 25 polymers was monitored by TGA under the

chosen conditions (20 °C/min, 450 °C, in air, see Table IV). In the case of the 8 references of

PP, the results reveal that thermal behaviour of the same thermoplastic can vary depending on

the additives and the type and rate of mineral loading (Table IV). For example, PP-4 and PP-5

which are respectively filled with 30 and 70% of mineral compounds, only loss 60 and 30%

of their initial weight at 450°C, while the other references loss between 90 and 97% of their

initial weight. The weight loss corresponds to the degradation of the polypropylene contains

in PP-4 and PP-5, therefore these two references are less emissive thank to their higher

mineral loading.

Industry measurements – Protocol validation

527 atmospheric and 71 polymer samples were taken on 13 industrial sites.

Table V shows the results obtained. Thermal degradations were conducted in the

laboratory based on the selected protocol and at the same temperature as in the industrial

process to validate protocol representativeness. Results obtained by laboratory analysis and by

measurements taken in workplace situations are in fact in fairly close agreement.

Some plastics (e.g. polycarbonate (PC) and thermoplastic polyurethane (TPU)) did not

decompose at the temperature applied in the laboratory. These results are consistent with

thermogravimetric analyses (Table IV) showing that no degradation occurs below 390 °C and

250 °C for PC and TPU respectively. Thus, the air pollutants detected on the samplers must

have originated from adjacent processes or other unrelated activities. This could also have

been the case when compounds were detected on the industrial premises but not in the

laboratory.

Table VI shows concentration levels of hazardous substances emitted at various industrial

processes. Most of the time, concentration levels of toxic compounds released from the

processes under study were low. Most operators control temperature carefully to prevent

11

thermoplastic degradation and this practice has a positive impact on emission levels. The

formaldehyde and benzene emissions are detailed in Table VII, depending on the production

process and the temperature of plastic transformation. The results show that injection-molding

of POM can generate high formaldehyde emissions. Under normal operating conditions –at

temperatures between 200 and 215°C– formaldehyde levels varied from 0.32 to 0.44 mg/m3 at

about 30 cm of the heating zone of POM. During the purging step, when the polymer is

flushed out of the press at 215°C, the formaldehyde concentration was of 10 mg/m3 at 30 cm

of the injection-molding furnace. The purge of the press requires the intervention of an

operator. The personal samplings made during this step exhibit a formaldehyde concentration

of 1.12 mg/m3 during 15 minutes. Therefore, even at temperatures below the initial

decomposition temperature, which is 254°C for POM (Table III), this polymer releases

formaldehyde, most probably corresponding to non-polymerized monomer molecules.

Laser cutting of PMMA appears to be another source of formaldehyde emissions. In the

laser cutting cabin, concentrations of formaldehyde vary from 0.34 to 0.53 mg/m3. Laser

cutting of PMMA also generates high concentrations of methyl methacrylate, up to 37 mg/m3

at about 30 cm of the laser. The large amount of energy supplied by the laser causes the

thermal degradation of the PMMA to its monomer. The personal samplings carried out for 4

hours on the operator working at the laser cutting station measured formaldehyde and methyl

methacrylate concentrations of 0.04 and 0.5 mg/m3 respectively.

Benzene concentration was also measured in the visited plastic production shops. Levels

were mostly less than 0.1 mg/m3. Concentrations between 0.18 to 0.32 mg/m

3 were obtained

in the case of injection-molding of PET and POM and during PMMA transformation by laser

cutting.

Due to the high temperatures involved, this process has the greatest potential to generate

harmful substances. Therefore, to prevent workers exposure to thermal breakdown products

12

released by laser cutting, it would be necessary to equip the station of a localized collection

system to capture the different pollutants generated at the emission source.

As shown in Table VII, formaldehyde and benzene emissions were detected in the

laboratory study for the same thermoplastic/temperature combination except for very low

level and in the cases of PET and POM transformation. Further analyses had shown that the

emission source of benzene in these cases was the vacuum pump of the injection-molding

press and not directly the thermal processing of the thermoplastics.

CONCLUSION

The laboratory study of the influence of thermal breakdown parameters allowed us to

establish a decomposition and analytical protocol.

Comparison of emissions measured during real industrial processing and when applying

the laboratory protocol to the same thermoplastic samples provided similar results.

Thermal breakdown product characteristics and temperature behaviour were determined

for the 25 thermoplastics studied.

These combined data effectively contribute to prevention on two sides:

- Thermal degradation control by knowing the degradation onset temperature

- Workplace exposure assessment at thermoplastic production facilities: the

protocol is a tool available to prevention personnel for assessing the toxic compounds

likely to be released by thermal processing of plastics

REFERENCES

1. http://www.plasticseurope.org/documents/document/20150227150049-

final_plastics_the_facts_2014_2015_260215.pdf

13

2. Rosenberg, N.: Allergic occupational respiratory disorders in the plastics industry.

Occupational allergology-pneumology sheet TR 46. Doc. Méd. Trav. 118:235-251 (2009).

3. Lafon, D., and R. Garnier: (2008) - Toxicity of thermal degradation products of

plastics. Medical-surgical encyclopedia. Toxicology - Professional Pathology 16-541-C-10.

Issy-les-Moulineaux: Elsevier Masson, 2008.

4. Forrest, M.J., A.M. Jolly, S.R. Holding, and S.J. Richards: Emission from processing

thermoplastics. Ann Occup Hyg. 39:35-53 (1995).

5. Farhi, R., C. Morel, and J. Chéron: Plastics and additives- Health and safety. 3rd

edition. Paris: Édition INRS, ED 638, 2006.

6. Mairesse, M., J.M. Petit, J. Chéron, and M. Falcy: Thermal degradation products of

plastics. Hyg. Sécur. Trav. 174: 47-57 (1999).

7. Masson, A., and J.C. Protois: Thermal degradation of polymers. Hyg. Sécur. Trav.

197:31-38 (2004).

8. Price, D., G.J. Milnes, P.J. Tayler, J.H. Scrivens and T.G. Blease: Degradation of

polymers studied by mass spectrometry. Polym. Degrad. Stab. 25:307-323 (1989).

9. Unwin, J., M.R. Coldwell, C. Keen and J.J. Mcalinden: Airborne emissions of

carcinogens and respiratory processing of plastics. Ann. Occup. Hyg. 57:399-406 (2013).

10. Wampler, T.P.: Applied pyrolysis handbook. 2nd edition. CRC Press, Taylor & Francis

group, 2006.

11. Yamashita, K., N. Yamamoto, A. Misukoshi, M. Noguchi, Y. Ni, Y. Yanagisawa:

Compositions of volatile organic compounds emitted from melted virgin and waste plastics

pellets. J. Air & Waste Manage. Assoc.59:273-278 (2009)

14

12. Prati, S., S. Smith, and G. Chiavari: Characterisation of siccative oils, resins and

pigments in art works by thermochemolysis coupled to thermal desorption and pyrolysis GC

and GC-MS. Chromatographia 59(3-4):227-231 (2004).

13. Bower, N.W., and C.J.K. Blanchet: Analytical pyrolysis - Chromatography: something

old, something new. J. Chem. Educ. 87:467-469 (2010).

14. Bhaskar, T., J. Kaneko, A. Muto, Y. Sakata, E. Jakab, T. Matsui, and M.A. Uddin:

Pyrolysis studies of PP/PE/PS/PVC/HIPS-Br plastics mixed with PET and dehalogenation

(Br, Cl) of the liquid products. J. Anal Appl. Pyrol. 72:27-33 (2004).

15. Blazek, J.: “Study of the reaction kinetics of the thermal degradation of polymer.” PhD

diss, Institut National Polytechnique de Toulouse, University of Toulouse, 2005.

16. Conesa, J.A., A. Marcilla, R. Font, and J.A. Caballero: Thermogravimetric studies on

the thermal decomposition of polyethylene. J. Anal Appl. Pyrol. 36: 1-15 (1996).

17. Cervantes-Uc, J.M., J.V. Cauich-Rodriguez, W.A. Herrera-Kao, H. Vázquez-

Torres, and A. Marcos-Fernández: Thermal degradation behavior of polymethacrylates

containing amine side groups. Polym. Degrad. Stabil. 93:1891-1900 (2008).

18. Method 2549 of Niosh Manual of Analytical methods, Fourth Edition, 5/16/96.

19. Cervantes-Uc, J.M., J.V. Cauich-Rodriguez, H. Vázquez-Torres, and A. Licea-

Claverie: TGA/FTIR study on thermal degradation of polymethacrylates containing

carboxylic groups. Polym. Degrad. Stabil. 91:3312-3321 (2006).

15

0

20

40

60

80

100

0 100 200 300 400 500

we

ight d

erivative as a fu

nctio

n o

f the

tem

pe

rature

Sam

ple

we

igh

t(%

)

Temperature (°C)

Ti

Tmax1

Tmax2

We

ight lo

ss

Figure 1. Thermogram interpretation. Ti: onset degradation temperature; Tmax: temperature

of maximum weight loss

16

0 20 40 60Time (min)

Figure 2. Chromatograms of polypropylene thermal breakdown products obtained by TGA

(−) and pyrolysis (−)

17

PE

SANPS

PV

DC

PV

C

PP SB

AB

S

PM

MA

PA

N

PO

M

PA

6

PA

12

PA

6.6 PC

PET

PB

T

PTT

CA

B

PV

DF

PTF

E

TPU

NA

S

Mineral acids

Hydrocarbons

Oxygenated compounds

Aromaticcompounds

Monomers

Aldehydes

Figure 3. Main thermal degradation compounds at 450 °C with respect to thermoplastic type

( : 1 compound, : 2 compounds, : 3 compounds, : 4 compounds, : 5 compounds)

18

Table I. Polymer, name, abbreviation, % European consumption (*) and processes

implemented during atmospheric measurements at plastic transformation facilities (polymers

with no process reported were only studied in the laboratory)

Polymer Name Abbreviation Cons. % * Process

Polyolefins

Polyethylene

PE 29%

Extrusion

Injection-moulding

Welding

Hot wire cutting

Polypropylene PP 19%

Extrusion

Injection-moulding

Welding

Polyvinylic

Polyvinylchloride PVC 11%

Extrusion

Injection-moulding

Welding

Poly-vinylidene-chloride PVDC 1% -

Polystyrenic

Polystyrene PS 7.5% Injection-moulding

Poly(styrene/acrylonitrile) SAN

2%

-

Poly(styrene/Butadiene) SB -

Polyacrylonitrile-butadiene-styrene ABS Injection-moulding

Poly(styrene/ethylene/butylene) SEBS Injection-moulding

Polyacrylic Polymethylmethacrylate PMMA

1%

Injection-moulding

Laser-cutting

Polyacrylonitrile PAN -

Polyacetal Polyoxymethylene POM 0.5% Injection-moulding

Polyamides

Polyamide 6 PA 6

3%

Extrusion

Polyamide 12 PA 12 -

Polyamide 6.6 PA 6.6 Injection-moulding

Extrusion

Polycarbonates PC 1.3% Injection-moulding

Extrusion

Linear polyesters

Polyethylene terephthalate PET

7%

Injection-moulding

Extrusion

Hot wire cutting

Polybuthylene terephthalate PBT Injection-moulding

Polytrimethylene terephthalate PTT -

Cellulosiques Acetobutyrate cellulose CAB <0.5% -

Polyfluoroethene

Polyvinylidene fluoride PVDF

<0.5%

-

Polytetrafluoroethylene PTFE -

Ethylene-Tetrafluoroethylene ETFE Extrusion

Thermoplastic polyurethane TPU 7% Parboiling

Other Styrene acrylic copolymer NAS <0.5% -

19

Table II. Summary of thermogravimetric analysis and pyrolysis characteristics

TGA Pyrolysis

Atmosphere Oxidative or neutral Oxidative or neutral

Heating rate 5 to 50 °C/min Up to 1000 °C/sec

Analytical coupling Delayed GC/MS analysis Online thermal desorption of the

cold trap and GC/MS analysis

Sampling mode of

released compounds

At the outlet on sampling media

at ambient temperature

On integrated cold trap of

Tenax at -30 °C

Weight loss yes no

Several temperature steps no yes

20

Table III. Air sampling and analysis methods

Compound Sampling media Sampling flow

rate (L/min)

Sampling

time Analytical method References

Formaldehyde Cartridge with

silica gel coated

with 2,4-DNPH

0.25 15min to

8h HPLC/UV

Metropol

method 001 Acetaldehyde

methyl

methacrylate Solid sorbent tube

of AC 0.05

15 min to

8 h GC/MS

Metropol

method 021 Methyl acrylate

Tetrahydrofuran Solid sorbent tube

of AC 0.05

15 min to

8 h GC/MS

Metropol

method 064

Benzene Solid sorbent tube

of AC 0.05

15 min to

8 h GC/MS

Metropol

method 012 Toluene

Styrene

1,3-butadiene Solid sorbent tube

of AC 0.05

15 min to

8 h GC/MS

Metropol

method 076

Vinyl chloride

monomer

2 tandem tubes of

AC 0.05

20 min to

5 h GC/MS

NIOSH

method 1007

Caprolactam

Solid sorbent tube

of XAD-2 resin

(100 mg/50 mg)

0.2 3 to 8 h GC/MS Metropol

method 072

Chlorhydric acid

Quartz fiber filters

impregnated with a

Na2CO3 solution

2 2 to 8 h Ionic

chromatography

Metropol

method 009

DNPH: dinitrophenylhydrazine

AC: activated coconut charcoal (100 mg/50 mg)

HPLC/UV: High performance liquid chromatography with ultraviolet detection

GC/MS: Gas chromatography with mass spectrometry

21

Table IV: Initial decomposition temperature (Ti), temperature of maximal weight loss

(Tmax), mean and variance of final weight loss, number of replicates (number of experiments

carried out for each reference) for the 23 thermoplastics studied

Polymers Ti (°C) Tmax 1 (°C) Tmax 2 (°C) Final weight loss (%)

Replicas mean Variance

PE 274 391 435 90 1.81 11

PP-1 258 363 NA 92 3.19 4

PP-2 262 374 NA 93 0.40 5

PP-3 248 291 NA 95 4.25 4

PP-4 201 290 420 59 4.92 6

PP-5 290 375 NA 30 7.89 6

PP-6 260 368 438 92 10.16 5

PP-7 248 290 NA 97 23.19 4

PP-8 247 286 NA 97 8.19 4

PVC 260 317 446 63 5.51 7

PVDC 210 278 NA 73 19.76 5

PS-1 257 341 NA 96 8.64 5

PS-2 239 382 427 95 13.88 9

PS-3 244 334 NA 96 21.00 6

SAN 258 392 NA 95 3.88 13

SB 308 439 NA 82 NA 4

ABS-1 285 424 NA 90 7.56 10

ABS-2 271 422 NA 77 8.50 4

PMMA 286 351 NA 96 15.93 13

PAN 299 343 NA 27 NA 2

POM 254 294 NA 99 6.36 8

PA 6 318 440 NA 78 NA 2

PA 6,6 304 444 NA 70 NA 2

PA 12 238 446 NA 73 NA 2

PC 391 NA NA 22 NA 3

PET 313 442 NA 83 17.69 4

PBT 302 416 NA 94 NA 2

PTT 301 416 447 85 NA 2

CAB 144 380 NA 94 NA 2

PVDF 408 NA NA 38 10.69 4

PTFE NA NA NA 0 NA 4

TPU 254 283 348 83 NA 2

NAS 274 391 435 90 1.81 11

22

Table V. Thermal decomposition products identified from atmospheric samples taken on

industrial sites and laboratory-based degradations

Polymer On industrial processes In laboratory

Temperature of the industrial process Laboratory degradation temperature

Main product only find on industrial site Main product only find in laboratory

Minor product only f ind on industrial site Minor product only f ind in laboratory

PE 170-200°C 180-200°C

C3 aldehyde

Alcohols (ethanol, ethoxypropanol)

Furane

PP 200-285°C 200-285°C

Benzoic acid Propanoic acid

Tetrahydrofurane Dimethylfurane

PVC 180-215°C et 380°C 180-210°C et 450°C

Ketones (methylethylcetone, methylvinylcetone) hydrogen chloride

Alcohols ( ethanol, ethyhexanol) Aromatic hydrocarbons (benzene, toluene, styrene)

C6-C13 saturated hydrocarbons

PS 210-230°C 230°C

C3-C16 saturated hydrocarbons Acrylaldehyde

Acetophenone

ABS 230-305°C 230-305°C

Isocyanonaphtalene

Acrylaldehyde, benzaldehyde

Phenol, cresol


SEBS 240-340°C 240-340°C

alcohols (methanol, ethanol) Toluene

Propanoic acid and butanoic acid

C3-C9 ketones

PMMA 240-260°C 240-260°C

Benzaldehyde Acrylaldehyde

Styrene C8 and C16 alcohols

C3-C5 saturated hydrocarbons C6-C12 insaturated hydrocarbons

Methyl esters

Methacrylonitrile, isobutyronitrile

POM 190-215°C 190-215°C

Methylal Isopropanol

Phenylmethanol Ketones (acetone, cyclohexanone)

Methyl esters

Aldehydes (formaldehyde, acetaldehyde)

alcohols (methanol, ethanol)

C3-C5 insaturated hydrocarbons


Alcohols (methanol, ethanol)

Acetic acid

Hydrocarbures aromatiques (styrene, benzene)

Aldehydes (formaldehyde, acetaldehyde, benzaldehyde)


acetic acid

Methyl methacrylate, acrylate de methyle

Aromatic hydrocarbons (styrene, benzene, toluene, methylstyrene)


alcohols (methanol, ethanol, propanol)


Brominated compounds


Methyl methacrylate

Acetic acid


Aromatic hydrocarbons (styrene, ethylbenzene)

Aldehydes (acetaldehyde, benzaldehyde)


Aldehydes (formaldehyde, acetaldehyde, C3-C7)

Ketones (acetone, butanone)

Alcohols (methanol, ethanol, butanol)

Acetic acid

Furane

Identified thermal degradation products

Main common product

Minor common product


Acetic acid


23

Polymer On industrial processes In laboratory

Temperature of the industrial process Laboratory degradation temperature

Main product only find on industrial site Main product only find in laboratory

Minor product only f ind on industrial site Minor product only f ind in laboratory

PA 6 270°C 270°C

C4-C9 ketones

Methacrylaldehyde

Ethanol

Acetic acid


PA 6.6 250-350°C 280-340°C

Acetaldehyde Aldehydes (formaldehyde, acrylaldehyde, C11-C14)

Insaturated hydrocarbons (C4-C19) Cyclohexanone

N-butylbenzensulfonamide Acetic acid

Butyle acetate Acrylonitrile

Dihydrofurane

PC 250-320°C 250-320°C

alcohols (ethanol, phenol, methoxypropanol) Chlorinated compounds (traces)

Methyl methacrylate

Aromatic hydrocarbons (toluene, methylstyrene)

Benzofurane, hydroxyacetophenone


PET 190-285°C 285°C

Benzaldehyde

Propanol, benzyl alcohol Butanol

Aromatic hydrocarbons (benzene, toluene, Acetic acid

monoterpenes ) Ketones (acetone, cyclohexanone)

C3-C10 saturated hydrocarbons C3-C4 insaturated hydrocarbons

PC/PBT 215-260°C 260°C

C3-C16 saturated hydrocarbons Aldehydes (acetaldehyde, acrylaldehyde)

Ethyl acetate, butyl acetate

ETFE 320°C 320°C

Hydrogen fluoride

TPU 160°C (TPU solution) 160°C (TPU alone)

Acetaldehyde Acetic acid (traces)

Alcohols (methanol, ethanol) Tetrahydrofurane (traces)

Limonene


Siloxanes

Tridecafluorohexane

Saturated and insaturated f luorinated hydrocarbons


Acetaldehyde

C3-C4 ketones

Furane, tetrahydrofurane

Alcohols (methanol, ethanol)

Methyl methacrylate


Aromatic hydrocarbons (benzene, toluene)

Caprolactame

Acetaldehyde

Caprolactame

Alcohols (methanol, ethanol, propanol, butanol)

Ketones (acetone, cyclopentanone)

Identified thermal degradation products

Main common product

Minor common product

Products in green text were found in both the field and laboratory studies

24

Table VI. Concentration levels of hazardous substances emitted at various industrial

processes.

Compound

Airborne concentration of detected compounds (mg/m3)a WEL

(mg/m3)

8h/15min PE PP PVC PA 6.6 ABS PC PET POM PS PMMA

Formaldehyde

0.01 to

0.33 (20)

<0.01

to 0.19

(18)

0.01 to

0.23

(11)

<0.01

to 0.10

(9)

0.01 to

0.02

(8)

0.01 to

0.03

(3)

0.01 to

1.98 (18)

0.10 to

10.30

(11)

0.02

(2)

<0.01

to 0.53

(8)

0.625/1.25

Acetaldehyde

0.03 to

0.35

(20)

0.01 to

0.20

(18)

<0.01

to 0.08

(11)

0.01 to

0.03

(9)

<0.01

(8)

<0.01

(3)

0.09 to

1.94

(18)

0.01 to

0.57

(11)

0.009

(2)

0.02 to

0.04

(8)

180/-

Acrylonitrile / / / / n.d.

(9) / / / / /

Methyl

methacrylate / /

0.01 to

0.92

(6)

n.d. to

0.03

(8)

0.01 to

0.29

(3)

<0.03

(2) / /

0.01

(2)

1.24 to

37.03

(10)

205/410

Methyl acrylate / / n.d.

(6)

n.d.

(8)

n.d.

(3)

n.d.

(2) / /

n.d.

(2)

n.d to

0.52

(10)

18/36

Tetrahydrofuran /

0.05 to

0.13

(14)

0.03 to

0.28

(6)

/ /

0.13 to

0.23

(6)

0.02

(6) / / 150/300

Vinyl chloride

monomer / /

<0.03

(6) / / / / / / / 2.59/-

Benzene

n.d. to

0.02

(14)

<0.10

(20)

<0.01

to 0.03

(12)

<0.01

to 0.01

(12)

<0.04

(9)

<0.10

(3)

<0.01

to 0.19

(16)

n.d. to

0.32

(10)

<0.01

to 0.01

(2)

n.d to

0.26

(8)

3.25/-

Toluene

n.d. to

0.26

(14)

n.d. to

<0.02

(20)

n.d. to

0.03

(12)

n.d. to

<0.01

(12)

<0.01

to 0.01

(9)

n.d. to

<0.01

(3)

0.06 to

2.26

(16)

n.d. to

0.01

(10)

n.d to

0.01

(2)

0.06 to

0.20

(8)

76.8/384

Styrene / / / /

0.03 to

16.25

(9)

/ / / 0.02

(2)

0.02 to

3.01

(8)

215/-

1,3-butadiene / / / / <0.10

(9) / / / /

Caprolactame / / / 0.1

(5) / / / / / 10/40

Chlorhydric acid / /

n.d to

0.02

(6)

/ / / / / / -/7.5

n.d: not detected

WEL: Workplace Exposure Limit aValues in parentheses indicate number of air samples collected for each chemical for each

plastic

25

Table VII. Concentration levels of formaldehyde and benzene emitted as a function of the

plastic type and the production process

Plastics Process

Processing

temperature

(°C)

Mass

processing

rate (kg/h)

Airborne (mg/m3)a Found in laboratory

Formaldehyde Benzene Formaldehyde Benzene

PE

Extrusion 200-205 / 0.01 (5) n.d. (5)

Yes No Injection-moulding 200 20-150 0.01 to 0.02 (6) /

Welding 240 / 0.01 (1) <0.03 (1)

Hot wire cutting 170-190 / 0.03 to 0.33 (8) 0.01 to 0.02 (8)

PP Extrusion 220-260 60 0.01 to 0.19 (8) n.d. to <0.10 (8)

Yes No Injection-moulding 220-285 130 <0.01 to 0.07 (10) <0.01 to <0.20 (12)

PVC

Extrusion 190-215 280 0.01 to 0.23 (9) <0.01 to 0.03 (9)

Yes Yes Injection-moulding 180-200 / 0.01 (1) <0.01 (2)

Welding 380 / 0.01 (1) <0.01 (1)

PA 6.6 Injection-moulding 290-350 12-22 <0.01 to 0.10 (7) <0.01 to 0.01 (10)

Yes Yes Extrusion 280 20 0.02 (2) <0.1 (2)

ABS Injection-moulding 260-305 33 0.01 to 0.02 (8) <0.01 to <0.04 (9) Yes Yes

PC Injection-moulding 290-320 24 0.01 (2) <0.01 (2)

No No Extrusion 250 20 0.03 (1) <0.10 (1)

PET Injection-moulding 230-285 160 0.01 to 1.49 (12) <0.01 to 0.18 (10)

Yes No Extrusion 275-290 158 0.02 to 1.98 (6) <0.02 to 0.19 (6)

POM Injection-moulding 200-215 / 0.10 to 10.30 (11) n.d. to 0.32 (10) Yes No

PS Injection-moulding 230 / 0.02 (2) < 0.01 to 0.01 (2) No No

PMMA Injection-moulding 230-260 15-23 <0.01 to 0.04 (4) n.d. to <0.02 (4)

Yes Yes Laser-cutting unknown / 0.34 to 0.53 (4) 0.04 to 0.26 (4)

n.d.: not detected aValues in parentheses indicate number of air samples collected for each chemical for each

plastic

Documents

Identifying thermal breakdown products of thermoplastics