Document JNB0wdzQ6kV6m6jeJRdgdwd56

REPORT ON ENCAPSULATION OF LEAD CHROMATE PIGMENT IN PLASTIC RESINS RELATIVE TO THE WORKPLACE HAZARDOUS MATERIALS INFORMATION SYSTEM (WHMIS) CLASSIFICATION PRESENTED BY G.P. ROBSON ENVIRONMENTAL SCIENCE DUPONT CANADA INC. sr BOR 000404 report on encapsulation of lead chromate pigment IN PLASTIC RESINS RELATIVE TO WHMIS CLASSIFICATION INTRODUCTION The attached report was presented to the SPI Canada Ad Hoc Group of Manufacturers of Concentrates and Compounds on 1990 February 20. The information contained in the report was developed at the request of the Ad Hoc Group in an attempt to shed light on the WHMIS sums of 'encapsulated* hazardous ingredients in plastic resin concentrates or compounds, using lead chromate pigment as a model. This was needed to decide whether workers could be exposed to lead pigment during anticipated use of lead chromate-pigmented plastic resins. The Ad Hoc Group arranged for the manufacture of the pigmented resins, the supply of the regrind materials and dau from acid extraction tests. My responsibility has been to arrange for electron microscopy and photon spectroscopy and to interpret the results. At the February 20 meeting, we all recognized that although the dau developed provide an incomplete picture of the overall situation, and there is no will to continue the investigation at this time, they may represent an important change in our understanding of the availability of additives in plastics resins versus the position we have adopted to date. The Group agreed that the report should be shared with SPI (U.S.) in order to inform them and to solicit their comment and feedback on any implications versus the OSHA Hazard Communication Standard. BOR 000405 1990 March 01 ENCAPSULATION OF LEAD CHROMATE PIGMENT IN PLASTIC RESINS AND ITS IMPACT ON WHMIS CLASSIFICATION Summary WHMIS regulations in Canada require that untested mixtures be classified according to the toxic properties of ingredients in the mixture. The Plastics Industry considers that it is unreasonable to apply this concept to plastic resin concentrates and compounds which contain toxic ingredients, without regard for the availability of such ingredients to cause worker exposure. An SPI Canada Ad Hoc Committee of manufacturers of compounds and concentrates proposed that data be developed to provide a better understanding of how toxic ingredients are distributed in plastic resins and their availability to be released and cause worker exposure under anticipated use conditions. Part of that activity involved the preparation and testing of lead chromate pigmented PVC, ABS and PE resins as 3 mm dia. cylindrical pellets. Lead chromate was selected as a 'model* substance since it is commonly used and is a WHMIS Controlled Product as a carcinogen. Resin pellets were examined by Scanning Electron Microscopy (SEM) and X-Ray Photon Spectroscopy (XPS) to determine whether lead chromate pigment particles were exposed at thepellet surface or were truly encapsulated in the resin. SEM showed that pigment particles were at or close to the surface but due to the slight surface penetration of the electron beam - up to l micron - it was not known whether they were actually exposed at the surface. XPS provides quantitative information on elements at surfaces and only penetrates monolayers at the surface approx. 50A. 1 micron is equivalent to 10.000A. XPS showed that the lead pigment concentration at PVC pellet surfaces was similar to the bulk concentration; PE pellets showed traces of lead pigment at the surface. In ABS pellets no lead was detected at the surface since silicate-encapsulated lead chromate pigment was used. Similar XPS results were obtained from frac tured pellet surfaces which revealed the interior or bulk composition. Extraction of whole pellets in 5% aqueous hydrochloric acid for 10 minutes, nuing i standard to stomach acid digestion, showed rhar some lead was extraaed in all cases, in approximate proportion to the lead pigment con centration in theresin. The average proportion of lead extraaed from whole pellets was 0DB% of the total lead in the resin but the actual levels of lead extraaed were below 0.03%. The effea of resin size reduction, eg. as dust, on lead by arid, is not known. Assuming that dust inhalation is the principal workplace exposure route of concern for plastic resins, tests were conducted to measure the inspirable dust concentration (i.e. <100 micron particle size) in commercial reground resin. Samples of ABS, PVC and HDPE commercial regrind resin (not lead chromate pigmented) contained a maximum of 0.002% dust less than ISO microns. Testing of one grade of S mm dia. PE resin pellets by rolling in a steel can for one hour to simulate attrition in steel containers or pneumatic conveying lines produced 0.0012% dust less than 177 microns. It is not known how a broad range of plastic resinswould rate in evaluating their potential to produce inspirable dust. BOR 000406 Several possible ways of using the 0.1% (or 1.0% as appropriate) minimum concen tration for WHMIS classification of mixtures are suggested which could apply to plasticresins. Some discussion will be required to decide on their value and applicability. Based on these data we should reconsider the use of theterm 'encapsulated' for additives in plastic resins. It is clear that some lead chromate pigment particles are actually exposed at the surface and are not covered by a layer of plastic resin. I suggest the term 'bound' would be more accurate. INTRODUCTION The regulation relating to WHMIS Classification and the use of the arbitrary 0.1% or 1.0% by weight ingredients minimum concentrations for classification of untested mixtures for toxicity and corrosivity, involve the inherent assumption that such ingredients are available in a form such that workers can be exposed to them.Also these cut-off concentrations recognize that ingredients present below these concentration values will not be consideredin classifying untested mixtures because they will not contribute significantly to the hazards of the mixture. The concept of hazardous ingredients present in materials to which workers cannot be exposed under anticipated use conditions is the basis for the exclusion of manufactured articles from WHMIS classification. This concept has not beenextended so far to cover materials other than manufactured articles. Commercial thermoplastic resins such as polyethylene, polypropylene, polyvinyl chloride, polystyrene, acrylonitrile/butadiene/styrene copolymers (ABS), and acetals, are tough materials which do not shatter easily under impact or grinding conditions to form fine powders, and which are essentially insoluble in water and body fluids. When such resins contain WHMIS Controlled Product ingredients such as non volatile pigments, fillers, stabilizers or reinforcing agents (fibres), the ingredients are dispersed in the polymer matrix by melt compounding and when resin compound pellets are produced for commercial use, the ingredients are bound firmly within the pellet and are not available in any significant amount to causeexposure ofworkers to the ingredients under normal use conditions. In order to demonstrate experimentally that non-volatile additives arebound in plastic ream matrices, we must recognize the three human exposure routes for plastics resin ingredients, viz: (i) inhalation, (ii) ingestion, and (iii) eye or skin contaet/abaorption. Note; Re Scone of Bound Ingredients Ingredients which arevolatile under melt processing conditions to release gases, vapours or aerosols (i.e. small liquid or solid particulates which condense in air from vapour released) are not considered as being bound in the plastic resin. Such ingredients will be subject to the normal WHMIS classif ication rules for untested mixtures, eg. residual styrene in styrene-based polymers. BOR 000407 3- - The following criteria are proposed for additives which can be considered to be bound in plastic resins: 1. Additive has not significant vapour pressure at worst case melt processing temperatures (what is significant?). 2. Additive has decomposition temperature well above melt processing temper atures. 3. Additive can bewell dispersed in the resin, i.e. no large agglomerates exist. 4. Additive does not migrate or diffuse to the surface of resin pellets on storage. 5. Grinding of scrap or reclaim does not liberate the free additive or expose the additive in a way that can cause ham to workers. Note: In most cases all five conditions should be met Exceptions may occur, eg.azodicarbonamide blowing agent is a WHMIS controlled product ingredient used in some resins, but is intended to decompose to release nitrogen gas under normal melt processing conditions. Skin or eye contact is not considered to be a concern provided that the additive does not migrate significantly to the surface of the resin pellet. For inhalation and ingestion we are concerned primarily with exposure to dust, since pellets will not be ingested or inhaled. Dust can be generated in several ways: i) during resin compounding by manufacturer fines from attritionof pellets during handling, pneumatic conveying and packaging; and dust from grinding of product or grinding of recycle or reject material ii) during use by customer; fines from handling and pneumatic conveying; and dust from grinding recycle material EXPERIMENTAL WORK Experiments were carried out: 1. Commercial ream pellets containing lead chromate pigment were examined by electron microscopy, x-ray photon spectroscopy and acid extraction to determine the location and availability of the lead pigment at the pellet surface. 2. Commercial resin pellets containing azodicarbonamide were to deter mine the potential far dust production by attritionand release of the free azodicarbonamide ingredient 3. Commercial resin regrind samples were examined to determine the amount of fine dust produced. BOR 000408 RESULTS 1. PVC. ABS AND PE RESINS CONTAINING LEAD CHROMATE PIGMENT a) Scanning Electron Microscopy of Surface of 3 mm_ dla. x 3 mm long Pellets Scanning electron microscopy (SEM) of resin pellets was carried out to determine if the lead chromate pigment could be detectedat, or near, the surfaces. Some pellets were fractured in liquid nitrogen and the fractured surfaces examined to determine if the interior was significantly different from the external surface. Resins containing 33 to 30%pigment (concen trates) and also 2% pigment (compounds) were examined. Since different sources of concentrate were used to prepare the pellets it was found that ABS resin contained silica-encapsulated lead chromate pigment. PE and PVC resins contained non-encapsulated lead chromate pigment. The detailed results and copies of the photo-micrographs obtained on the SEM at Queen's University are in Appendix 1. Note that SEM is not a quanti tative technique. In summary the following conclusions were made: i) Lead pigment particles were detected at or close to the surface in all cases. For PE and ABS the pigment concentration did not appear signifi cantly different between the fractured interior and exterior samples. For PVC there appeared to be a greater pigment concentration at the exterior surface compared to the fractured, internal surface. ii) SEM showed pigment particles at or close to the pellet surface but, because the electron beam penetrates the resin surface to a depth of up to 1 micron, it is not possible to know whether any pigment particles are actually exposed or whether they are covered by a thin (approx. 1 micron) layer of plastic resin. b) X-Ray Photon Spectroscopy of Snrihce of 3 mm dla. x 3 mm long Pellets X-Ray Photon Spectroscopy (XPS) of resin pellets was carried out at the Surface Science Laboratory of the University of Western Ontario (Surface Science Western). XPS was used because it can examine the outermost surface of materials with penetration only of approx. 5QA (Note: 1 micron equals 10.00QA) and provide quantitative data. The results of the XPS work are in Appendix II. In summary, the following conclusions were made (all concentrations are weight %): i) In the PVC samples high concentrations of lead were detected at the external surface vis 15% in Pb in the 33% lead chromate sample and 3.6% Pb in the 2% lead chromate samples. These levels were reduced by 90% after the measured surface was sputtered, Le. cleaned to a depth of 10QA by the x-ray dux. This indicates that higher levels of pigment are at the external pellet surface. BOR 000409 5- - ii) In the PE samples the 50% PbCr04 pellets showed approx. 0.5 to 1.0% Pb at the surface. The 2% PbCr04 pellets showed approx. 0.2% Pb at the surface at the detection limit of the method. iii) In the 2% PbCr04 in ABS sample, lead pigment was not detected in any surface, at a detection limit of approx. 0.2% Pb (see c) ii) below). There are several inconsistencies in the numerical values for atomic composi tion provided by XPS. For example, lead concentrations decrease after sputtering in the PVC samples but increase greatly in the PE sample. It is difficult to explain such effects based on the few samples examined. c) X-Ray Photon Spectroscopy of Surface of Lead Chromate Pigment Used In These Tests i) A non-encapsulated grade of pigment was used for the PE and PVC samples containing approx. 60% Pb. XPS of this pigment showed the surface composition to be: 17.2% by weight Pb 1.9% by weight Cr 13.8% by weight A1 36.2% by weight C 30.9% by weight 0 ii) A silica-encapsulated grade of lead chromate pigment was used in the AES sample and was also examined by XPS. The surface composition found was: 27.6% by weight 0 35.1% by weight C 23.9% by weight Si 13.4% by weight A1 Lead was not detected. Sputtering of the surface to remove the original surface and re-examina tion did not detea the presence of lead. Therefore based on this limited test, this pigment is clearlycoated with a material which appears to encapsulate the lead chromate pigment entirely. d) Simulated Stomach-Add Extraction of Lend Chromate-Contalnlne Resina and Plmnents Samples were extracted with 5% aqueous hydrochloric add at 20*C for ten minutes according to the method described by Consumer and Corporate Affairs Canada in the HazardousProducts Act, Part 1, Section 9. Work was done by Technitrol Canada Ltd, Report No. 194363, 1969 August 25. Results were: BOR 000410 6 Samcls % Pb in Resin ppm wt. of Pb* Leached Proportion of Total Lead in Sample Extracted bv Arid as % 33% PbCi04 in PVC 2% PbCr04 in PVC 50% PbCi04 in PE 2% PbCr04 in PE 3% PbCr04 in ABS 2% PbCr04 in ABS Regular PbCr04 Pigment Encapsulated PbCi04 Pigment 19.8 202 1.2 8 30.0 249 1.2 4 1.8 27 1.2 7 60 9400 40 6900 0.10 0.07 0.08 0.03 0.15 0.06 Average of above values - 0.08 1.57 1.73 * ppm Pb is calculated on weight of sample pellets/pigment 2. COMMERCIAL POLYETHYLENE RESIN CONTAINING 0-5 TO 1.0% A2QDICARBONAMIDE a) Dust Production bv Attrition of Resin Pclleta Resin in the form of approx. 5 mm dia. lens-shaped pellets, produced by melt cutting, was tested to simulate attrition caused by pellets impacting other pellets or impacting metal surfaces such as in bulk containers and pneumatic conveying lines. A one-gallon steel paint can was half-filled with resin and rolled at 60 rpm for one hour. The resin was then screened to measure dust production with results: % weight Through 18 mesh and Retained on 80 mesh (177 micron opening) % weight Through 80 mesh 0.0096 0-0012 b) CouwooMou of Fine Dust Produced The less than 80 mesh were cammed by infrared microscopy. In this method infrared spectra are produced for individual particles. The spectra showed the material to be principally polyethylene with small levels of fTiyfinAffumiH* present. Thus the izodiearhonamide (which is a respiratory sensitizer) is bound in the polyethylene resin. Free particles of azodicarbonamide were not detected. BOR 000411 -7. 3. DUST tN REGRIND SAMPLES OF ABS. PVf AND HPPE RESIN Samples of resin from a commercial regrind operation were collected directly from the grinder discharge and the size range measured at Queen's University. Results were: % weight retained oo sieve Resin Mesh 5 Mesh Opening l4 mm) ABS PVC HDPE 9.6 48.5 41.7 18 Mesh 100 Mesh 1L_siblI fl50 micron) 89.7 0.65 51.2 03 57.6 0.68 (75 micron) 0.0013 Nil 0.002 Less than 200 Mesh Nil Nil Nil Airborne particles which can be inhaled must be less than 100 micron in size. This is known as `inspirable* dust (see ACGIH TLV Booklet, App. E). The fraction between 100 miaon and 10 micron in size will be deposited in the upper respiratory tract and will usually be swallowed with the mucus. Particles less than 10 micron in size will be deposited in the midrespiratory tract and lungs. The data show that only 13 to 20 ppm by weight of regrind resin is less than 150 micron in size. No detectable quantities of dust less than 75 micron could be measured. PtSCUSStQM 1. CORRELATION BETWEEN XPS/SEM AND ACID EXTRACTION TESTS The acid extraction test simulates the extraction of ingredients from samples that could occur if the samples were ingested and subjected to stomach add. The data in the Table above show clearly that the amount of lead extracted from resin pellets is proportional to the lead content and also partide size. The surface area to volume ratios for pellets and pigment particles are: 3 mm dia. x 3 nun long peUet volume * 2L2 mar surface area - 42.4 mm2 Therefore surface area: volume ratio * Z1 mm*1 10 miaon dia. pigment agglomerate volume 0.52 x 10^ mnr surface area 3.14 x 10*4 mm2 Therefore surface area: volume ratio 6 x 102 Note: Selection of 10 micron size for the agglomerate is a simplifying assumption. -1 Therefore pigment particles have a surface area to volume ratio approx. 300 times greater than the pellets. BOR 000412 8 The greater surface area of pigment particles should enhance their solu bility in acid, however, it is not possible to make a direct comparison with pellets due to unknowns such as solubility rate and presence of some aluminum-containing coating on the pigment particles, as indicated by the XPS analysis see page 5. About 0.08% of the total lead present in pellets was extracted compared to approx. 1.6% for the lead chromate pigment, i.e. a ratio of 20:1 for pigment: pellets. Although the XPS measurements showed no detectable Pb at the surface of the encapsulated pigment and very low to non-detectable Pb levels at the surface of the PE and ABS samples, lead was still extracted from these materials in the hydrochloric add tesL Therefore there is no good correla tion between XPS results and HQ extraction test results with respect to the availability of the lead pigment. 2. ARGUMENTS AROUND THE 0.1% INGREDIENT CUT OFF CONCENTRATION PbCr04 present at <0.1% weight in a resin would not be subject to WHMIS regulations. 0.1% PbCr04 is equivalent to 0.06% Pb, assuming the material to be PbCr04. Is it possible to demonstrate that the lead in these materials which is in an available form, is less than 0.06% (600 ppm weight) Pb? I) Composition of Dust from Resin The SEM micrographs show that the lead pigment is at or very dose to the pellet surface. The XPS measurements show that leadpigment con tent of the outermost pellet layer varied greatly with the resin used, being highr for PVG This is probably a function of the dispersibil ity of the pigment in each resin, the melt viscosity characteristics, and the manufacturing process used. It is probable that different results would be achieved with the same nominal composition but differ ent pellet manufacturing techniques. Therefore we cannot extrapolate reliablyfrom these data the pigment of dust produced by pellet attrition. to estimate Assume that dust has the same composition as the bulk resin. II) Qaandty/Perceutage of Inspinblc Dust Predated from Raaln or Reerind Material Data obtained show; Attrition of PE pellets far one hour produced 12 ppm weight of dust less 1T7 micron particle size, Le. inspirable dust Grinding ABS, PVC and HDPE scrap produced 13 to 20 ppm of dust less than 130 micron particle size. BOR 000413 -9 Thus in all four samples the inspirable dust is well below 0.1% by weight. Therefore, regardless of the composition of this dust, it can be claimed that the 'available* pigment which could cause worker expo sure is well below 0.1% and therefore the material should be excluded from WHMIS classification. The level of inspirable dust produced by a wide range of resins is not known. Therefore the broad applicability of this argument is not known. iii) Percentage of Lead. Extracted from Resin bv 5% HC1 In all cases for resins tested by extracting whole pellets with 5% aqueous HG at 20*C (see Table) the amount of Pb extracted was well below 0.06% Pb ( 0.1% PbCrO^) calculated on a pellet basis. Therefore the `available* pigment is below 0.1% and provided that this test is an acceptable criterion for extractability, the material should be excluded from WHMIS classification. The extraction level for smaller resin particles and for other formulations is not known, and therefore the broad applicability of the argument is not known. 3. HQW RELIABLY CAN A SUPPLIER PREDICT USES FOR PLASTIC RESIN PRODUCTS? Arguments have been made that the use of professional judgement, in class ifying resin products based on the potential for exposure, is not valid because suppliers cannot understand or predict end uses to which customers may subject their products. This is an important and basic issue which SPI must deal with in a positive manner. Arguments that plastic resin suppliers can reasonably predict end uses include the following Plastic resins have limited high temperatures for use consistent with preserving the useful, physical properties of the resin. At excessively high temperatures, plastics decompose or depolymerize producing a variety of low molecular byproducts and the hazardous consequences of such high temperature treatment are described on MSDS. Resins are sold in granular form, have particle sizes ranging from cylin drical pellets approx. 1/4* diameter, to spherical beads, to free flowing powder. Customers may grind resin to produce fine particles for unusual purposes and the adverse or hazardous consequences of grinding ream should be known to the supplier and described on MSOS. Commercial plastic resins are relatively chemically inert and MSDS should describe their chemical reaaivity and incompatibility. It is difficult to identify any other end use issues apart from tempera ture, size reduction and chemical reactivity which could apply to plastic resins. If these three topics cover all significant end use issues then it should be possible to claim that suppliers can confidently consider end uses in using professional judgement to make WHMIS classifications. BOR 000414 - 10 4, disclosure of generic chemical identity OF HAZARDOUS INGREDIENTS IN PLASTIC RESINS la cases where the detailed identity and concentration of hazardous ingre dients in resins is not disclosed because WHMIS exemption if claimed, it is recommended that the generic chemical identity of these ingredients be provided so that users may take appropriate precautions in the event of fires, disposal by incineration, contamination of soil or water, etc. BOR 000415 APPENDIX 1 SCANNING ELECTRON MICROSCOPY OF PLASTIC RESINS CONTAINING LEAD CHROMATE PIGMENT A summary of conclusions obtained from SEM photomicrographs is given below. Photographs of the Polaroid photomicrographs are also provided (not included). DEFINITION OF TERMS USED ON PHOTOMICROGRAPHS SEI * Secondary Electron Image * Surface Topography BSE * Back Scattered Electron Image - More dense particles at or near surface show up as lighter on back* ground Pb MAP * Light coloured spot identifies Pb-containing material * at or near surface In SEM the electron beam penetrates the surface and causes backscattering. Penetration is a function of density of material and may be up to approx. 1 micron - 10,000 A. BOR 000416 2- - 2%_PbCr04 in ABS SEM by Queen's University 1. Cylindrical Surface BSE image shows pigment particles at, or close to, surface and corres ponding to high points on secondary electron image (SEI). Pb MAP confirms that particles contain Pb and correspond to SEI and BSE image. 2. Fractured Surface BSE matches SEI for location of pigment particles. Number of particles is not significantly different from cylindrical surface at same magnification. PbCrO< In PE SEM bv Queen's University 1. m pfectf4 a. Cylindrical Surface Pigment particles visible at/near surface by SEI and BSE. clusive - not exposed long enough (see #1, 2). Pb MAP incon b. Fracture Surface Pigment particles visible at/near surface by SEI and BSE. *Pits* visible at fracture point where pigment particles were located. Pb MAP inconclusive since not adequately exposed (see #4, S, 7). c* Flat End of PdUt Pigment particles visible at/near surface by SEI and BSE. Pb MAP inconclu sive. Stnmures like pits visible (see #8, 9). 2. 2% PbCrO< a. Cylindrical Snrface Pigment panicles visible at/near surface by SEI/BSE. particle identification (see #11, 12, 13). b. Flat End of Pellet Pigment particles visible by SEI (see #14). Pb MAP confirms BOR 000417 3- - PbCrOj in Polvethvlene fPE) SEM- by University of Western Ontario 1- 50<fr PbCrOj Cylindrical Surfact Pigment particles visible at or near surface. SEI only. No BSE or Pb MAP made (see photos #33, 34, 35). 2. 2% PbCrOj a- Cylindrical Surface Pigment particles visible at/near surface by SEL. Pb MAP not very clear but supports identification. High mag. view shows particles bound in resin (see photos #22, 23, 25). b. Fracture Surface Pigment particles visible at/near surface by SEL cation. Pigment particles more visible, i.e. drical surface. 'Pits* visible where pigment photos #26, 30). Pb MAP supports identic* at surface, than at cylin particles were located (see PKrt4..la PVC SEM bv Queen's University 1. 2% PbCrOj - SEM Cylindrical Surface . Sample M High pigment level at or dose to surface. Pb MAP confirms BSE location of pigment particles (see #1001, 1003, 1002, 1004, 1005). 2. Fractured Surface Appears to be leu pigment than at surface. 33% PbCrO| - SEM . Sample N Cylindrical Surface Very high pigment level at or dose to surface. view. BSE confirms high pigment at/dose to surface. Some surface occlusions in one BOR 000418 APPENDIX II X-RAY PHOTON SPECTROSCOPY OF PLASTIC RESINS CONTAINING LEAD CHROMATE PIGMENT A summary of conclusions obtained from XPS analysis of resins is given in the attached Surface Saence Western report dated August 1989. XPS provides quantitative results in Atom % & weight %. Atom % is converted to weight % by multiplying atom % by element atomic weight and normalizing these values to calculate % weight by each element involved. BOR 000419