Ann. occup. Hyg., Vol. 45, No. 5, pp. 381–384, 2001 2001 British Occupational Hygiene Society

Published by Elsevier Science Ltd. All rights reservedPrinted in Great Britain.

0003-4878/01/$20.00PII: S0003-4878(00)00059-4

Dustiness of Different High-Temperature InsulationWools and Refractory Ceramic FibresP. CLASS†*, P. DEGHILAGE† and R. C. BROWN‡†Thermal Ceramics de France S.A., Route de Lauterbourg, BP 148, F-67163 Wissembourg Cedex,France; ‡Toxicology Services, 4 Bramble Close, Uppingham, Rutland LE15 9PH, UK

Recent regulations are encouraging the replacement of older types of man-made mineralfibre by more soluble and, thus, less biopersistent compositions. In order for there to be anyhealth benefits from this policy and to gain maximum benefit from such substitutions, theuse of the new materials should not increase exposure. The work reported here was under-taken to investigate the use of new high-temperature glass insulation wools in place of refrac-tory ceramic fibres (RCF). Airborne fibre levels occurring during the manufacture of bothRCF and calcium magnesium silicate wools (CMS) were compared using measurements ofgenuine workplace exposure from a routine monitoring operation on the same plant.Exposures during use were compared in one customer facility where RCF and CMS blanketswere used for the same task. Further comparisons were made in a laboratory test of dustinessusing a “shaking box test”. For some manufacturing tasks there are only a few workplacesamples and there are few opportunities for genuine comparisons with both RCF and CMSin identical uses. However, both materials produced very similar exposure levels duringmanufacture, use and in the laboratory test. The novel magnesium silicate fibre was signifi-cantly dustier in the laboratory test. 2001 British Occupational Hygiene Society. Publishedby Elsevier Science Ltd. All rights reserved

Keywords: mineral fibres; fibres; dustiness; substitution

INTRODUCTION

In December 1997 the European Commission pub-lished their hazard classification of various glass insu-lation wools (Directive 97/69/EC). The names usedin the scheme adopted are applied to fibres defined,or distinguished from one another, in a way differentfrom that used in any previous classification scheme.For example, most of the “special purpose fibres” inthe IARC (IARC, 1988; IPCS, 1988) scheme wouldbe classified as mineral wools under these EU rules.Although the fibres concerned only cause itching,they are classified as irritants, that is, a descriptionpreviously limited to materials causing skin inflam-mation or damage. Most importantly, for carcinogenclassification the European Union distinguishesbetween two different types of “fibres with randomorientation” (that is, they are wools, not continuous

Received 10 April 2000; in final form 23 June 2000.*Author to whom correspondence should be addressed. Tel.:+33-8854-9550; fax: +33-8854-2920; e-mail: phclass@easy-net.fr

381

filaments). The two types are distinguished by com-position, as follows:

1. Those with an alkaline oxide and alkaline earthoxide (Na2O + K2O + CaO + MgO + BaO) con-tent less than, or equal to, 18% by weight. Theseare called “refractory ceramic fibres and specialpurpose fibres”, and are classified as category 2carcinogens.

2. Those with a content of more than 18% of theseoxides; these are called “mineral wools”, and areclassified as category 3 carcinogens.

There is further provision for the use of appropriatetoxicological or biopersistence data in order that someof the second group “mineral wools” may be exoner-ated from the carcinogen classification. This is notpossible for the other fibres.

Alkaline metal oxide and earth alkali oxide contenthas an effect on the water solubility of glasses andso is itself a determinant of the persistence of inhaledfibres in the lung—so-called biopersistence. By mak-ing further exoneration available the directive con-firms the benefit of low biopersistence as a determi-nant of low hazard. This could be due to an effect

382 P. Class et al.

of dissolution on some aspect of surface chemistryreducing some chemical activity. Such an effect couldresult in a reduction in the hazardous properties ofthe material itself with each individual fibre being lesshazardous, but such a mechanism is speculative. Amore important role of low biopersistence is probablyin reducing the number of fibres in the lung. For equalexposures, the less biopersistent the fibre the lowerthe dose; and this will remain true whether the doseis expressed as a peak tissue concentration or as someintegral of fibre number over time.

There is an obvious potential for commercialadvantage in offering an exonerated fibre for appli-cations previously using fibres now classified as car-cinogens. This is especially so when the choice isbetween an exonerated fibre and a fibre in category2. The regulations that apply when using category 2carcinogens are far more onerous. However, wheretechnical requirements allow substitution by a lessdurable fibre, a genuine reduction in risk requires thatthe use of such fibres should not result in an increasein exposure. This could negate the advantages of lowbiopersistence and, perhaps more trivially, wouldmake the workplace less comfortable due to irritationor itching from skin contact.

The purpose of the present study was, therefore,to compare the airborne fibre levels that would begenerated by the use of a fibre with reduced biopers-istence in some applications previously using refrac-tory ceramic fibres (RCFs). To do this, the dustinessof RCFs was compared to that of the newer fibresusing both workplace measurements and in a labora-tory test.

METHODS

MaterialsRCFs are alumino-silicate glasses used for insu-

lation applications above 1000°C. For some of theseapplications, various calcium magnesium silicate(CMS) wools are potential substitutes. These containvarious proportions of calcium and magnesium, andat the extreme, only one of these elements may bepresent. While RCFs are classified as category 2, theother fibres used in this study fit the EU mineral wooldescription and have been exonerated from any car-cinogen classification by animal testing.

Workplace fibrous dust levelsThe opportunity to compare dust levels during the

use of fibrous insulation at a customer’s factory wasprovided by an oven producer willing to evaluate thepossibility of replacing RCF by CMS. Sampling con-ditions were organised so that, as far as possible,identical tasks were carried out with RCF and CMSwools and all conditions remained unchanged. Inparticular, the type of oven and production rate werekept constant. The insulation material—a needled

blanket—had the same shape and density for bothRCF and CMS. The same workers performed thesame tasks during both evaluations in order to reducethe influence of individual handling practices on dustconcentration; such an effect is often observed in rou-tine occupational hygiene surveys. Measurementswere carried out at approximately the same time ofthe day and were of a similar duration during thesame shift on two consecutive days. The productsused in this comparison are described in Table 1.

Measurements were also made during fibre pro-duction on manufacturing lines that produced bothRCF and CMS, albeit at different times. Thesemeasurements were a part of the industry’s qualityassured sampling and measurement system (the“CARE Programme”). Thus they were not targetedspecifically at this study but collected using the sam-pling strategy independently devised to ensure fullystatistically valid dust measurements at plants manu-facturing RCFs and CMS fibres (Burley et al., 1997).

Intrinsic dustiness of fibre blanketsDustiness was estimated using a method based on

the shaking “dust box” described in Dodgson et al.(1987). The basic design was modified and the actualequipment and methods used are described in Class etal. (1996). Briefly, ten regularly distributed samples(about 30 g each) were taken from a band, 54 by 90cm, removed from the central part of different CMSand RCF fibre blanket rolls (total dimension 60 by710 cm). The samples were shaken at about 3.3 Hzin a mesh cage, contained in an electrically conduc-tive 70 cm cube PVC box (called here the “shakingbox”). Airborne fibres generated were kept in suspen-sion using a fan with a flowrate of about 50–100 m3

h�1. This method simulates extreme handling con-ditions, and samples were weighed before and aftertesting; the results were discarded if the test samplehad lost more than 3% of its weight during testing.The products tested are described in Table 2. Samples2, 3, and 6 were all calcium magnesium silicatesamples differing in the proportion of calcia to mag-nesia and in the total content of these two oxides.

During testing, the airborne fibre concentrationgenerated in the box was measured by drawingknown volumes of air through a membrane filter con-tained in the chamber by means of a sampling pumplocated outside the box.

Fibre countsBoth workplace and shaking box airborne fibre

concentrations were measured by counting and sizingfibres on the filters using the phase-contrast opticalmicroscope WHO/EURO method (WHO, 1997). Thefibre-counting laboratory was a member of theWHO/EURO Reference Scheme (Crawford et al.,1987).

383Dustiness of High-Temperature Insulation Wools and RCF

Table 1. Description of products used as oven insulation in the comparison study

Product (trade name) Nature Nominal density/thickness (kg m�3)/(mm) Nature of the product

Cerablanket Spun RCF 96/25 Needled blanketSuperwool 607 Spun CMS 96/25 Needled blanket

Table 2. Products examined in the shaking box. All thetest samples were needled blankets manufactured from spun

fibre

Fibre Nature of product Nominaltype density/thickness

(kg m�3)/(mm)

1 Aluminium silicate (RCF) 128/252 Calcium magnesium silicate 128/25

(CMS)3 Calcium magnesium silicate 96/25

(CMS)4 Calcium magnesium silicate 128/25

(CMS) with some zirconia5 Calcium magnesium silicate 96/25

(CMS) with some zirconia6 Calcium magnesium silicate 128/25

(CMS)7 Magnesium silicate 128/25

RESULTS

Occupational hygiene survey resultsTable 3 gives the airborne fibre concentrations dur-

ing operations using CMS and RCF in both ovenassembly and during manufacture.

Shaking box test resultsAfter rejection of three values, one each from fibres

2, 5 and 7, due to excessive loss of sample, the resultsare presented in Fig. 1. All the blankets used were 25mm thick. Two of the fibre types (CMS and CMSwith some zirconia) were available at densities ofboth 96 and 128 kg m�3. The other types were onlyavailable at 128 kg m�2. All the values obtained havebeen plotted so as to avoid making assumptions aboutthe distribution of the values. It can be seen that fibre

Table 3. Results from occupational hygiene samples taken at various workplaces

Workplace No of samples (task length Mean concentration (std dev.) Range f ml�1

weighted averages) f ml�1

RCF CMS RCF CMS RCF CMS

Customer premisesOven insulation mounting 5 5 0.46 (0.25) 0.36 (0.17) 0.12–0.77 0.16–0.63Fibre manufactureFurnace operator 2 2 0.07 0.06 0.05–007 0.05–0.08

Needle operator 3 4 0.07 (0.02) 0.10 (0.07) 0.06–0.18 0.06–0.09End of line operator 10 8 0.22 (0.06) 0.18 (0.07) 0.12–0.37 0.14–0.33Dye cutting operator 25 12 0.39 (0.16) 0.39 (0.17) 0.20–0.83 0.15–0.67

Fig. 1. Aerosol concentration in the shaking box dustiness test.All the values available are plotted, the horizontal dotted linesare the minimum and maximum values for the first six fibreswhich are calcium–magnesium silicate or aluminium silicate

fibres- while fibre 7 is a magnesium silicate fibre.

7 released significantly more fibres than all the otherfibre types.

DISCUSSION

Data collected during production and installation ofCMS showed that the airborne fibre concentrationsoccurring during the use of these products were notdifferent from those occurring when producing orusing RCF. Again in laboratory scale dustiness test-ing, both products behave similarly [sample 1 (RCF)and sample 2 (CMS) both having a density of 128 kg

384 P. Class et al.

m�3 and a thickness of 25 mm] . This is perhaps tobe expected since dustiness depends on fibre fragilityand the coherence, or friability, of the insulation pro-duct as a whole. To be physically suitable for anyapplication these properties cannot be allowed to varyover a wide range.

The data obtained during these tests does appearentirely consistent with the view that the shaking testmethod can be used to reflect the behaviour of pro-ducts in the workplace, at least when comparing pro-ducts with one another. However, the airborne dustconcentrations measured in the shaking box test canonly provide comparative values; these would needextensive calibration against standard fibres used instandardised tasks before they could be used to pre-dict airborne fibre concentrations in the workplace.

For two fibre types, dustiness was measured at twodensities. Fibre types 2 and 3, as well as 4 and 5, dosuggest an inverse relationship, but further studies areneeded in this area. However, the greatest differencesin this study were between different products. Themagnesium silicate fibre (sample 7) is much dustierthan any of the calcium magnesium silicate or RCFblankets, even with the same density and thicknesssamples; conversely blanket sample 6, the mostrecently commercialised composition of calciummagnesium silicate, was less dusty than the otherblankets tested.

We recommend that attention to exposure should

be included in any decision process aimed at replac-ing one material with another, whatever the hazardclassification.

REFERENCES

Burley CG, Brown RC, Maxim LD. Refractory ceramic fibres:the measurement and control of exposure. Annals of Occu-pational Hygiene 1997;41(Suppl 1):267–72.

Class P, Brown RC, Alexander IC, Jubb G, Deghilage P.Evaluation of the relation between the nominal fibre diameterof bulk refractory ceramic fibres and airborne fibre concen-tration (dustiness) using a laboratory shaking test box.Gefahrstoffe, Reinhaltung der Luft 1996;56:319–21.

Crawford NP, Kello D, Jarvisalo JO. Monitoring and evaluat-ing man-made mineral fibres: work of a WHO/EURO refer-ence scheme, In: Walton WH, editor. Man-made mineralfibres in the working environment. Proceedings of an Inter-national Symposium, Copenhagen, 28–29 October 1986.Oxford: Pergamon Press; 1987. p. 557–65 [[Annals of Occu-pational Hygiene 31(4B)]].

Dodgson J, Cherrie J, Groat S. Estimates of past exposure torespirable man-made mineral fibres in the European insu-lation wool industry. Annals of Occupational Hygiene1987;31:567–82.

IARC. Man-Made Mineral Fibres and Radon. Monograph Ser-ies Vol. 43. Lyon (France): International Agency for CancerResearch; 1988. ISBN 92 832 1243 6.

IPCS. Man-made Mineral Fibres. Environmental Health Cri-teria Series No. 77. Geneva (Switzerland): International Pro-gramme for Chemical Safety; 1988.

WHO. The WHO/EURO Technical Committee for Monitoringand Evaluating Airborne MMMF. Reference methods formeasuring airborne man-made mineral fibres (MMMF).Copenhagen: World Health Organisation; 1997.