Ann. occup. Hyg., Vol. 47, No. 4, pp. 305–312, 2003 © 2003 British Occupational Hygiene Society Published by Oxford University Press DOI: 10.1093/annhyg/meg043

Permeation of 70% Isopropyl Alcohol Through Surgical Gloves: Comparison of the Standard Methods ASTM F739 and EN 374 ERJA A. MÄKELÄ*, SINIKKA VAINIOTALO and KIMMO PELTONEN† Department of Industrial Hygiene and Toxicology, Finnish Institute of Occupational Health, Topeliuksenkatu 41 a A, FIN-00250 Helsinki, Finland Received 1 July 2002; in final form 17 December 2002 Standard test methods ASTM F739 and EN 374 were compared by assessing the permeation of 70% isopropyl alcohol (2-propanol) through seven brands of surgical gloves. The two standards differ in the flow rates of the collection medium and in the chemical permeation rate at which the breakthrough time (BTT) is detected, the EN detection level being 10 times higher than the permeation rate used by ASTM. In a departure from the EN standard method, a 4 h testing time was used instead of 8 h. All of the tested gloves were from the same manufacturer and were made from either natural rubber (NR) (six brands) or chloroprene rubber (CR) (one brand). Two of the NR glove brands were double layered. For the thin NR gloves (0.22, 0.28 and 0.27 mm) the permeation rates were higher throughout the tests with a flow rate of 474 ml/min (EN) of the collection medium (nitrogen) compared with the permeation rates obtained with a flow rate of 52 ml/min (ASTM). These resulted in BTTs of 4.6, 6.5 and 7.6 min (EN) and 4.8, 6.5 and 9.1 min (ASTM), respectively. No statistical difference could be observed between the BTT values obtained with the two standard methods for any of the thin gloves. Thus, although the ASTM standard has a lower criterion for the detection of permeation, it does not necessarily produce shorter BTTs. For the better barriers the methods yielded more equivalent permeation rate curves and thus the EN BTTs were longer than the ASTM BTTs: the EN results were 21, 80, 122 and >240 min compared with the ASTM results of 12, 32, 38 and 103 min for glove thicknesses of 0.37 (NR), 0.22 + 0.22 (double layered NR), 0.31 + 0.29 (double layered NR) and 0.19 mm (CR), respectively. Keywords: gloves; isopropyl alcohol; permeation; 2-propanol; standards; surgical

Chemical permeation of a volatile test chemical is measured periodically from the nitrogen or air collection medium. Both standards have been developed in order to assess how quickly and to what extent a chemical can gain access to the skin through protective clothing. Breakthrough time (BTT) and steady-state permeation rate (SSP) have been established as criteria of the resistance to permeation. The BTT is the time between the application of the test chemical onto the outer surface of the sample material and detection of the test chemical permeating the material with a specified rate (Pb). In the ASTM standard, this definition applies to ‘the normalized breakthrough time’. The SSP is the value from which the permeation rate does not grow and only the ASTM standard requires its measurement. The major differences in the two standards are the Pb and the flow rate of the gaseous

INTRODUCTION

The European standard EN 374 (European Committee for Standardization, 1994) and the American Society of Testing and Materials (ASTM) standard F739 (ASTM, 1999) describe similar test methods for measuring chemical permeation through protective materials. Both standards define the same test cell with two chambers, a flow-through chamber for the collection medium and a chamber with an inlet for a test chemical. A sample of the protective material being studied is clamped between the chambers.

*Author to whom correspondence should be addressed. E-mail: [email protected] †Current address: National Veterinary and Food Research Institute, PO Box 45, FIN-00580 Helsinki, Finland 305

306

E. A. Mäkelä, S. Vainiotalo and K. Peltonen

collection medium. The EN standard has a Pb that is 10 times higher than the ASTM standard. The flow rate in the EN standard (5 test cell vol/min corresponding to 420–520 ml/min) can also be 10 times higher than in the ASTM (50–150 ml/min). The measurement of the maximum permeation rate demands that additional factors that can restrict the mass transfer must be minimized. A slow desorption rate of a chemical from the material surface decreases the permeation rate. The desorption rate can in some cases be improved by increasing the flow rate of the gaseous collection medium (Schwope et al., 1988b; Que Hee, 1996; Anna et al., 1998). On the other hand, this leads to dilution of the test chemical in the collection medium, thus affecting the detection limit of the permeation rate. There are a few previous studies about the influence of the flow rate on the permeation results (Mellström et al., 1989, 1991; Mellström, 1991a,b,c; Zellers and Sulewski, 1993). Only five tests in these studies (Anna et al., 1998) have been carried out with the collection medium flow rates as defined by the two standards and the large standard test cell (glove sample diameter inside the cell 5.1 cm). These studies clearly state that in some tests an increase in the flow rate can yield higher SSP values, but the differences in BTTs have not been considered significant. A theoretical evaluation has shown that the BTTs are longer if the flow rates are increased (Schwope et al., 1988b). However, at the time of the Schwope study, calculation of the BTT was based on the detection limit of the analytical method. The current calculation is now based on a fixed permeation rate (Pb) and thus a reevaluation is needed. Isopropyl alcohol (IPA) (2-propanol) is frequently used as a disinfectant or a solvent for other chemicals in a number of healthcare products. IPA can be an irritant and can dry the skin or permeate through

the skin to the systemic circulation (Jensen, 1981; Fiserova-Bergerova et al., 1990). It is also known to be able to permeate through rubber gloves. The danger in its permeation through the gloves to the skin is that it may transfer other hazardous components in solution. For industrial gloves, the permeation of 100% IPA has been frequently studied, but only a few permeation tests have been reported for surgical or examination gloves. BTTs as low as 1 min have been reported (Mellström et al., 1992; Forsberg and Keith, 1999). Due to the high expected permeability, short BTTs and the common use of diluted IPA, we selected 70% IPA for use in these tests instead of 100%. The permeation of 70% IPA was studied through single layer natural rubber (NR), double layer NR and single layer chloroprene (CR) surgical gloves according to the ASTM F739 and the EN 374 standards. The permeation results were compared between single and double layer gloves and gloves of different materials and thicknesses. As all the tests were carried out by both standards, the effects of the different flow rates could be evaluated for the permeation of IPA. MATERIALS AND METHODS

The seven brands of surgical gloves (Table 1) were supplied by SSL International plc (Oldham, UK). Six of the glove materials had a polymeric Biogel® inner coating. Two brands were double gloves with a dark green glove to be worn underneath a pale yellow one; this allows holes in the gloves to be seen during surgical operations. The test chemical was made by mixing together isopropyl alcohol (p.a. grade; Merck) and HPLC grade water in the proportions 70:30 (v:v). Test samples with an average diameter of 9.0 ± 0.1 cm were cut from the palms of the gloves. The

Table 1. Glove materials Glove

Material

No. of test samples

Thickness (mm) (mean ± SD)

Weight per unit area (g/m2) (mean ± SD)

Biogel® Super Sensitive

NRa + BCb

40

0.22 ± 0.01

197 ± 6

Regent Surgical

NR

40

0.28 ± 0.01

220 ± 7

Biogel®

NR + BC

21

0.27 ± 0.02

244 ± 11

Biogel® Orthopaedic

NR + BC

21

0.37 ± 0.01

302 ± 6

Biogel® Indicator

NR + BC

Inner glove

21

0.22 ± 0.01

199 ± 5

Outer glove

21

0.22 ± 0.01

202 ± 5

Inner glove

21

0.31 ± 0.01

251 ± 5

Outer glove

21

0.29 ± 0.01

241 ± 6

40

0.19 ± 0.01

255 ± 8

Biogel® Reveal

Skinsense™ N aNatural

rubber. coating. cChloroprene rubber. bBiogel®

NR + BC

CRc + BC

2-Propanol permeation through surgical gloves

diameters were measured in three directions for every sample. Glove samples were weighed using an analytical balance with a precision of 0.1 mg. The weight per unit area values were calculated. The sample thicknesses were measured (Ames M034 E; Shirley Developments Ltd) using a pressure of 3.4 kPa and measuring diameter of 28.7 mm (Table 1). Three values were measured for each sample. The samples were not inspected for structural imperfections before the chemical permeation testing took place. As required by both standards, ASTM F739 and EN 374, a circular glove sample was mounted between the two chambers of the standard test cell. Collection medium was set to flow through the chamber, which had the inner surface of the glove sample as one wall. Sampling of the collection medium for chemical analysis was then started. The test chemical was then injected into the chamber with the outer surface of the glove sample, at the starting time for the BTT determination. Sampling was terminated after 4 h, or earlier if the test chemical was found to permeate the glove material at a steady rate. For the EN standard, values in excess of a Pb of 1.0 µg/cm2/min would have sufficed. The use of a more appropriate testing time considering the usage of disposable gloves, i.e. 4 instead of 8 h, was the only modification made to the standards in this study. The laboratory and the collection medium temperatures were kept at 23.0 ± 1.0°C and the test chemical temperature was adjusted before the test to 23.0 ± 0.1°C using a water bath. BTT is the time between the application of the test chemical into the test cell and the time when the permeation rate exceeds a Pb of 0.1 µg/cm2/min (ASTM) or 1.0 µg/cm2/min (EN). At least three parallel tests were carried out and the final BTT result is the average of the three measurements. The permeation rate (P, µg/cm2/min) is calculated as follows in the open loop methods (i.e. without re-circulating the collection medium). P = c × v/A

(1)

where c is the test chemical concentration (mg/l), v is the collection medium flow rate and A is the exposed area of the test sample. According to the EN standard it is assumed that the entry of the test chemical into the collection medium is constant between two consecutive samplings. Thus the BTT (min) is calculated from the first degree equation: BTT = t1 + (t2 – t1) × (Pb – P1)/(P2 – P1)

(2)

where t is the time from the injection of the test chemical into the test cell (min), index 1 refers to the first measuring point just before permeation exceeds the Pb defined by the standard and index 2 refers to the

307

first measuring point after which P has exceeded the Pb. The ASTM standard requires the calculation of the average SSP, if the permeation rate reaches a constant level. The SSP is calculated from a few points at the end of the three parallel tests. t-Tests (Microsoft® Excel 97 SR-2, T-test: TwoSample Assuming Equal Variances) with a significance level of 0.05 were applied for statistical comparisons between the ASTM and the EN standard results or between results obtained with different flow rates and the same permeation rate criteria. Samples were injected (EC6W valve with valve actuator E60; VICI Valco Instruments Co. Inc., Houston, TX) from the selected collection medium flow line [six port valve SF6P (VICI Valco Instruments Co. Inc.) with a Smart Drive valve actuator (Omnifit Ltd, Cambridge, UK)] into a gas chromatograph (Micromat HRGC 412, flame ionization detector; HNU Nordion, Helsinki, Finland). Both valves were controlled by a computer program written for the system (Valve Actuator Program, GC Version 1.0, Matti Jussila). A HP-5 column (30 m × 0.53 mm, film thickness 2.65 µm; Hewlett Packard, Little Falls, Wilmington, DE) was used at a constant oven temperature of 45°C and a single chromatographic run was used for the whole test. The sample loop had been spliced from 1/16 inch and 1/8 inch tubing (HNU Nordion) and its volume was ∼1 ml. The flow rate of the helium carrier gas was 8.5 ml/min and an injection split of 1:3 was used (Pekari et al., 1992). The system allows the testing of between one and three protective material samples at the same time. The collection medium was nitrogen gas, with flow rates (Electronic flow meter 5182-0879; Hewlett Packard) of 52.0 ± 1.4 (ASTM) or 474 ± 2 ml/min (EN). The flows were split before the regulator valve to avoid a build up of pressure in the system. The splitting system consisted of plastic T-pieces and flow restrictors (flow rate 474 ml/min, disposable 0.80 × 38 mm needle; flow rate 52 ml/min, 0.45 × 12 mm needle) for the free ends of the T-pieces. The tests of rapid permeation (BTT < 20 min) were carried out with one glove sample at a time and the tests with slower permeation with two glove samples at the same time. In both cases the sampling frequency into the gas chromatograph was once every 2 min. The sampling programs included regular injections of pure nitrogen to ensure that the system had not become contaminated. For both the ASTM and EN methods, quantification was carried out using the same range of calibration samples. Isopropyl alcohol calibration samples were prepared in plastic laminated aluminum foil bags at concentrations of 30–1500 µg/l in nitrogen. Calibration samples were introduced through the valve system into the chromatograph with plastic

308

E. A. Mäkelä, S. Vainiotalo and K. Peltonen

Table 2. Isopropyl alcohol (70% v/v) breakthrough times measured for surgical gloves with two collection medium flow rates and calculated at two permeation rate levels Breakthrough time (min) (mean ± SD) Standard used

EN 374

Flow rate (ml/min)

474

ASTM F739 52

474

52

Permeation rate (µg/cm2/min1)a

1.0

0.1

0.1

Biogel® Super Sensitive

4.6 ± 0.6b

4.8 ± 0.9b

2.1 ± 0.1

Regent® Surgical

6.5 ± 0.5b

6.5 ± 0.9b

3.4 ± 0.3

12 ± 1

Biogel®

7.6 ± 0.5b

9.1 ± 1.5b

3.5 ± 0.7

18 ± 3

Biogel®

Orthopaedic

Biogel® Indicator

21 ± 2

12 ± 0c

80 ± 13d

32 ± 2c

30 ± 5c

38 ± 3c

37 ± 3c

122 ± 6

Biogel® Reveal Skinsense™ N

>240

103 ±

7c

9.8 ± 1.4c

97 ± 2c

1.0 9.6 ± 1.7

25 ± 2 121 ± 26d 153 ± 18 >240

The results which have been obtained under different test conditions and which do not differ statistically (P > 0.05) have been indicated. aThe level of the rate used for the determination of the BTT. bThe ASTM result does not differ from the EN result statistically. cThe result with the flow rate 52 ml/min does not differ statistically from the result with the flow rate 474 ml/min using Pb = 0.1 µg/cm2/min. dThe result with the flow rate 52 ml/min does not differ statistically from the result with the flow rate 474 ml/min using Pb = 1.0 µg/cm2/min.

luer-lock syringes (10 ml, Discardit™ II; Becton Dickinson, Madrid, Spain). Regression analyses (Microsoft® Excel 97 SR-2) were applied to the calibration responses and they produced linear lines which were forced to 0 (r = 0.999 ± 0.001, P = 3.2 ± 0.1 × 10–15). Calibration was performed daily before testing and was checked after each test. RESULTS

The thicknesses of the gloves were 0.22, 0.28, 0.27 and 0.37 mm for the single layered NR gloves Biogel® Super Sensitive, Regent® Surgical, Biogel® and Biogel® Orthopaedic, respectively. Weight per unit area values were 197, 220, 244 and 302 g/m2. The Biogel® Indicator and Biogel® Reveal gloves were double gloves, whose inner and outer gloves had about the same thickness and weight per unit area. Both Indicator gloves were equal in their thickness and weight per unit area to the Super Sensitive gloves and both Reveal gloves were approximately the same as the standard Biogel® gloves. The CR gloves, Skinsense™ N, were the thinnest of all at 0.19 mm, but they had a relatively high weight per unit area value of 255 g/m2. Further information on the gloves is shown in Table 1. The BTTs using EN 374 to determine the permeation of 70% IPA were 4.6, 6.5, 7.6, 21, 80 and 122 min for the NR gloves Biogel® Super Sensitive, Regent® Surgical, Biogel®, Biogel® Orthopaedic, Biogel® Indicator and Biogel® Reveal, respectively. Following the same order of increasing weight per unit area, the ASTM test results for the NR gloves were 4.8, 6.5, 9.1, 12, 32 and 38 min. The permeation rate of IPA did not reach the Pb of 1.0 µg/cm2/min

of the EN standard in the 4 h test with the Skinsense™ N gloves, but reached the ASTM Pb of 0.1 µg/cm2/min after 103 min. The BTTs and the corresponding standard deviations are shown in Table 2. Figure 1 illustrates the increase in the permeation rates during the tests. The points where the curves cross the Pb of 1.0 µg/cm2/min can be observed as the corresponding BTTs. In all of the EN testing, the BTTs for the parallel tests were within the required range ± 20% of the mean. The BTT results were calculated for both of the Pb values and for both of the flow rates. When viewed with the same flow rate (Table 2 and Fig. 2), the Pb of 1.0 µg/cm2/min yielded about double or higher BTT values than the Pb of 0.1 µg/cm2/min. When comparing the ASTM results with the corresponding EN results, it was found that the results of the thin NR gloves (Biogel® Super Sensitive, Regent® Surgical and Biogel®) did not differ statistically even though there was a 10-fold difference in their Pb values. On the other hand, for the different flow rates but the same Pb, a statistical difference was found between the BTTs at the Pb of 0.1 µg/cm2/min for the above mentioned thin NR gloves and at the Pb of 1.0 µg/cm2/min for all the gloves except the Biogel® Indicator gloves. SSP rates according to the ASTM standard were attained in the single layered NR testing. The average SSPs of 70% IPA were 4.2, 4.2, 3.1 and 2.7 µg/cm2/min for Biogel® Super Sensitive, Regent® Surgical, Biogel® and Biogel® Orthopaedic, respectively. The remaining gloves did not reach the SSP during the 4 h test. SSP values were also calculated for the tests using the EN flow rate of 474 ml/min: the SSPs of IPA were 7.2, 5.3, 5.3, 3.2 and 1.7 µg/cm2/min for

2-Propanol permeation through surgical gloves

309

Fig. 1. Permeation of 70% isopropyl alcohol through surgical rubber gloves. The ASTM F739 test was made with a collecting medium flow rate of 52 ml/min (crosses) and the EN 374 test with a rate of 474 ml/min (diamonds).

310

E. A. Mäkelä, S. Vainiotalo and K. Peltonen

Fig. 2. Breakthrough times of 70% isopropyl alcohol through surgical rubber gloves calculated for different flow rates and breakthrough detection levels. The test conditions are indicated for each series of points as collection medium flow rate (ml/min)/breakthrough detection level (µg/cm2/min). The error bars represent the standard deviations.

Table 3. Average (n = 3) steady-state permeation rates (SSPs) for isopropyl alcohol (70% v/v) SSP (µg/cm2/min) (mean ± SD) Standard used

EN 374

ASTM F739

Flow rate (ml/min)

474

52

Biogel® Super Sensitive

7.2 ± 0.8

4.2 ± 0.2

Regent Surgical

5.3 ± 0.3

4.2 ± 0.6

Biogel®

5.3 ± 0.2

3.1 ± 0.5

Biogel® Orthopaedic

3.2 ± 0.0

2.7 ± 0.2

Biogel® Indicator

1.7 ± 0.1

Not detected

The SSP values are required by the ASTM F739 standard but not by EN 374.

Biogel® Super Sensitive, Regent® Surgical, Biogel®, Biogel® Orthopaedic and Biogel® Indicator gloves, respectively (Table 3). The SSPs differed statistically significantly for the tests with different flow rates. DISCUSSION

Surgical or other thin rubber gloves may often provide insufficient protection against small organic molecules. If the glove is to have any protective func-

tion against hazardous chemicals, it should have a BTT of at least 10 min and twice the working time with the chemical. The difference between the recommended time of use for the gloves and the BTT arises from the differences in conditions between everyday actual use and the standardized test method (Leinster et al., 1990; Leinster, 1994). Thus the tested single layered NR gloves cannot be recommended against 70% IPA solutions, except for occasional very brief tasks. Even then there should not be dermally toxic substances present which could permeate along with the IPA. For a short period of time, the double layered and the CR gloves should provide protection. This study shows that flow rate is a critical factor in tests in which a gaseous collection medium is used. Compared with the flow rate of 474 ml/min, the lower flow rate of 52 ml/min yielded lower permeation rates throughout the tests and thus made the BTTs longer in five of the seven test pairs when detected at the Pb of 1.0 µg/cm2/min and in three of the seven test pairs when detected at the Pb of 0.1 µg/cm2/min (Table 2). The differences in the BTTs with the different flow rates at a Pb of 1.0 µg/cm2/min were so marked that they would influence how three

2-Propanol permeation through surgical gloves

of the gloves (Regent® Surgical, Biogel® and Biogel® Indicator) would be classified for protection against IPA according to the EN standard. The BTTs for the three thin NR materials (Biogel® Super Sensitive, Regent® Surgical and Biogel®) were about the same when tested using the two different standards, despite the 10-fold difference in the Pb values (Table 2 and Fig. 2). It seems that in the tests of these three materials with the collection medium flow rate of 52 ml/min the vaporization of IPA suffered from the low mixing efficiency in the test cell, which has been discussed by Anna et al. (1998). Thus their BTTs became about twice as long as those measured with the higher flow rate, if calculated at the same Pb. Statistical examination of the BTTs calculated at the same Pb revealed that the BTTs of the better barriers (Biogel® Orthopaedic, Biogel® Indicator, Biogel® Reveal and Skinsense™ N) did not differ between the two flow rates in all but two cases (Biogel® Orthopaedic and Biogel® Reveal, Pb 1.0 µg/cm2/min). The slow permeation probably allowed better mixing and vaporization. Thus, because of the difference in the Pb values, it was possible that the ASTM method could yield clearly shorter BTTs than the EN method for these good barriers, as expected. Anna et al. (1998) examined the effect of the collection medium flow rate in the ASTM F739 method. They performed 176 chemical permeation tests and concluded that since the SSP values increase for many test chemicals when the flow rate is increased, the standard method must be changed to determine the maximum SSP. They reported BTTs for five chemical/glove pairs with collection medium flow rates of 50 and 500 ml/min and noticed that the higher flow rate appeared to shorten the BTT, although they did not detect any actual statistically significant differences. The differences of our findings compared with the Anna et al. study are due to the different test chemicals and glove materials. In other studies there is less evidence of the BTTs shortening with an increase in the collection medium flow rate, but the tests have been performed under different test conditions (Mellström et al., 1989, 1991; Mellström, 1991a,b,c; Zellers and Sulewski, 1993). In those studies a narrower range of collection medium flow rates or much higher flow rates were used compared with the EN and the ASTM standard test conditions, if the flow rates are calculated as the test cell volume changes per minute. In the case of NR materials, the thicker the glove material the longer the BTT values. The Regent® Surgical glove, which is manufactured differently, was in line with the other gloves. Its SSP and BTT values measured under the standard conditions did not differ from the corresponding values of the Biogel® gloves. The weight per unit area was in between the Biogel® Super Sensitive and the Biogel® gloves and its thick-

311

ness was the same as the Biogel® gloves. Thus there was no evidence that the Biogel® polymer layer made the NR material less permeable to IPA. Skinsense™ N proved to be the best barrier against IPA of the tested gloves. It was the thinnest of the materials and the least permeable. CR surgical gloves have also proved to be effective against methacrylates and formaldehyde (Schwope et al., 1988a; Mäkelä et al., 1999, 2003). However, the CR gloves meant for healthcare are only sold as sterilized surgical gloves, although they could also have potential in providing protection against hazardous chemicals, e.g. for laboratory workers. At present their price may hinder their more extensive use outside the operating theatre. The BTT values for the double layered gloves were about five times longer than for the single layered gloves of similar thicknesses in the ASTM test and about 16 times longer in the EN test. Thus it can be concluded that irrespective of the testing method, the protection provided by double layered gloves against IPA is more than double the protection provided by single layered gloves. This observation can be explained by the chemical having to adsorb, diffuse and desorb twice and the differences in the concentration gradients driving the chemical across the systems. The advantage of double gloving has been noted previously for protection against microorganisms and chemicals (Mellström, 1991a; Gerberding et al., 1995; Heller and Greer, 1995; Page, 1997; Mäkelä et al., 1999, 2003). Even if the regulations require the use of a specific standard test method when certifying chemical protective gloves, they do not forbid the use of other available test results when selecting the gloves. Both the standards, ASTM F739 and EN 374, can be used to differentiate totally unsuitable gloves from gloves which provide adequate protection. Often there is a shortage of information when gloves are to be selected. Therefore, the instructions for use, in addition to the required standard information, should also include information obtained with other test methods. However, if the permeation test results are to be of value, the applied methods must also be described. Both standards are meant to specify a test method which yields reproducible and comparable results that describe the resistance of a protective material to chemical permeation. The collection medium flow rate of the ASTM method is 100 ± 50 ml/min, which corresponds to 1.0 ± 0.5 chamber vol/min. This is low and the range is also wide, which should be considered as these factors may influence the permeation results. On the other hand, the low flow rate and thus the high concentration of the test chemical in the collection medium allows the use of a low Pb (0.1 µg/cm2/min). Thus permeation which cannot be determined by the EN standard test can sometimes be measured using the ASTM test. The higher flow rate

312

E. A. Mäkelä, S. Vainiotalo and K. Peltonen

of the EN test lessens the impact of the desorption rate on the permeation results. The higher permeation rate produced by the higher collection medium flow rate also results in a steeper slope of the permeation rate versus time curve, which makes determination of the BTT more accurate. Thus the different conditions of the two standards both have their advantages and without a large number of comparable test results, it is quite difficult to define which represents the better way of determining the BTT. Acknowledgements—The authors wish to express their gratitude to Matti Jussila, the scientist, the technician and the computer programmer. The study was made possible by the kind support of SSL International plc, Oldham, UK.

REFERENCES ASTM. (1999) Standard test method for resistance of protective clothing materials to permeation by liquids or gases under conditions of continuous contact, ASTM F 739-99a. West Conshohocken, PA: American Society of Testing and Materials. Anna DH, Zellers ET, Sulewski R. (1998) ASTM F 739 method for testing the permeation resistance of protective clothing materials: critical analysis with proposed changes in procedure and test-cell design. Am Ind Hyg Assoc J; 59: 547–56. European Committee for Standardization. (1994) European standard EN 374-3: 1994. Protective gloves against chemicals and micro-organisms—Part 3: Determination of resistance to permeation by chemicals. Brussels: CEN. Fiserova-Bergerova V, Pierce JT, Droz PO. (1990) Dermal absorption potential of industrial chemicals: criteria for skin notation. Am J Ind Med; 17: 617–35. Forsberg K, Keith LH. (1999) Chemical protective clothing— performance index, 2nd Edn. New York: John Wiley & Sons. Gerberding JL, Quebbeman EJ, Rhodes RS. (1995) Hand protection. Surg Clin N Am; 75: 1133–9. Heller ET, Greer CR. (1995) Glove safety: summary of recent findings and recommendations from health care regulators. South Med J; 88: 1093–8. Jensen O. (1981) Contact allergy to propylene oxide and isopropyl alcohol in a skin disinfectant swab. Contact Dermatitis; 7: 148–50. Leinster P. (1994) The selection and use of gloves against chemicals. In Mellström GA, Wahlberg JE and Maibach HI,

editors. Protective gloves for occupational use. Boca Raton, FL: CRC Press, pp. 269–82. Leinster P, Bonsall JL, Evans MJ, Lewis SJ. (1990) The application of test data in the selection and use of gloves against chemicals. Ann Occup Hyg; 34: 85–90. Mäkelä EA, Väänänen V, Alanko K, Jolanki R, Estlander T, Kanerva L. (1999) Resistance of disposable gloves to permeation by 2-hydroxyethyl methacrylate and triethyleneglycol dimethacrylate. Occup Hyg; 5: 121–9. Mäkelä EA, Vainiotalo S, Peltonen K. (2003) The permeability of surgical gloves to seven chemicals commonly used in hospitals. Ann Occup Hyg; 47: 313–23. Mellström G. (1991a) Protective gloves of polymeric materials. Experimental permeation testing and clinical study of side effects. Acta Derm Venereol Suppl (Stockh); 163: 1–54. Mellström GA. (1991b) Comparison of chemical permeation data obtained with ASTM and ISO permeation test cells. I. The ASTM standard test procedure. Ann Occup Hyg; 35: 153–66. Mellström GA. (1991c) Comparison of chemical permeation data obtained with ASTM and ISO permeation test cells. II. The ISO/DIS standard test procedure. Ann Occup Hyg; 35: 167–80. Mellström GA, Landersjö L, Boman AS. (1989) Permeation of neoprene protective gloves by acetone: comparison of three different permeation cells in an open-loop system. Am Ind Hyg Assoc J; 50: 554–9. Mellström GA, Landersjö L, Boman AS. (1991) Permeation testing of protective gloves by using two different permeation cells in an open-loop system (neoprene-toluene). Am Ind Hyg Assoc J; 52: 309–14. Mellström GA, Lindberg M, Boman A. (1992) Permeation and destructive effects of disinfectants on protective gloves. Contact Dermatitis; 26: 163–70. Page D. (1997) Study points to latex glove permeability. Nephrol. News Issues; 11: 28–30. Pekari K, Vainiotalo S, Heikkilä P, Palotie A, Luotamo M, Riihimäki V. (1992) Biological monitoring of occupational exposure to low levels of benzene. Scand J Work Environ Health; 18: 317–22. Que Hee SS. (1996) Simple glove permeation models. Appl Occup Environ Hyg; 11: 117–24. Schwope A, Costas PP, Mond CR et al. (1988a) Gloves for protection from aqueous formaldehyde: permeation resistance and human factors analyses. Appl Ind Hyg; 3: 167–76. Schwope A, Goydan R, Reid RC, Krishnamurthy S. (1988b) State-of-the-art review of permeation testing and the interpretation of its results. Am Ind Hyg Assoc J; 49: 557–65. Zellers ET, Sulewski R. (1993) Modeling the temperature dependence of n-methylpyrrolidone permeation through butyl- and natural-rubber gloves. Am Ind Hyg Assoc J; 54: 465–79.