C L I N I C A L
A N D
E X P E R I M E N T A L
OPTOMETRY ORIGINAL PAPER
Effect of multiple antireflection coatings on impact resistance of Hoya Phoenix spectacle lenses Clin Exp Optom 2006; 89: 2: 86–89 B Ralph Chou*† MSc OD FAAO Jeffery K Hovis* OD PhD FAAO * School of Optometry, University of Waterloo, Waterloo, Ontario, Canada † School of Optometry and Vision Science, The University of New South Wales, Sydney, Australia E-mail:
[email protected]
Submitted: 26 April 2005 Revised: 5 August 2005 Accepted for publication: 20 August 2005
DOI:10.1111/j.1444-0938.2006.00013.x Purpose: To investigate how the impact resistance of Hoya Phoenix spectacle lenses is affected by centre thickness and the application of a multilayer antireflection (MAR) coating. Methods: Four groups of plano lenses were tested: dress thickness with scratch resistant (SR) coating on both surfaces, dress thickness with SR and MAR, industrial thickness with SR and industrial thickness with SR and MAR. Lenses were edged to a clear circular aperture of 50 mm with a 0.5 mm hidden bevel and mounted in a specially-designed lens support. A pneumatic gun was used to propel a 6.35 mm steel ball at the centre of each lens. Impact speed was varied using the ZEST protocol to determine the threshold breakage speed. Results: The threshold breakage speeds of the dress and industrial thickness SR lenses were 55.1 and 63.2 m/s, respectively and the corresponding threshold breakage speeds for SR-MAR lenses were 50.1 and 54.7 m/s. All comparisons were statistically significant using Student’s t-test with a rejection level of p < 0.005. Unlike polycarbonate lenses, dress thickness Phoenix lenses do not display ‘oilcanning’ deformation on high energy impact and therefore are less likely to be dislodged from their mountings. Conclusions: We found that the mean impact resistance of the Phoenix lenses was greater than the level required of eye protector lenses by the standards AS/NZS 1337:1992, ANSI Z87.1-2003 and CSA Z94.3-02. Similar to CR39 and polycarbonate, the application of MAR to Phoenix lenses reduces their impact resistance, however, they provide an acceptable level of impact protection in industrial settings, where there is little danger of exposure to pointed or sharp-edged high-speed missiles.
Key words: coatings, impact resistance, Phoenix, spectacle lenses, Trivex
We have reported previously that the combination scratch resistant (SR) antireflection (AR) coatings on CR39 and polycarbonate lenses can significantly reduce the resistance of both materials to impact by small high-speed projectiles.1,2 Recently, Trivex polymer lenses have been introduced with the claim that they have impact resistance comparable to polycarbonate.3 Clinical and Experimental Optometry 89.2 March 2006
86
Trivex is a material developed by PPG Industries.4 It has an index of refraction of 1.53 at 546 nm and an Abbé number of 43 or 45, depending on the lens manufacturer. Its specific gravity of 1.11 is the lowest value of the ophthalmic plastics on the market. The Bayer abrasion index is 0.4, compared to 1.0 for CR39 and 0.2 for polycarbonate. Trivex lenses of 1.0 mm
thickness pass the FDA drop-ball impact resistance requirement. Industrial lenses are claimed to meet the ANSI Z87.1-20035 requirements for impact resistance.4 Hoya Vision Care has modified the Trivex polymer to achieve the desired Abbé number and in North America the lenses are marketed under the brand name Phoenix. © 2006 The Authors
Journal compilation © 2006 Optometrists Association Australia
Phoenix lens coatings Chou and Hovis
In this study, we report on the effect of SR and AR-SR coatings on the impact resistance of Phoenix lenses of dress and industrial thickness. MATERIALS AND METHODS
Test lenses Four groups of plano Phoenix lenses were examined in this study. All lenses were processed to plano power from Hoya semifinished blanks at the Hoya Vision Care facility in Lewisville, TX, and provided as uncut finished lenses. The specific characteristics of each group are summarised in Table 1. Lenses with SR only were processed from semi-finished blanks with a factory applied front hard coat. The ocular side was given a UV cured spin SR coat after surfacing (B Harris, Hoya Vision Care North America, personal communication). Lenses with SR and multiple AR (MAR) coatings were processed from uncoated semi-finished blanks. After surfacing, a thermally cured SR dip coat was applied to both surfaces before the proprietary MAR coating (HiVision AR) was deposited.
Procedures Each lens was edged to a diameter of 50 mm with a hidden bevel using a Weco Edge 450 automatic edger to meet the spec-
ifications for testing of prescription safety lenses in section 11.2.2.4.2 of the CSA Z94.3-02 standard.6 The lens was visually inspected for any surface or edge flaws and its thickness confirmed with lens callipers before being mounted in a specially constructed lens holder. The lens holder was designed according to the requirements of CSA Z94.3-02 for testing lens material impact resistance. The peak of the hidden bevel fits into the groove formed between the base of the lens holder and the screw-on cap, which acts as the lens retainer. Based on the reports that the Phoenix lens had an impact resistance comparable to polycarbonate, we carried out preliminary testing using a high-speed needle. These results indicated that impact resistance was closer to that of CR39 rather than polycarbonate, so that it was possible to assess the impact resistance with the 6.35 mm steel ball. The steel ball is the missile specified in the AS/NZS 1337:19927 medium impact test and ANSI Z87.1 high velocity impact test as well as the CSA Z94.3 ballistic test for impact resistance. Missiles were fired at the centre of the lens by an air gun system described elsewhere.1 Speed was set by varying the air pressure applied to the system. The initial test speed for the missile was selected randomly from values ranging from 45 m/s to 70 m/s. After this impact, a new lens was selected and a new impact speed was set using the ZEST protocol derived by
Parameter
Lens group 2 mm SR 2 mm SR + MAR 3 mm SR 3 mm SR + MAR
Axial thickness* (mm) 1.92 ± 0.05 2.00 ± 0.03 2.89 ± 0.05 2.94 ± 0.03
* mean ± 1 standard deviation
Table 1. Test lens groups
Range Step size Initial probability density function Peak velocity Decay constant Psychometric function Slope False positives False negatives
Value 5 log units 0.025 log units Hyperbolic secant
RESULTS 2.3 log units 2 Logistic 20 0.01 0.01
Table 2. Parameters used in the Zest program
© 2006 The Authors Journal compilation © 2006 Optometrists Association Australia
King-Smith, Johnson and Good8 and King-Smith and colleagues.9 The ZEST computer program is based on the ascending and descending staircase psychophysical method for determining thresholds. After the first impact, the subsequent impact velocity for a fresh lens was determined based on whether a ‘break’ of the lens occurred. If the lens broke, the program selected a lower impact speed and if the lens did not break, a higher impact speed was selected. The actual speed as measured by the timing gates and the result of the impact is recorded and used in the calculations. This allows the mean and standard deviation of the impact speed to converge rapidly to the threshold breakage speed after approximately 20 lenses.8 The parameters used by the program in determining these values are shown in Table 2. Standard deviations calculated from the final probability distribution function are influenced by the slope of the ‘psychometric’ function assumed by the program.9 This means that for a fixed number of lenses, the standard deviation is influenced by the slope of the function, however, we controlled for this by ending the evaluation based on a fixed variability criterion rather than a fixed number of lenses. The process was terminated when the standard deviation fell below six per cent. Although this procedure usually produces unequal sample sizes, it happened that sample sizes were all equal to 18 when this criterion was met. At this point, the changes in missile speed called for by the ZEST program were smaller than the variability in speed produced by the gun at a given system pressure. The criterion for lens failure was full thickness cracking or breakage of the lens on impact.
Figure 1 shows the threshold breakage speeds for the different groups of Phoenix lenses tested. The groups are labelled according to their nominal centre thickness followed by the types of coatings on the lenses. Except for the 2 mm SR and 3 mm SR-MAR lenses, all other comparisons between test groups of the mean Clinical and Experimental Optometry 89.2 March 2006
87
Phoenix lens coatings Chou and Hovis
Threshold breakage speed (m/s)
80 Uncoated CR-39 60
40
20
0 2 mm SR
2 mm SR MAR
3 mm SR
3 mm SR MAR
Lens type Figure 1. Mean threshold breakage speed by lens group. The error bars represent one standard deviation above and below the mean value. The horizontal line is the mean threshold breakage speed for uncoated 3 mm CR39.1
Lens group
Mean impact speed (m/s)
Impact energy (J)
63.24 ± 3.71 54.74 ± 3.20 55.10 ± 3.28 50.10 ± 2.89 n/a n /a 40.0 46.5
2.03 1.52 1.55 1.28 0.20 0.80 0.82 1.10
3 mm SR coated 3 mm MAR/SR coated 2 mm SR coated 2 mm MAR/SR coated 5/8-inch drop ball (FDA) 1-inch drop ball (ANSI Z87.1) Medium energy ballistic test (AS/NZS 1337) Ballistic test (CSA Z94.3-02, ANSI Z87.1)
Table 3. Results by lens thickness and coating
threshold speeds were statistically significant by t-test with an adjusted rejection level of p < 0.005. Table 3 compares the threshold breakage speed and threshold impact energy with the energy levels associated with the ANSI drop ball, the Australian medium energy impact and CSA ballistic impact requirements for industrial spectacle lenses. DISCUSSION The average impact speed for failure of the Phoenix lenses ranged from 50 to 62 m/s, Clinical and Experimental Optometry 89.2 March 2006
88
which indicates that its impact resistance was closer to that of 3 mm uncoated CR39 than the 180 to 270 m/s range often quoted for polycarbonate.1,2 We confirmed qualitatively that polycarbonate lenses have a much higher impact resistance in that we were unable to break any 2 mm polycarbonate lenses with blunt missiles at the maximum speed of the apparatus, which was approximately 100 m/s.2 Another characteristic that Trivex shares with the more rigid CR39 was that it did not deform on high energy blunt impact like polycarbonate. None of the Phoenix lenses that we
tested displayed the phenomenon of ‘oilcanning’ (flexure of the lens on impact that leads to dislodgement from its mounting) that is commonly seen after high-speed blunt impacts on thin polycarbonate lenses.2 Phoenix lenses often showed a radial pattern of internal cracks centred on the impact point when struck with blunt highspeed missiles. Six of the 2 mm SR lenses broke into pieces at impact speeds between 55 and 60 m/s but no other lenses in the test sample were shattered at these speeds. The 3 mm SR lenses cracked at speeds exceeding 62 m/s but did not break into fragments. Thus, even though the impact resistance of 3 mm SR lens is marginally less than the 3 mm uncoated CR39, the Phoenix lens would be a better choice than CR39 as the uncoated 3 mm CR39 shatters at a mean breakage speed of 63.97 m/s for the 6.35 mm steel ball.1 Nevertheless, the 3 mm SR Phoenix lens impact resistance is poorer than that of 2 mm SR polycarbonate, which breaks at 180 m/s.10 As with other resin lenses,1,2,10,11 the presence of MAR coatings decreases the impact resistance of Phoenix lenses. The reason for the decrease in impact resistance with the combination coatings may be an increase in the surface tension on both sides of the lens. Stephens and Davis12 speculated that an anti-scratch coating on the back surface puts the rear surface in tension, which probably decreases the impact resistance. Antireflection coatings that are applied to both lens surfaces over a base anti-scratch coating can be expected to increase the level of surface tension on both surfaces, further weakening the impact resistance of the lens to blunt impact. At a nominal thickness of 2 mm, lenses are weakened by 17.3 per cent in terms of impact energy, while 3 mm lenses show a decrease in impact resistance of 25.1 per cent. Nevertheless, the 3 mm SRMAR has an impact resistance that is essentially identical to that of the 2 mm SR lenses, suggesting that a decrease in impact resistance due to the addition of the MAR can be offset by a reasonable increase in centre thickness. The decrease in the impact resistance of the 3 mm lens with the two types of coatings reinforces © 2006 The Authors
Journal compilation © 2006 Optometrists Association Australia
Phoenix lens coatings Chou and Hovis
our view that the cosmetic and visual advantages of antireflection coatings are offset by a decrease in impact protection for a Phoenix lens of given thickness. Our findings are limited to surfaced Phoenix lenses coated as described. It is uncertain whether unmodified Trivex material has the same impact resistance characteristics. There may be slight differences in the performance of virgin Trivex compared to Phoenix. We anticipate that different coating materials and procedures have a larger effect on the impact resistance of finished lenses. This hypothesis is based on previous results with CR39 that showed considerable variability in impact speeds with different SR coatings, that is, dip versus spray versus spin coat.1 We anticipate that the Phoenix lenses will show a similar variability with different SR coatings. Our results strongly suggest that coated surfaced Phoenix lenses are an alternative to polycarbonate when the risk is exposure to missiles of low to moderate impact energy.
CONCLUSIONS Our data confirm that multiple antireflection coatings significantly reduce the impact resistance of Phoenix lenses at both dress and industrial thicknesses. Phoenix spectacle lenses are a better alternative to CR39, however, multiple antireflection coated lenses should not be used in eye protectors for industry or sports, particularly at 2 mm centre thickness, in situations where there is a high risk of exposure to high energy impacts. If antireflection coated lenses are preferred by the patient, the best option remains polycarbonate despite its poorer Abbé number. ACKNOWLEDGEMENTS
REFERENCES 1. Chou BR, Hovis JK. Durability of coated CR39 industrial lenses. Optom Vis Sci 2003; 80: 703–707. 2. Chou BR, Gupta A, Hovis JK. Effect of multiple antireflective coatings and center thickness on resistance of polycarbonate spectacle lenses to penetration by pointed missiles. Optom Vis Sci. 2005; 82: 964–969. 3. http://www.eyeglasslenses.com/content/ knowledgebase/kb_trilogyQnA.asp accessed 22 March 2005. 4. http://corporate.ppg.com/PPG/ opticalprod/en/monomers/products/ Trivex.htm accessed 22 March 2005. 5. American National Standards Institute. ANSI Z87.1-2003 Standard Practice for Occupational and Educational Eye and Face Protection. New York: ANSI; 2003. 6. Canadian Standards Association (2002). CSA Z94.3-02 Industrial Eye and Face Protectors. Toronto: CSA; 2002. 7. Standards Australia. AS/NZS 1337:1992. Eye protectors for industrial applications. Sydney: Standards Australia; 1992. 8. King-Smith PE, Johnson YM, Good GW. Lens fracture testing using the ZEST threshold method. Optom Vis Sci 1993 70(Suppl): 152–153. 9. King-Smith PE, Grigsby SS, Vingrys AJ, Benes SC, Supowit A. Efficient and unbiased modifications of the QUEST threshold method: theory, simulations, experimental evaluation and practical implementation. Vision Res 1994; 34: 885–912. 10. Davis JK. Perspectives on impact resistance and polycarbonate lenses. Int Ophthalmol Clin 1998; 28: 215–218. 11. Lamarre DA. Development of Criteria and Test Methods for Eye and Face Protective Devices. Cincinnati: NIOSH; 1977. 12. Stephens GL, Davis JK. Spectacle lenses. In: Tasman W, Jaeger EA, eds. Duane’s Clinical Ophthalmology. Revised ed. 1993. Philadelphia: JB Lippincott; 1993. p 45.
Corresponding author: Dr B Ralph Chou School of Optometry University of Waterloo Waterloo, Ontario N2L 3G1 CANADA E-mail:
[email protected]
This study was supported by the Canadian Optometric Education Trust Fund. We thank Sam Odom and David Pietrobon of Hoya Vision Care, North America, for supplying the lenses used in this study. The authors have no proprietary or commercial interest in Hoya Vision Care, North America. © 2006 The Authors Journal compilation © 2006 Optometrists Association Australia
Clinical and Experimental Optometry 89.2 March 2006
89
