ION EXCHANGE TEMPERING OF GLASS OPHTHALMIC LENSES

DONALD P. RENALDO, M.D. Philadelphia, Pennsylvania

AND A R T H U R H . K E E N E Y , M.D., AND H . L Y L E DUERSON, J R .

Louisville, Kentucky

Drop-ball tests of assembled lenses and frames on an anthropomorphic dummy head have demonstrated far greater impact resis­tance than tests of isolated lenses on a test block.1 Conventional heat-treated crown glass of semi-ophthalmic thickness (minimum thickness, 1.8 to 2.6 mm) in four dioptric powers mounted in plastic frames with pos­terior retaining lips (buttress groove) with­stood considerably greater impact than simi­lar lenses mounted in either conventional cel­lulose acetate frames without posterior re­taining lips, or in wire-rim frames of cur­rent vogue.1

Other drop-ball tests with isolated, new lenses, placed on the American National Stan­dards Institute (ANSI) Z80 test block, showed heat-treated and air-quenched oph­thalmic crown glass to be consistently strong­er than untempered ophthalmic crown glass, when critical quality controls of surfacing, edging, heating, and quenching were main­tained.

Experimentation on isolated new lenses on the ANSI Z80 test block showed even greater strength in glass lenses, chemically tempered by ion exchange, than conventional heating and quenching. These results, compared with an earlier study,2 verify the greater protec­tive quality of this type of lens measured by drop-ball impacts.

From Wills Eye Hospital and Research Insti­tute, Philadelphia, Pennsylvania (Dr. Renaldo), and University of Louisville School of Medicine, Louisville, Kentucky (Dr. Keeney and Mr. Duer­son). This study was supported in part by the Na­tional Society for the Prevention of Blindness, and Research to Prevent Blindness, Inc.

Reprint requests to Arthur H. Keeney, M.D., University of Louisville Health Science Center, School of Medicine, Louisville, KY 40201.

MATERIAL AND METHODS We used 80 crown glass lenses of semi-

ophthalmic thickness, treated by ion ex­change. Groups of 20 lenses each in four dif­ferent dioptric powers ( + 2.50 spherical, — 2.50 spherical, — 2.50 cylindrical, and piano) were used. Prior experiments estab­lished baseline figures for these strengths which may be used in comparison with pres­ent results. All lenses were S-V-7 configura­tion (48 mm X 41 mm) and were heat treated in potassium nitrate baths reportedly at 470°C for 16 hours. Each lens was checked for (1) optical strength, (2) minimum and maximum thickness to ±0 .1 mm, (3) visible surface flaws or impregnated concretions as deter­mined under 3X magnification in ultraviolet light, and (4) lack of birefringence under polarized white light. All lenses were placed, convex side up, on a J^-inch neoprene gasket on a standard ANSI Z80 test block. Drop-ball testing apparatus consisted of an electro­magnetic chuck release, adjustable to a height of 122 inches. Chromium steel balls, %, %, and 1 inch in diameter, were delivered pre­cisely within a ^-inch diameter circle lo­cated in the geometric center of each lens. The initial drop (No. 1) was at the Food and Drug Administration (FDA) minimum performance requirement of 50 inches with a 5i-inch diameter steel ball, delivering 0.152 ft-lb of energy. Successive drops were made at increments of 6 inches. If failure did not occur with the «HS-inch ball, then we used a %- and subsequently a 1-inch ball through drop No. 39 when a maximum of 122 inches with the largest steel ball delivered an impact of 1.493 ft-lb of energy. If breakage oc­curred, we noted the form of breakage and

2Q1

292 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST, 1975

TABLE 1 DROP-BALL TEST RESULTS ON GLASS OPHTHALMIC LENSES TEMPERED BY ION EXCHANGE

+2.50 Spherical

-2.50 Spherical

-2.50 Cylindrical

No. of lenses tested

Optical tolerance, diopters

Minimum thickness, mm

Maximum thickness, mm

Average thickness, mm

Minimum break energy in ft-lbs and (drop No.)

Maximum break energy in ft-lbs and (drop No.)

Average break energy in ft-lbs and (drop No.)

No. of no failures

20

0.17 to 0.25 more positive

than indicated optical strength

1.50

3.0

2.25

1.346 (37)

1.493+(39+)

1.493+ (39+)

19

20

0.17 to 0.25 more negative

than indicated optical strength

2.1-2.3

3.8

2.84

1.199 (35)

1.493+(39+)

1.493+(39+)

18

20

0 to 0.06

2.2

3.4

2.5

0.979(32)

1.493(39)

1.052(33)

1

discarded the lens. Breakage was defined as follows:

A lens is considered to have fractured if it cracks through its entire thickness, including a laminar layer, if any, and across a complete diameter into two or more separate pieces, or if any lens material visible to the naked eye becomes detached from the ocular surface.'

A total of 2,707 drops at SO to 122 inches were made.

RESULTS

On pretest visual inspection, the + 2.50 and — 2.50 sphere lenses frequently showed small surface scratches and edge flaws. The — 2.50 cylindrical and piano lenses were free of such visible defects. All 20 of the + 2.50 spherical lenses exceeded the indicated opti­cal power by 0.17 to 0.25 diopters and 17 of 20 were marred by surface flaws, scratches, pits, rainbow shadows, and edge chips visible to the naked eye without magnification. All 20 of the — 2.50 spherical lenses exceeded the specified optical power by 0.17 to 0.25 diopters and exhibited surface flawing simi­lar to the + 2.50 spherical lenses. Flaws were

not of uniform type or location as a specific element in processing would suggest. The re­maining two lots of 20 lenses each were of indicated optical strength (within 0.06 diop­ters) and without visible flaws when exam­ined under 3X magnification in white and ultraviolet light. None of the 80 lenses dis­played any type of birefringence when viewed between two sheets of Polaroid film against a conventional incandescent household light. Despite suboptimal surface appearance, the + 2.50 and — 2.50 spherical lenses proved to be almost indestructable against our maxi­mum impact (drop No. 39) of 1.493 ft-lb. The — 2.50 cylindrical lenses, although less impact resistant than the + 2.50 spherical lenses, but as similarly well surfaced as the piano lenses, faired comparatively well against the 1-inch diameter steel ball (average break energy, 1.052 ft-lb; average drop, No. 33). We compiled the minimum, average, and maximum energy for breaks (Table 1). Vari­ations in minimum thickness were less than 0.1 mm in the — 2.50 cylinders and the + 2.50 spheres; the — 2.50 spheres showed

VOL. 80, NO. 2 TEMPERING OF LENSES 293

minimum thickness variations within 0.2 mm. The 20 piano lenses (Table 2) presented

greater spread in values of minimum thick­ness, from 1.8 to 2.3 mm (ten lenses were precisely at 2.2 mm, and nine ranged below). Maximum thickness varied between 2.1 and 2.4 mm. Average thickness was 2.17 mm. This was consistent with sampling of indus­trial production of piano lenses but exceeded the spread of thickness values among the three prescription groups ground to refrac­tive powers of 2.5 diopters. These lenses were free of visibly detectable flaws and showed a variation in refractive power of less than 0.06 diopters. The piano lenses could only infrequently withstand impact from the %-inch diameter steel ball (average break en­ergy, 0.814 ft-lb; average drop, No. 23).

Pilot studies done from 1962 to 1964 with other ophthalmic lenses (Chemcor) (Table 3) were limited to a %-inch diameter steel ball at drop distances of 36 to 78 inches or a maximum energy impact of 0.6 ft-lb. Thirty-five piano lenses were subjected to 271 drops on a standard Z80 test block. Only two lenses functionally failed under these condi­tions, and they were at drop No. 7 (72 inches) and drop No. 8 (78 inches) after edge chips on drop No. 6 (66 inches) and drop No. 4 (54 inches), respectively. Edge chips, cracks, or fractures were produced in

TABLE 3 CHEMCOR PLANO LENSES

Base curve No. of lenses tested Minimum thickness Maximum thickness Minimum break energy

in ft-lbs and (drop No.) Maximum break energy

in ft-lbs Average break energy in

ft-lbs and (drop No.) Edge chips No. of no failures

3/11/62

- 6 9 2.0 2.0

0.43(4)

>0.6

0.58(7+) 3 6

4/23/64

-6 .12 to -6 .37 6 3.4 3.5

0.6(8)

>0.6

>0 .6 (7+) 3 3

1963

-6 .12 3 3.5 3.6

0.43 (4)

>0.6

>0.54(6+) 2 1

12/17/64

-6 .12 8 2.0 + 2.1

0.6(8)

>0.6

> 0 . 6 ( 8 + ) 1 7

12/17/64

-6 .12 9 3.5 3.6

0.57(7)

>0.6

>0 .6 (8+) 2 7

Total

35

11 24

Drop-ball test with 7/8-inch steel ball; maximum test height, 78 inches (approximately 0.6 ft-lb); ambient temperature, 72 to 79°F; relative humidity, 32 to 42%.

TABLE 2 DROP-BALL TEST RESULTS ON PRODUCTION

PLANO GLASS OPHTHALMIC LENSES TEMPERED BY ION EXCHANGE

Piano

No. of lenses tested Optical tolerance, diopters Minimi, n thickness, mm Maximum thickness, mm Average thickness, mm Minimum break energy in ft-lbs

and (drop No.) Maximum break energy in ft-lbs

and (drop No.) Average break energy in ft-lbs

and (drop No.) No. of no failures

20 0 to 0.06 1.8-2.3 2.1-2.4 2.17 0.626(19)

1.420 (38)

0.814 (23)

0

11 lenses at drops No. 4 through 8 (average drop, No. 6 or 7). Although drops were di­rected to the lens center, equipment for those tests gave less accurate control of the impact location when drop distances ex­ceeded 50 inches. All of these 35 lenses, however, showed energy attenuating capacity greatly exceeding the 1972 FDA minimal performance requirements.

DISCUSSION

Because glass is inherently strong in com­pression, a basic method of strengthening glass is by compressing its surface. This

294 AMERICAN JOURNAL OF OPHTHALMOLOGY AUGUST, 1975

principle has been extended to ion exchange methods or "surface stuffing" techniques.*"7

More than 40 U.S. patents have been issued in this field through 1972.*

Ion exchange is basically a slow process that progresses over many hours at a tempera­ture range of 350°C (660°F) to 550°C (920°F). Finished lenses are immersed in a hot, molten salt bath where large monovalent ions such as potassium are exchanged for smaller monovalent ions such as sodium. The ionic radius of sodium is 0.095 nm while that of potassium is 0.133 nm at 0 to 20°C.8 Ex­changing larger for smaller ions results in in­creased compression of the surface of the lens due to the stuffing of larger ions into the smaller spaces that previously housed the smaller ions. Exchangeable ions must be con­geners, or of the same group in the periodic table of elements. For most ophthalmic crown glass, the salt is potassium nitrate (KN0 3 ) of grade NF-11 or better with additives. The depth of the ion exchange increases with the square root of the time the glass is in the molten salt bath (Fick's diffusion law9). After about four hours, a thickness of 25 to 30 u. is achieved; this progresses with some significant increase over 20 to 24 hours, be­yond which little practical increase in resis­tance is reported. The process has been ex­perimentally continued for 150 hours at which time the compression layer may be as much as 250 (x (0.25 mm) thick. As in conventional tempering, the process is carried out after all other work on the lens, except reflectance coating, is completed. Temperatures of the salt bath are considerably below the viscosity range of glass and further below the soften­ing range, so there is no heat distortion, warp­ing, or alteration in the shape of the lens re-

*A U.S. patent issued Jan. 29, 1957, to Hood and associates is considered pivotal. A "prior art" patent was issued to Research Corporation as the Neill Webster Patent, Nov. 16, 1965, in association with Brockway Glass Corp., Pittsburgh Plate Glass Corp., and Corning Glass Works. The trade name of Chemtempered has been adopted by Re­search Corporation, superseding the Corning designation, Chemcor.

gardless of the time cycle. There is essential­ly no stricture of lens shape or thickness on development of the surface layer because the ion stuffing progresses uniformly from the surface toward the interior of the lens at all points, and apparently even into surface flaws. Modified and separate chemical baths must be used for some particular glass formu­lations (Kalichrome, Photosun, and didym-ium). In addition to the surface compression effect, however, both lens geometry and total lens mass (convexity, concavity, and thick­ness) contribute to the overall or total impact resistance of the lens.

Prior to ion exchange development, most glass lenses toughened for impact resistance were electrically heated and chilled in cold air. Such heat tempering is done after sur­facing, edging, and polishing are completed. The lens is heated to the low viscosity range of glass (below the softening range) for 50 to 300 seconds according to its weight, size, average thickness, and color. Both surfaces are then rapidly chilled in blasts of air at room temperature. The surfaces cool first and create a compression layer or "skin" while the inside cools more slowly. As the cooling progresses, the inner mass of glass restrained physically by the rigid outer skin forms radial strain or tension lines. A compression-ten­sion (surface-core) ratio is established. This may be verified and quantitated visually by examining the biref ringent pattern created by the glass when examined in plane polarized light. This is somewhat analogous to a bi­cycle wheel with its spokes establishing a radial and balanced tension system against the rim. This pattern indicates the compression-tension distribution within the lens and the symmetry or asymmetry of the system across the plane of inspection. In contrast, ion ex­change processes establish surface compres­sion without the internal stresses against the surface layer. Therefore, ion exchange lenses, if previously well annealed, charac­teristically show no birefringence when ex­amined under conventional polarized light and have little internal tension or strain.

VOL. 80, NO. 2 TEMPERING OF LENSES 295

Recent batches of ophthalmic crown glass lenses tempered by ion exchange show frac­ture resistance in the range of 0.63 to greater than 1.5 ft-lb on impacting with %- and 1-inch steel balls from drop heights up to 122 inches. These results show that lenses tem­pered by ion exchange exhibit greater en­ergy attenuation by approximately 100% in piano lenses, 140% in + 2.50 spherical lenses, 300% in - 2.50 spherical lenses, and 350% in — 2.50 cylindrical lenses when compared to similar drop-ball tests of lenses of com­parable optical power tempered by conven­tional heat treating and air quenching.

These findings are specifically in terms of new and pristine lenses. Data have not been evolved for lenses in use or damaged by pit­ting or significantly deep scratches and flaws. Preliminary trials suggest that ion exchange treated glass lenses may withstand repeated pitting blows from small or sharply pointed objects even when such pits are easily visible to the naked eye. Pits on the front surface of the lens cause less loss of impact resistance than do pits on the back or eye side. Drop-ball test results may be greatly reduced by these pits.

Our 1971 to 1972 drop-ball studies1 of lens-frame combinations mounted on an anthropo­morphic dummy head indicated that a pos­terior lip behind the eyewire groove retained the lens more firmly and reduced posterior dislocations by about 85%. However, of the 100 lens-frame combinations tested at that time, 61 resulted in fractured lenses, though at relatively high impact levels. With the greater impact resistance of ion exchange lenses, mounting in resilient cellulose acetate or butyrate frames with posterior retaining

lips should yield even greater security to the wearer.

SUMMARY

We performed low velocity drop-ball tests using -Hi-, %-, and 1-inch diameter steel balls on ophthalmic crown glass lenses chemically tempered by the ion exchange process. Four representative dioptric strengths ( + 2.50 spherical, — 2.50 spherical, — 2.50 cylindri­cal, and piano) were studied with the iso­lated lenses mounted, convex side up, on the American National Standards Institute Z80 test block. New ion exchange lenses exhibited a 100 to 350% greater capacity for attenua­tion of energy from low velocity, large size missiles than matched lenses of similar strength prepared by the conventional heat-treating and air-quenching process.

REFERENCES

1. Keeney, A. H., and Renaldo, D. P. : Newly defined factors of lens-frame relationship in the reduction of impact injuries to the eyes. Ear Eye Nose Throat Mon. 52:19, 1973.

2. Keeney, A. H.: Lens Materials in the Preven­tion of Eye Injuries. Springfield, Illinois, Charles C Thomas, 1957.

3. Code of Federal Regulations, Title 21 (F.D.A.) Section 3.84, U.S. National Archives, Office of the Federal Register 37:2503, 1972.

4. Kistler, S. S.: Stresses in glass produced by nonuniform exchange of monovalent ions. J. Am. Ceramic Soc. 45:59, 1962.

5. Olcott, J. S.: Chemical strengthening of glass. Science 140:1189, 1963.

6. Burggraaf, A. J., and Cornelisseu, J.: Strengthening of glass by ion exchange. Physiol. Chem. Glasses 5:123, 1964.

7. : Strengthening of glass by ion ex­change. Physiol. Chem. Glasses 7:169, 1966.

8. Pauling, L.: General Chemistry, 3rd ed. San Francisco, W. H. Freeman and Co., 1970.

9. Stanworth, J. E.: Physical properties of glass. London, Oxford University Press, 1950.