Dental Research

Color change in light-cured resins exposed to daylightMichael F. Burrow* / Owen F. Makinson**

The clinical yellowing of anterior composite resin restorations with time is often thecause for their replacement. This work examined in vitro the effect of sunlight andwater separately and together on color changes in the lighter shades of 16 resins. Thenatural radiation (ultraviolet and visible) was greater than would be expected in clini-cal use. and the color changes were generally only slight for most resins, as judged bystandardized photographic recording and visual inspection of transparencies over time.The changes were greater when sunlight and water were combined. The light-cured sur-

face was less affected than the distal surface. The hydrolysis action by water was par-ticularly marked for one resin, which also showed a high release of camphoroquinone-type catalyst. Marked clinical discoloration may be as much or more affected byabsorption of food colorants than by sunlight and water.(Quintessence Int 1991,-22:447-452.)

Ititroduction

Previous studies have indicated that both chemicallyand light-cured composite resins have the propensityto change color over time by either extrinsic or intrin-sic discoloration.'"* Thts occurs despite satisfaction ofcurrent standard specification methods for color sta-bility evolved from those originally designed for den-ture acryhc resins.

In vitro studies using accelerated aging techniqueshave clearly shown that conventional aud tnicrofilledresins change color.'"^ Astnnssen*' studied 15 resins,using an accelerated aging technique, and indicatedthat color change in resins stored in water for 1 monthat ôO^C correlate well with specimens stored for 2months at 37 ̂ C. Brauer^ determined change of colorin resins stored in water or air occurred after exposure

' Tutor in Conservative Dentistry, Department of Dentistry, Uni-versity of Adetaide, GPO Box 498, Adelaide, Soulh Austraiia500t. Australia.

'* Reader in Conservative Dentistry. Department of Dentistry,University of Adelaide.

Adress all correspondence to Dr O. F. Mattinson.

to a xenon lamp. The visible light-activated resinswere found to be more color stable than were thechemically cured resins, and lighter shades showed agreater change.

Clinical studies have also shown that resins discolor.Ameye et al'" reported after 18 months that 98% of455 restorations were acceptable for color. A different6-year follow-up study^ indicated unacceptable colormatching ranging frorn 3.5% to 79.7%, depending onthe resin used, requiring an overall replacement ratioof 5.7% of restorations. An evaluation of ultraviolet-cured restorations over 10 to 12 years deterrnined thatfailures toward the end of the study period were duemostly to color change. This was cited as the majorreason for patients' dissatisfaction with the restora-tion.-

In this study, 16 brands o flight-activated resin wereselected on the basis of acceptibihty under the Inter-national Standards Organization,"'- American Den-tal Association,'-' or AustraUan .standards.''' The prin-ciple recommended use of the selected resins is for therestoration of anterior teeth. For this study it wasdecided to determine the effect of accelerated aging ofthese resins when exposed to daylight, in water or dry,and with diurnal temperature changes. Sunlight as theirradiation source was wanted for Australian condi-

Ouintessence International Voltjme 22, Number 6/1991 447

Dental Research

Tttble 1 Composite resitis tested

Resin

Aurafil

Dutalll VSHerculite XR

LC33

LCIOOO

Liteftl A

LumiforMultifil VS

PoíofilPrisma FilPrisma

Micro-fineSilux Plus

ValuxVisio-DispersVisio-Fil

Shade

Light

A20U enamelU dentinLSUYLGGBDYU

UGYBUA20BIOULYLY

ILUYGDY

SS

in the study

Batch No,

L505165

02283137782124504504504, 507504504504504504102

06862596862105861906862265213A02102175371202870908872

7D177C2P7Y1P8G1R8Y6P7D1P8602270077Qmi

Particle size

Fine

MicrofilledHybrid

Microfilled

Hybrid

Microfiilcd

HybridHybrid

HybridFineMicrofilled

Microfilled

HybridMicrofilledFine

Manufacturer

Johnson Si, JohnsonDental Care CoKulzer, IncKerr/Sybron Corp

Southern DentalIndustries Inc

Southern DentalIndustries IncShofu

Bayer DentalKulzer, Inc

Voco ChemieDentsply InternationalDentsply International

3M Dental Products Div

3M Dental Products DivESPE GmbHESPE GmbH

tions, because the current Xenon lamp test is but asun substitute derived from the early testing of dentureresins for color stability.

Method and materials

Changes in color may be assessed visually or by pho-tometers. Visual inspection is used in the ISO stand-ard, because only a change large enough to be dis-

cerned visually is of interest to the patient or operator;this subjective method was thus used for the main trial.

In addition, a refiection densitotneter {Super SpeedMaster, Model R75C, Electronic Systems Engineer-ing) was used as a sensitive, objective method for in-vestigating color change. This photometric methodachieves a reproducible tneans for determining whenchange in color occurs below visual perception levels.

The 16 light-activated composite resins, shades.

448 Quintessence International Volume 22, Number 6/199"

Dental Research

batch numbers, and manufacturers are listed inTable 1,

Three sets of specimens were made. The tirst groupof 3-mm thickness was subjected to the four environ-ments listed below. In the second group, of the samethickness, half the specimen was exposed to sunlightand water and half was protected from light, A thirdgroup of specimens of 1.2 mtn thickness were exposedto two of the test conditions to evaluate the effect ofspecimen thickness.

For the first group of specimens, to be placed in thefour environments, eight specimens were made of eachshade and resin, that is, two specimens for eaeh en-vironment. These were placed in white acrylic resinmolds (Perspex sheet, ICI Pty Ltd). Each specimenhad a diameter of 6.25 mm and a thickness of 3 mm(Fig I). The mold material was chosen after tests ona range of sheet materials; the white acryhc resin wasbest for visual comparison of photographs used as therecording medium.

A second series of specimens in this group was pre-pared from a range of shades in three resins to deter-mine the effect of shade on color stability.

Before they were cured, the samples of resin Vi'erecovered with polyester matrix strips and pressed flatbetween two glass slabs. Each sample was then curedfor 40 seconds from one surface only using a VisiluxII curing light (3M Denial Products Div). The samplewas cured on a glass slab to prevent surface reflectionfrom enhancing polymerization on the side of thespecimen away from the hght. The specimens in acrylicresin molds were divided into four lest groups for ex-posure to one of the following four environments: sun-light and water, darkness and water (control), sunlightand air, or darkness and air (control).

Those samples stored in tap water had the waterreplaced weekly and filled up when necessary becauseof evaporation. All specimens were placed in exposedconditions that had the effect of thermocycling overthe diurnal range. The period of sunlight exposureranged from approximately 6 hours during winter to10 hours during summer.

The specimens in the second group were of the samesize; two samples were made for each of 14 shades andresins. These were placed in water so that half thesurface was exposed to sunlight and half was coveredwith metal foil to illustrate color changes more easily.

The third group of samples was prepared to deter-mine the effect of thickness on color change. Thesespecimens were plaeed into washers (1.2 mm thick),cured as before, and stored in water (sunlight anddark) only.

Fig 1 Resin specimens in the white acrylic resin sheetwith the standard color chip tor assessing repeatability ofthe color processing.

Each specimen was photographed at basehne andat intervals of 1 week and 1, 3, and 6 months usingEktachrome (ASA64 and later ASAIOO) color trans-parency film (Eastman Kodak). The wet speeimenswere kept damp during photography. A standard colorchip (Color Control Patches, Eastman Kodak) wasincluded to assess constancy in the color processing.

The photographs were compared and each resinsample was rated as showing substantial darkening ofeolor, slight darkening of color, no detectable changein color, slight loss of color, or marked loss of color.

One operator was responsible for rating all photo-graphs, which included a replication test of 40 sam-ples. The percentage of concordance for the replica-tion test was 91%,

Eor the reflection densitometry, readings using theneutral, cyan, magenta, and yellow filters were ob-tained over the ñve recording times. Again, the wetspecimens were kept damp during measurements. Thereadings were taken with a neutral gray backgroundbeneath each specimen. These readings were averagedbetween the two specimens to produce a result for eachbrand and shade of resin tested. The same neutral graybackground was used as a null setting for the densi-tometer.

Results and Discussion

The results for the fir.st group of specimens are givenin Table 2 for the effect of the four environment com-binations of sunlight or darkness and air or water onthe 16 brands of resin. Table 3 provides the same in-

Quintessenoe International Volume 22, Numher 6/1991 449

Dental Research

Table 2 Resin samples showing colorchanges

Table 3 Resin samples of various shades showingcolor changes

Resin

AurafilDurafil VSHerculiteHe re til i teLC33LCI 000Litefil ALumiforMultifil VSPolofilPrisma FilPrisma M-fSilux PlusValuxVisio-DispVisio-FilSubjective colo

Shade

LA20U (e)U(d)UUUUA20ULYLYUUS

s

Dark(dry)

Û

000000000004-000

Dark(wet)

+000

-1--1-

0

n+0

00-1-0

Sunlight(dry)

++--1-

0000I)-1-

+++-00

r change evaluation of tbe resin samplessubstantial darkening of color; (I - ) = sliEht 11;jss of color: Í

+ ) = si ¡ght darkenin

Sunlight(wet)

+ +-H-H----H00

+

0-

++

-F-H

+.. I + +) =g of eolor;

1 = marked loss of color; 101 =

Resin Shade

LC33 LSUYLGGBDY

Litefil A UGYB

Silux Plus ILUYGDY

Dark(dry)

00

000

-I- +

+0000

-1-000ô0

Dark(wet)

++0-1-

+00

+++00000000

Sunlight(dry)

+++-1-0

+++0-00

0_l_

00•

Sunlight(wet)

H--1-——~———0+Û

++0_l_-1-

+1

no detectable change in color. Subjective color change evaluation of tbe rasin samples. (-I--I-) =substantial darkening of color; ( + ) = slight darkening ofcolcr.( —) = slight loss of color; (--) = marked loss of color; (0) = nodetectable change in color.

Fig 2 Aurafil L samples showing color changes: (ieñ) orig-inal state; (right) 3 months.

Fig 3 Discernible color change on an LC33(U) specimetafter drying (equivalent to mouth breathing); (top halt) Btposed to sunlight and in water; (bottom half) kept in thfdark and in water.

450 Quintessence International Volume 32, Nunii)e|. 6/19E

Dental Research

formation for the range of shades in three brands ofresin also frotn the first grotip of specitnens.

Because the main group of samples was limited totwo specitnens for each shade of resin in each of thefour environments, no statistical analyses are mean-ingful.

Sunlight and water alone had about an equal effectin producing a color change (14.1% and 10.9%. re-spectively) when compared to the control of dark anddry (1,6%). These were slight color changes and notreadily noticeable. When the effects of sunlight andwater together were examined, considerably more res-ins were atïected (20,3%), For three (4,7%) of these,the changes were slightly greater even in the first fewweeks (Figs 2 and 3),

The reflection densitometer readings indicated thatany changes in color tended to occur after the first 3months. Two specimens (AuratH and Prismafil) hadgreater changes than did all other resins for the read-ings taken with the neutral density and cyan filters.The reflection densitometer readings with color filterswere not readily reduced to a graphic form and arenot illustrated. Thns, for the brands of resin shown inTable 2 and the shade variations shown in Table 3,these results are nonquantitative and indicate the qual-itative change that an observer might see on closevisual examination; such close scrutiny would prob-ahly not be usual to most patients. The densitometerwas only used to ascertain the time at which the colorchanges started to occur.

The changes in surface reflectance from these resultsappear to be independent of sunlight or water. Thischange in reflectance could be due to chemicalehanges within the resin over time. The results forTable 2 are for samples approximately 3,1 mm thick,ahout the depth of Class III or Class IV restorations.In thin samples (approximately 1,2 mm thick), similarto labial veneers, a color change, when detected, tend-ed to be a change to yellow. In other thin samples,detection of color change was almost impossible be-cause of their translucent nature, which allowed thehacking used during photography to show through.The yellowing effect was dramatically evident on theunderneath surfaces away from the curing light ofsome resin specimens (Fig 4), It was also noted thatthe intensity of the yellowing decreased slightly overthe 6 months. It is believed this is probably due toeamphorquinone. which apparently varies with theamine balance in the setting reaction from batch tobatch of material and between materials, Uncuredsamples left in water for some weeks show a similar

Fig 4 Leaching of yellow camptioroquinone from (rightside) the underside (least-cured side) of the resin samplecompared with (left side) the top surface, which was nextto the curing iight

yellowing and may have an envelope of yellowed so-lution.

For three brands of resins, separate sets of speci-mens were made with the range of shades to see ifhghter or darker shades might be more affected bysunlight and/or water. The results indicated more no-ticeable color changes in one brand of resin (LC33),The results for the dark control specimens withoutsunlight might have been due to some slight moisturecondensation diurnally within the envelope container.For the wet specimens of the same resin, sunlightcaused the opposite color change.

The reflection densitometer readings of differentshades gave similar results. The large number of den-sitometer readings were not readily reduced to agraphic form. Little ehange was noted until the 3-month reeordings, and there was httle variation be-tween hrands or shades. Again, the results appearedto be relatively independent of sunlight or water.

Conclusions

For the resins tested, only slight eolor changes werenoted on some specimens with sunhght and/or waterin a rigorous test of color stabihty. The free cam-phoroquinone component from the undersurfacecould be expected to leach into adjacent tooth sur-faces.

In the harsh sunlight conditions of the Austrahanclimate (north of the 37th parallel), color changes inresin restorations and veneers are a eommon cause of

Quintessence International Volume 22, Number 6/1991 451

Dental Research

replacement. Considering the severity of the test con-ditions, it is surprising that there were so few colorchanges, and these, except in three instances, weremild. The main causes of color change in vivo might,therefore, be of dietary origin.

Acknowledgments

We wish to thank all Ihc niainiractiireis and suppliers for the supplyoFniaterials and their generous cooperation. This project was fundedby the Australian Dental Reseai-cb Fund for which the atitbors aremost grateful.

RefereDces

t. Asinusseii £\ Clinical relevance of physical, ctiemical and bond-ing properties of composite resins. Opet Dent 1985;10:61-73.

2. Craig RG: Restorative Dental Materials. St Louis, CV MosbyCo. t985, pp 237-238.

3. Draughn RA, Bowen RL, Moffa JP: Composite restorative ma-terials, in Reese JA, Valega TM (eds): Restorative Dental Ma-terials: An Ovcrvieiv. vol t. London, Quintessence PublishingCo. t98S, pp 75-tO7.

4. Phillips RW (ed): Spinners Science of Denial Materials, ed 8.Phitadetphia, WB Sautiders Co, 1982, pp 221-226.

5. Stokes AN, Robb NT, Shorter HR. Ultraviolet-light cured cora-posite-resin restorations on fractured incisor teeth: an eva!tia-tion aRei ta-12 years. NZ Dent J 1986;82:t5-16.

6. van Dijken JWV: A clinical evaluation of anterior conventional,microtiller, and hybrid eomposite resin tlllings. A 6 year follow-up study. Acta Odontot Scand ]984;44:3 57-367.

7. Powers JM, Dennison JB. Koran A: Color stability of resto-rative resins under accelerated aging. J Dent Res ]978;57:964-970.

8. Asmussen E: An aeceleratedlest for color stability of restorativeresins. Acta Ondontot Scand t981;39:329-332,

9. Brauer GM : Color ctianges of composites on exposure to var-ious energy sources. Dent Mater 1988;4:55-59.

to. Ameye C, Lambrecbts P, Vanherle G: Conventional and tnicro-filled cotnposite resins. Part 1. Color stability and marginal ad-aptation. J Prosthet Dent 198t;46:623-630.

11. Intemationat Organization for Standardization: Delermlnationof Color Stahitity of Denial Polymeric Materiat.i. tSO 749t-t9a5.Geneva, International Organization for Standardization, 19S5.

12. International Organization for Standardization: Rcsin Ba.fe Fill-ing Materials. ISO 4Ü49-1988. Geneva, Intemational Orgatii-zation for Standardization, 1988.

13. American Dental Association: Spediication No. 12 for denturebase resin. Guide to Dentul Materials, ed 2. Chicago. AmericanDental Association. 1964. pp 117-122.

14. Standards Australia: Rcsin-Based Dentat Restorative Materials.Australian standard 127B-t9B2. Sydney, Standards Australia,t982. D

Moving Soon?

Please tiotify us promptly of any change of address to assure an uninterrupted subscription.

Cotnplete the form below and send to: 0/ Subscription Department, Quintessence Publishing Co.Inc., 551 North Kimberly Drive, Carol Stream, IL 60188

MOVING FROM:

(OLD ADDRESS).NEW ADDRESS.

CITY, STATE, ZIP. (CITY, STATE, ZIP).

NOTE: To expedite delivery, send ttie change ot address as soon as you know it, and allow 6 weeks for processing.

452 Quintessence International Volume 22, Number 6/1991