d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1594–1601

avai lab le at www.sc iencedi rec t .com

journa l homepage: www. int l .e lsev ierhea l th .com/ journa ls /dema

Slumping during sculpturing of composite materials

Yu-Chih Chianga,c, Alena Knezevicb, Karl-Heinz Kunzelmannc,∗

a School of Dentistry and Graduate Institute of Clinical Dentistry, College of Medicine,National Taiwan University and National Taiwan University Hospital, Taipei, Taiwanb Department of Endodontics and Restorative Dentistry, School of Dental Medicine, University of Zagreb, Croatiac Department of Restorative Dentistry and Periodontology, Dental School of the Ludwig-Maximilians University,Goethestr. 70, D-80336 Munich, Germany

a r t i c l e i n f o

Article history:

Received 26 February 2008

Received in revised form

15 April 2008

Accepted 18 August 2008

Keywords:

Composite materials

Handling characteristics

Slumping

Sculpturing

Rheologic property

a b s t r a c t

Objectives. This study investigated the slumping characteristics of four composite materials

during sculpturing prior to their polymerization.

Methods. Four different composite materials were used to measure shape deformation due

to slumping. Silicon impressions of the occlusal plane of three different molars were used

as a mould for the composite samples. The surface of the samples was digitized with a

laser scanner (400 slices, lateral resolution: 25 �m). Scans were made after 1–4 min. The

3D data sets were numerically superimposed with matching software and differences were

calculated relative to the baseline measurement.

Results. The amount of surface deformation increases with increasing observation time. The

average coefficient of variation was 0.2. The largest mean amount of slumping was observed

for ELS with tooth mould 1 (150.0 �m), and for Clearfil Majesty with tooth mould 2 (98.3 �m)

and mould 3 (42.8 �m). Miris 2 Dentin and Synergy D6 Enamel were rather similar and seem

to exhibit little deformation. The slump flow of ELS and Clearfil Majesty was up to 400%

higher than the formers. The deformation could be sorted in the following order “mould

3” < “mould 2” < “mould 1” for all materials and all observation time. There was a significant

influence (p < 0.05) of the three factors, time, mould and composite type (ANOVA).

Significance. This specific method provides a reproducible approach for the assessment of

the handling characteristics of composite materials. The results can identify slumping dif-

ferences and assist in collecting information about the feasibility of a material for certain

indications.

emy

the different monomer systems are relatively small when

© 2008 Acad

1. Introduction

Composite materials became available to dentistry in the six-ties of the last century. The physical, mechanical and clinicalbehavior of composite materials depends on their compo-

sition. Their main components are an organic matrix andinorganic fillers. The filler surface is treated usually with asilane to enhance the bond between the fillers and the matrix

∗ Corresponding author: Tel.: +49 89 51609346; fax: +49 89 51609302.E-mail address: karl-heinz@kunzelmann.de (K.-H. Kunzelmann).

0109-5641/$ – see front matter © 2008 Academy of Dental Materials. Pudoi:10.1016/j.dental.2008.08.008

of Dental Materials. Published by Elsevier Ltd. All rights reserved.

[1–3]. The fillers themselves are usually a mixture of differentglasses and size distributions.

The type of resin will affect the viscosity of the resinand the amount of crosslinking. The differences between

they are properly polymerized and will therefore have nogreat influence on the mechanical properties of the material[4].

blished by Elsevier Ltd. All rights reserved.

4 ( 2

mfipi[

iiipamtayaIhiaccsisLddw

cvttt

npdadrb

mrqpmtmaenct

qi

d e n t a l m a t e r i a l s 2

The fillers, however, largely influence the physical andechanical properties of composite material. The inorganic

llers reduce the thermal expansion and curing shrinkage,rovide radioopacity, improve handling characteristics and

nfluence the aesthetic appearance of composite restoration2,3,5].

Composites vary not only in their composition but alson their handling characteristics. Each of these factors willnfluence their clinical performance. They are available in var-ous consistencies and packages. Currently the mechanicalroperties of all major brand composite restorative materi-ls are rather similar. Therefore, the dentist can decide whichaterial to choose based on his personal preferences. Fac-

ors that may be important include handling characteristics,vailability of shades, packing or even price. In the last fewears, there is a tendency to reduce the volumetric shrink-ge of composite materials by optimizing their filler systems.mproved filler loads usually result in higher viscosity. Theigher viscosity and therefore different handling character-

stic result primarily from alterations of the fillers shape, sizend size distribution [4]. Although it seems that the viscosity ofomposite materials increases with increasing inorganic fillerontent, Lee et al. [6] reported that there is no direct relation-hip between those two components, because smaller fillersnfluence the viscosity more then larger fillers. Therefore theize distribution is also rather important. Bayne et al. [7] andee at al. [6] concluded in their studies that there are largeifferences in viscosity and elastic properties not only amongifferent brands but also among composites of the same brandith different delivery modes/batches [8].

Most clinicians have been interested in the handlingharacteristics of the composites because they affect the con-enience of use to adapt to the cavity wall, sculpturability ofhe occlusal morphology and non-adherence to the applica-ion instruments, which in return can influence the treatmentime and quality [9].

Depending on the indication of the composite it might beecessary to keep a persistent shape of already sculpturedarts of occlusal surfaces during the creation of other occlusaletails until final light curing. On the other hand a certainmount of wettability of the tooth surface is necessary andemands a sufficiently reduced viscosity. Also, for anteriorestorations, a little bit of leveling of modeling marks mighte interesting to obtain a nice surface shape [10].

While handling characteristics of uncured compositeaterials are very important for the clinical application, cur-

ently little objective evaluation methods are available touantify and compare these properties [11]. Usually expertanels are involved during the final optimization step for newaterials. Sometimes, then the manufacturer has to adjust

he properties for different countries individually. Thereforeore methods to improve or adjust handling characteristics

re needed [12]. According to the importance of handling prop-rties efforts during the development of composite materialsowadays are focused not only on the enhancement of post-ured physical properties but also on the improvement of

hese rather subjective parameters [13,15].

The aim of this study was to introduce a new method touantify the deformation of uncured composites due to grav-

ty and viscosity during sculpturing. This property is usually

0 0 8 ) 1594–1601 1595

described as slumpiness. This parameter is important duringthe modeling of the final increment to shape the occlusal sur-face according to anatomical requirements. It is desirable thatthe material maintains its intended shape until the surface isfinally cured.

Based on clinical experience we formulated the followingworking hypotheses:

1. There are differences between the materials concerningtheir deformation due to flow after shaping them.

2. In case of any differences between the materials these dif-ferences are related to the height/volume of the modeledshape due to gravity.

3. In case of any differences between the materials these dif-ferences are related to the time between sculpturing andevaluation.

The clinical relevance of the hypotheses 2 and 3 is that adentist could compensate these two effects at least in part bylimiting the thickness of the last increment and by minimizingthe time to cure after sculpturing.

2. Materials and methods

2.1. Experimental design

Four different materials were selected for this study for whichthe development of the method was an essential part ofthe research goal. The composite materials were selectedbased on differences in their filler composition. Miris 2 Dentinand Synergy D6 Enamel are the nano-hybrid composite.ELS is micro-hybrid composite with silanized barium glass.Clearfil Majesty is microcomposite comprising surface treatedalumina microfiller. Details of the composite materials aresummarized in Table 1.

The most clinical relevant approach to evaluate the surfacechange after modeling an occlusal surface was to use nat-ural tooth surfaces. The occlusal surface of three arbitrarilyselected human molars severed as a template to fabricatedsilicon moulds for the experiments. From each molar theocclusal surfaces were cut with a water-cooled slow speed dia-mond saw. “Leco Varicut VC50” (Leco, Kirchheim, Germany).The data sets of the three different tooth moulds were ana-lyzed to compare the dimensions of the composite volumes.The data are listed in Table 2.

The tooth slices were glued to the bottom of a cylindri-cal PMMA mould with a diameter of 2 cm with cyanoacrylateand the surface was replicated with silicon impression mate-rial (Adisil Rose, Siladent, Goslar, Germany), which is used bydental technicians for precision replications in the dental lab-oratory. After hardening these silicon impressions were usedas a mould for the composite resin samples. For each sili-con mould and for each composite material, eight sampleswere made. The standardized procedure of measurement wasillustrated in Fig. 1.

The composite resins were heated to a temperature of 30 ◦Cwhich should approximate oral conditions regarding the vis-cosity of the materials. The temperature in the oral cavity ishigher than room temperature but does not reach the aver-

1596 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1594–1601

Table 1 – Composite materials and composition according to the manufacturers

Code Brand name Composition Shade Batch no./exp.date

Manufacturer

01 Miris 2 Dentin(nano-hybridcomposite)

Inorganic filler: 80 wt% (65 vol%), range ofparticle size: 0.02–2.5 �m, Methacrylate,Barium glass (silanized), Amorphoussilica (hydrophobed)

Shade S2 Lot 0112153 Coltene, Whaledent AG,Altstätten, Switzerland

02 Synergy D6 Enamel(nano-hybridcomposite)

Filler: 80 wt% (65 vol%), average fillerparticle size: 0.6 �m, range of particlesize: 0.02–2.5 �m,Bisphenol-A-diglycidylmetharylate,Bisphenol-A-diethoxymetharylate,Triethyleneglycol dimethacrylate,Urethandimethacrylate, Barium glass(silanized), Amorphous silica(hydrophobed), Prepolymerised filler

Universal Lot 0089441 Coltene, Whaledent AG,Altstätten, Switzerland

03 ELS (micro-hybridcomposite)

Barium glass (silanized): 74%, BisGMA,BisEMA, Silicium dioxide (silanized),Cataysts, Inhibitors, Pigments

A3op Lot 06.2011-16 Saremco AG, Rebstein,Switzerland

04 Clearfil Majesty(microcomposite)

Silanated glass ceramics, Surface treatedalumina microfiller,Bisphenol-A-diglycidylmetharylate(BisGMA),

A3 Lot 00010A Kuraray Medical INC.,Okayama, Japan

Table 2 – Descriptive data about the three different tooth slices, which were used to evaluate the shape changes ofcomposite materials

Tooth mould 1 Tooth mould 2 Tooth mould 3

Total volume [mm3] 444 271 133Basal area [mm2] 109 85 109Max. height [mm] 6.4 4.9 2.2

Mean height [mm]Max. height difference between cusps tip and fissure [mm]

age body temperature. The materials were moisture protectedwith a plastic bag and heated in a water bath.

2.2. Specimen preparation and 3D laser scan

The composite material was carefully filled into the siliconmould with dental hand instruments, small spatula and con-denser, to avoid air inclusions. A slight excess of materials wasapplied. The silicon moulds were pressed onto the surface of

Fig. 1 – Scheme of composite resin sample preparing forslumping measurements.

4.9 3.6 1.32.7 2.5 1.9

microscopic glass slides and the composite materials adheredadhesively to the glass surface after the silicon mould wasremoved.

For the materials Miris 2 Dentin and Synergy D6 Enamelthe samples could be removed easily from the silicon mold.The materials ELS and Clearfil Majesty sometimes adheredto the silicon mold and deformed during demolding. Sam-ples which were visually perceivable deformed were excludedfrom the experiment. Deformations which were not visiblewere accepted. Evaluation is not influenced by these varia-tions because the 3D results consider only relative changesbetween the different time intervals.

Throughout the preparation phase, the glass slides andthe composite materials together with the silicon mould wereplaced on a heating plate at 30 ◦C (Präzitherm PZ 34, Bachofer,Reutlingen, Germany). The surface was digitized three dimen-sionally with an optical sensor.

To maintain the temperature during the measurementand storage phase the sample stage of the laser sensor wasequipped with a thick metal block (width × depth × height:9 cm × 3 cm × 1 cm) which was also heated to 30 ◦C before eachmeasurement. The glass slides were fixed on top of this metalblock and remained there throughout the whole evaluation

phase.

Before the baseline measurement was started the surfaceof the composite replications of the tooth slices was coveredwith a thin layer of titanium dioxide powder (CEREC Propel-

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1594–1601 1597

Table 3 – Results of slumping measurements for each tested composite and tooth mould from image registration during1–4 min after modeling the occlusal layer

Composite Tooth mould 1 mean (S.D.) Tooth mould 2 mean (S.D.) Tooth mould 3 mean (S.D.)

Slumping [�m] after 2 mina

Miris 2 Dentin 18.9 (4.8) 17.2 (3.3) 15.0 (2.4)Synergy D6 Enamel 28.8 (2.6) 21.2 (1.6) 18.8 (2.7)ELS 85.1 (34.9) 32.5 (5.4) 24.3 (5.8)Clearfil Majesty 68.8 (22.0) 59.4 (13.2) 28.1 (5.0)

Slumping [�m] after 3 mina

Miris 2 Dentin 21.5 (4.5) 18.0(4.0) 16.1 (2.7)Synergy D6 Enamel 37.8 (5.4) 25.8 (2.2) 19.2 (2.1)ELS 124.9 (51.9) 44.9 (7.1) 30.9 (7.3)Clearfil Majesty 93.5 (35.1) 82.6 (22.4) 37.2 (7.3)

Slumping [�m] after 4 mina

Miris 2 Dentin 22.4 (3.6) 19.2 (3.0) 17.1 (3.3)Synergy D6 Enamel 41.4 (3.5) 29.1 (4.1) 20.3 (2.5)ELS 150.0 (66.4) 53.8 (8.2) 33.3 (6.7)Clearfil Majesty 109.9 (36.1) 98.3 (27.0) 42.8 (8.4)

afternce bmaxim

lnp

m6ipsTptr

gtpomabf

2

Tb4msiattSttA

a The data set after 1 min served as the baseline image. “Slumpingin micrometers. “Slumping after 3 min” corresponds to the differebetween the 4 and 1 min data set. The values correspond with the

ant Spray, Vita, Bad Säckingen, Germany). This coating wasecessary to obtain a diffusively reflective surface which is arerequisite for the optical scanning process.

The laser scanner (Laserscan 3D, Willytec, München, Ger-any) is a triangulation scanner which is equipped with a

50 nm laser diode. The wavelength of the laser and the lowntensity (5 mW) assured that the polymerization of the com-osite materials was not initiated by the laser beam. The lasercanner digitizes a light profile with 512 pixels with one scan.o measure the whole surface of the composite samples 400rofiles were acquired per single 3D scan. The lateral resolu-ion of the scanner is 25 �m in x- and y-direction. The verticalesolution in z-direction is better than 10 �m.

It took 50 s to apply the composite to the microscopiclass slide, mount the slide on the sample stage, positionhe stage below the scan line and enter the measurementarameters. Therefore, the baseline evaluation was finishedne minute after the composite was removed form theould. Three additional 3D scans were made at 2–4 min

fter demoulding. No further measurements were addedecause this would be far beyond a clinically relevant timerame.

.3. Image matching and registration

he slumping data were obtained by image registrationetween observed time 1 and 2 min, 1 and 3 min, and 1 andmin. The 3D data sets were evaluated with the custom madeatching software Match3D (Gloger, München, Germany). The

oftware uses principally the same algorithms as describedn Neugebauer [14]. In general, a rigid image registration ispplied to two successive data sets, which means that onlyranslation and rotation were allowed for the transforma-ion function from baseline to follow-up images. The Least

quare Error of the distance between the two surfaces alonghe surface normal of the baseline image is minimized itera-ively with the Levenberg–Marquardt optimization algorithm.

minimum of 20,000 iterations were performed. The vertical

2 min” represents the difference between the 2 and 1 min data setetween 3 and 1 min, while “Slumping after 4 min” is the difference

um difference between the two surfaces after matching.

difference between the two surfaces is calculated after imageregistration, named “slumping”. Based on a statistical crite-rion pixels with large differences to the original surface areexcluded from the matching process. In our experiment thethreshold for exclusion was set to a difference of 3.5 standarddeviations of all pixels. All surface pixels meeting this require-ment were used for the image registration. For the quantitativeevaluation the composite tooth shape was marked as theregion of interest and the mean difference between the base-line and follow-up image was calculated for all pixels withinthe region of interest.

2.4. Statistical analysis

The statistical analysis was performed with SPSS software14.0. Besides, descriptive statistics differences between meanswere evaluated with univariate multifactorial analysis of vari-ance (ANOVA). The three factors were composite material(composite), tooth shape (mould) and observation time (time).

3. Results

The results of the shape deformation, mean and standarddeviation, were calculated and are summarized in Table 3.The amount of deformation increases with increasing defor-mation time. The average coefficient of variation was 0.2,which means that the method is quite reproducible. How-ever, the coefficient of variation for the materials ELS andClearfil Majesty was higher in mould 1 with a maximum of0.44 (ELS, mould 1, after 4 min). Fig. 2 shows as an example of3D scanned profiles at baseline and after 4 min slumping ofClearfil Majesty, and an additional curve represents the pro-file plot of the difference image after matching. Comparing

the difference profile with the surface profiles reveals that inthe fissure we can observe a material gain (positive values),while the cups exhibit a material loss (negative values) due toslumping.

1598 d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1594–1601

Fig. 2 – Example profile plots along one single line of the 3Dscanned surface of Clearfil Majesty.

The three factorial analysis of variance showed that therewas a highly significant influence of all three factors “time”,“mould” and “composite” (all p < 0.001). The 2-way interac-tions between time and mould, time and composite, mouldand composite were also significant (p < 0.05). There was nosignificant influence of the 3-way interactions in these factors.

The significant influence of all three factors makes it morecomplicated to evaluate the results.

3.1. Time effect

Analyzing Table 3 in detail reveals increasing deformation foreach material within one tooth mould for each measuringinterval. Fig. 3 shows as an example of the amount of deforma-tion for each observation time of tooth mould 1. The data fortooth moulds 2 and 3 result in comparable curves. The over-all information is visualized in Fig. 4. Based on Fig. 4 one cansee that most of the deformation happens in the first minuteafter starting the measurement. For the materials which havemore deformation, ELS and Clearfil Majesty, more than 50%deformation occurs in the first time interval. The materials,which exhibit less deformation, Miris 2 Dentin and Synergy D6Enamel, experience even between 70 and 85% of deformationwithin the first time interval.

3.2. Shape effect

Tooth mould 1 had the most exaggerated deformation of thedifferent tooth shapes. In general the deformation can besorted in the following order “tooth mould 3” < “tooth mould2” < “tooth mould 1” for all materials and also all observation

Fig. 3 – Example of deformation due to flow for the testedcomposite materials over time for tooth sample “mould 1”.

times without a single exception. If we compare this rank orderwith the height differences between the cusp tip of the largestcusp and the fissure (Table 2) then we can identify an associ-ation between the amount of deformation and the increasingheight difference. The association is not linear, however, thedeformation is more pronounced with the higher cusp–fissuredifference of mould 1.

3.3. Composite type effect

The materials differ considerably concerning their deforma-tion. Miris 2 Dentin had the least dimensional changes inall time and tooth shape combinations. The deformationrank order was Miris 2 Dentin < Synergy D6 Enamel < ClearfilMajesty < ELS for the first time interval. The differencebetween the least and the maximum deforming material wasnearly 400%. It is interesting to note that ELS and ClearfilMajesty change their rank order in the second and third timeinterval for tooth moulds 2 and 3 but not for mould 1.

4. Discussion

Assuming comparable mechanical data, properties, which canbe summarized as “handling characteristics”, can influencethe decision of a dentist to use a certain material for a givenindication. One example of such a property is the deformationof a material due to flow during sculpturing the anatomy of therestoration [6].

No standard method is available to determine the shapechange during manipulation of the restoration. Standard rhe-ologic measurements, like for example dynamic mechanicalanalysis, use sample shapes and dimensions which are dif-ferent to occlusal geometries [9,15]. During the developmentof the method we discussed several alternatives. Initially wefavored idealized shapes which have some aspects in commonwith occlusal geometries. We considered, for example, flat

composite samples with a cut to simulate a fissure. Anothersuggestion was to use half spheres to simulate cusps and tocombine for example 3 or 4 half spheres to mimic an occlusalsurface. We also discussed to use a cone or cylinder with know

d e n t a l m a t e r i a l s 2 4 ( 2 0 0 8 ) 1594–1601 1599

Fig. 4 – Surface deformation of each composite material in each time interval was calculated (n = 8). Bar represent standarderror. Letter “b” denoted the slumping change between 1 and 2 min; “c”: between 2 and 3 min; “d”: between 3 and 4 min.V e in

decdfnt

tTtgc(su

uecdotopw

alues are percentage of slumping change at “c” and “d” tim

imensions which would allow a mathematical approach tovaluate the slump behavior just like it is routinely used inoncrete investigations [16]. However, it was not evident whichimensions would be appropriate for this experiment. There-ore, currently the most clinically relevant option was to useatural tooth surfaces which per se ensured the right propor-ions of cusps and fissures.

In addition to answering our initial working hypotheses,his first experiment also served to find new hypotheses.herefore, we used three different teeth, which were selected

otally randomly. The minimum thickness for the occlusaleometry was set to 2 mm because usually all light-curingomposite materials can be cured up to a thickness of 2 mmTable 2). In addition, two thicker slices were added becauseome manufacturers claim that their materials can be curedp to 5 mm or even more.

A further problem to solve was, which measurement tose for the measurement of the surface changes. We consid-red two alternatives. First we wanted to evaluate the heighthange and profile change as suggested in ASTM C143-90 Stan-ard Test Method for Slump of Hydraulic Cement concrete [16]r by Al-Sharaa and Watts [17]. However, this method seemed

oo limited, however, because only arbitrarily selected pointsr line profiles can be evaluated. It seemed to be more appro-riate to collect information of the surface changes of thehole occlusal area. Therefore, we decided to use an optical

tervals.

3D scanner which has a wide vertical measuring range andhigh vertical resolution at the same time. This scanner sup-plies data with a lateral resolution of 25 �m. But it has, justas all types of height sensors, a limitation in steep verticalareas. In addition, the triangulation angle of the optical setupcaused shadows in steep areas. For this reason areas with asurface inclination of more than 60◦ were eliminated automat-ically from the evaluation. This means that we could measurea detailed surface map to investigate with a high data qualitythe height changes, which are usually called “slump”, but itwas not possible to measure changes of the outer diameter ofthe tooth slices, which would allow to gather information onthe so-called “slump flow” [18].

The third arbitrarily defined condition, which affects theresults, was the temperature at which the experiments wereperformed. We decided to use the mean between room tem-perature and the body temperature because the temperatureat the surface of the teeth is lower than the body temperaturebut higher than the average room temperature. This temper-ature was kept constant through the whole experiment sothat we can compare the materials relative to each other forthis temperature. It might be interesting to investigate a wider

range of temperatures, because the viscosity of the materialschanges with temperature [9].

The 3D evaluation using a rigid registration of two surfacedata sets has the important advantage that the result will be

s 2 4

r

1600 d e n t a l m a t e r i a l

totally independent of the composite volume or the absoluteheight of the measured specimen. During the whole observa-tion period, the composite continuously flows due to gravity.The rigid registration eliminates differences which are causedby the overall height of the tooth itself. However, due to theflow, the surface shape changes too, which makes a rigid regis-tration prone to artifacts. The statistically defined eliminationof points with large deformations helps to match the major-ity of corresponding surface points. Therefore it is possibleto match “similar” surfaces, too. For each surface point thedistance to the closest corresponding point is calculated andafter successful matching, the sum of all these differences isminimal. This is important in order to understand, why thenumbers of our deformation values are so small. Therefore,we cannot say that a deformation of 100 �m means, for exam-ple, the flow of the cusp of 100 �m relative to the fissure. Thecorrect interpretation is that the deformation was 100 �m rel-ative to baseline shape which is fitted to the follow-up data setin order to minimize all calculated differences. These numbersmay be used only to compare the materials slumping behavior.

Based on our results, we find that the materials experienceincreasing deformation with increasing observation time. Inaddition to the confirmation of our working hypothesis, a clin-ically important aspect is that most of the deformation canbe observed at the beginning of the experiment. The prac-tical interpretation is that due to the fact that most of thedeformation appears at the beginning of the application of thematerials, it will be hardly possible to compensate the flow of amaterial by working faster. Another consequence of this mightbe that in future experiments a single observation with a shortobservation period might be enough to evaluate the slumpingbehavior of a material.

The influence of the tooth shape on the amount of defor-mation is also an interesting detail, confirming again ourworking hypothesis. We found an association between theamount of deformation and the height difference between thecusp and fissure. The term “association” was chosen, becauseno correlation was calculated. The reason for this is that theassociation is not linear as one can deduce from Table 3. Itmight be speculated that higher height differences might beassociated with a smaller cusp tip radius. In viscous fluids, thesurface tension tends to shape drops of similar size depend-ing on their viscous properties. Unfortunately, the rheologyof composite materials is very complex and it is not possibleto directly transfer the results of viscous fluids to compos-ite materials. It is plausible, however, that the amount ofdeformation and the deformation rate is influenced by the vis-cous properties of the composite material too. The influenceof the tooth shape will be an important criterion for futureexperiments, where the standardization of the slumpinessevaluation will be in the focus of interest. It will be neces-sary to find one or more geometrically exactly defined shapesfor standardized evaluations which mimic the findings of thecurrent research with natural tooth surfaces as close as possi-ble. Several alternatives might be feasible, like for example acone shape with a defined tip radius and cone angle. The best

combination has to be investigated yet.

The materials exhibit differences in the amount ofdeformation, which would be expected, based on clinicalexperience. Manufacturers can influence the rheologic prop-

( 2 0 0 8 ) 1594–1601

erties of the materials in a wide range as we can see from thedifferent application versions from flowable to condensablecomposite materials [19–21].

A dentist can choose which composite material is suitablefor a certain situation. For example, high viscosity materi-als are suitable for posterior restorations where the occlusalshape has to be sculptured and should maintain its intendedcusp shape and pit–fissure geometry before polymerization.In contrast, on flat surfaces, like direct veneering with com-posites on the labial surface of teeth, these materials willslide over the surface and do not attach themselves by flowto the surface. Here is the clinical benefit for materials likeELS and Clearfil Majesty. However, Miris 2 Dentin and SyneryD6 Enamel seem to be perfectly suited for posterior teeth witha lot of sculpturing of cusps. Therefore it is not intended touse the current data to claim that one material is superior theother materials. Instead, it is intended to collect informationabout the feasibility of a material for certain indications.

In summary:

1. There are significant differences between the four testedcomposite materials concerning their shape deformationafter demoulding. Miris 2 Dentin and Synergy D6 Enamelare rather similar and seem to exhibit little deformationonly. ELS and Clearfil Majesty are in another group. Theslump flow of the later materials is up to 400% higher asthe previous mentioned group.

2. “Time” is also a factor which influences the surface defor-mation of uncured composites. The largest deformationoccurs at the beginning of the demoulding.

3. The proposed slumping measurement has a low overallcoefficient of variation and has the potential to become astandard test method for characterizing the handling prop-erty “deformation due to flow” of dental composites. Theinfluence of the tooth shape, however, makes it inevitableto develop a standard tooth shape for a wider internationalapplication of this test. This geometry should be a clini-cally relevant shape, which has properties, comparable tothe described natural tooth surfaces.

Acknowledgements

The authors would like to thank Mrs. Eva Koebele for techni-cal guidance and extensive help during laboratory work, andDr. Hong-Jiun Chen who works in National Taiwan Universityfor statistical assistance. This study was supported in part bya Grant No. 065-0352851-0410, Ministry of Science, Educationand Sport, Croatia.

e f e r e n c e s

[1] Arksornnukit M, Takahashi H, Nishiyama N. Effects of silanecoupling agent amount on mechanical properties andhydrolytic durability of composite resin after hot water

storage. Dent Mater J 2004;23:31–6.

[2] Hervas-Garcia A, Martinez-Lozano MA, Cabanes-Vila J,Barjau-Escribano A, Fos-Galve P. Composite resins, a reviewof the materials and clinical indications. Med Oral Patol OralCirc Bucal 2006;11:E215–220.

4 ( 2

d e n t a l m a t e r i a l s 2

[3] Karmaker A, Prasad A, Sarkar NK. Characterization ofadsorbed silane on fillers used in dental compositerestoratives and its effect on composite properties. J MaterSci Mater Med 2007;18:1157–62.

[4] Roeters JJ, Shortall AC, Opdam NJ. Can a single compositeresin serve all purposes? Br Dent J 2005;199:114 [p. 73–79;quiz].

[5] Labella R, Lambrechts P, Van Meerbeek B, Vanherle G.Polymerization shrinkage and elasticity of flowablecomposites and filled adhesives. Dent Mater 1999;15:128–37.

[6] Lee IB, Son HH, Um CM. Rheologic properties of flowable,conventional hybrid, and condensable composite resins.Dent Mater 2003;19:298–307.

[7] Bayne SC, Thompson JY, Swift Jr EJ, Stamatiades P, WilkersonM. A characterization of first-generation flowablecomposites. J Am Dent Assoc 1998;129:567–77.

[8] Opdam NJ, Roeters JJ, Peters TC, Burgersdijk RC, Kuijs RH.Consistency of resin composites for posterior use. DentMater 1996;12:350–4.

[9] Lee IB, Cho BH, Son HH, Um CM. Rheologicalcharacterization of composites using a vertical oscillationrheometer. Dent Mater 2007;23:425–32.

[10] Jacobsen PH, Whiting R, Richardson PC. Viscosity of setting

anterior restorative materials. Br Dent J 1977;143:393–6.

[11] Tyas MJ, Jones DW, Rizkalla AS. The evaluation of resincomposite consistency. Dent Mater 1998;14:424–8.

0 0 8 ) 1594–1601 1601

[12] Wilson KS, Antonucci JM. Interphase structure–propertyrelationships in thermoset dimethacrylate nanocomposites.Dent Mater 2006;22:995–1001.

[13] Lee JH, Um CM, Lee IB. Rheological properties of resincomposites according to variations in monomer and fillercomposition. Dent Mater 2006;22:515–26.

[14] Neugebauer PJ. Feinjustierung von Tiefenbildern zurVermessung von kleinen Verformungen. Diplomarbeit:Universität Erlangen-Nürnberg; 1991.

[15] Ferracane JL, Moser JB, Greener EH. Rheology of compositerestoratives. J Dent Res 1981;60:1678–85.

[16] ASTM C1611 A. Standard test method for slump flow ofself-consolidating concrete annual book of ASTM standards;2005.

[17] Al-Sharaa KA, Watts DC. Stickiness prior to setting of somelight cured resin-composites. Dent Mater 2003;19:182–7.

[18] ASTM C143-90 A. Standard test method for slump ofhydraulic cement concrete. Annual book of ASTM standards;2005.

[19] Hussain LA, Dickens SH, Bowen RL. Properties of eightmethacrylated beta-cyclodextrin composite formulations.Dent Mater 2005;21:210–6.

[20] Sun Y, Zhang Z, Wong CP. Study on mono-dispersednano-size silica by surface modification for underfill

applications. J Colloid Interface Sci 2005;292:436–44.

[21] Kim JW, Kim LU, Kim CK, Cho BH, Kim OY. Characteristics ofnovel dental composites containing2.2-bis[4-(2-methoxy-3-methacryloyloxy propoxy) phenyl]propane as a base resin. Biomacromolecules 2006;7:154–60.