- Email: [email protected]
Photo-polymerization shrinkage-stress kinetics in resin-composites: methods development
Dental Materials 19 (2003) 1–11 www.elsevier.com/locate/dental
Photo-polymerization shrinkage-stress kinetics in resin-composites: methods development D.C. Wattsa,*, A.S. Marouf a,b, A.M. Al-Hindia a
University of Manchester Dental School, Higher Cambridge Street, Manchester M15 6FH, UK b Tishreen University Dental School, Lattakia, Syria Received 8 January 2001; revised 9 November 2001; accepted 06 December 2001
Abstract Objectives. Studies of free shrinkage-strain kinetics on restoratives have begun to multiply. However, there have been fewer investigations of the more difficult problem of concurrent stress-kinetic measurements. The aim was to outline design parameters for a new methodology for this problem, amenable especially to light-cured materials, and to present illustrative results for a range of restorative composites. Methods. Absolute values of stress measurable for a given material and geometry are dependent upon the stiffness of the measurement system. In an infinitely stiff system, the measured stress would also tend towards infinity. Real teeth and their cavities are not infinitely stiff; they have elastic and visco-elastic compliance. Consequently, it is important that some minimal, but essentially constant compliance be allowed, whatever the final or time-dependent modulus of the material may be. This goal has been realised by measurement of the timedevelopment, for a disk-geometry specimen (f ¼ 10, h < 1.0 mm) of stress (Sr), with a calibrated cantilever beam-geometry load cell. A novel specimen-holder design was used for this purpose, held in a rigid base assembly. Specimen thicknesses (or gap-widths) of 0.8 and 1.2 mm were specifically investigated on four representative resin-composites. Concurrent measurements were made of the enddisplacement of the cantilever load cell, relative to a lower glass plate retaining the specimen. Results. Load-calibration of the cantilever load cell gave an end-displacement per unit stress of circa 6 mm/MPa This compares with literature values for cuspal compliance or displacement of circa 20 mm. Re-normalisation of the stress-data was implemented. This was accomplished assuming Hooke’s law behavior at each instant and equivalent to a stiffer system, with a correction (multiplier) factor of 4 on the raw-stress values. For the materials examined, resultant maximum-stress levels determined were circa 5 – 8 MPa Stress-levels obtained at 1.2 mm thickness were slightly higher (11 – 15%) than the level of stress obtained at 0.8 mm thickness. This is attributable to the greater mass of material undergoing shrinkage at 1.2 mm, offset slightly by the different C-factors. Significance. The new device is a practical and self-contained system for rapid and accurate measurement of stress-kinetics in photopolymerising and also self-cure materials. q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved Keywords: Polymerisation shrinkage; Stress; Composite resin
1. Introduction Molecular densification during the polymerisation process of dental restoratives, and the macroscopic effects of shrinkage-strain and/or shrinkage-stress, continue to attract widespread international research interest [1 –14]. This is pursued at several levels. There is the need to characterize these properties of both candidate monomer molecules, synthesized with the aim of attaining reduced shrinkage materials [15], and also resin-composite materials – either commercially available or formulated experimentally [4]. In * Corresponding author. Tel.: þ 44-161-275-6749. E-mail address: [email protected] (D.C. Watts).
addition, different light-curing units and modes of operation require investigation with representative materials in the context of shrinkage phenomena [13,16,17]. These properties are strongly coupled and are controlled especially by the degree of conversion of the network [13] (Fig. 1). For more than a decade, several techniques have been developed and utilized for measurement of shrinkage-strain. The dilatometer, linometer and ‘bonded-disk’ methods have been widely employed [18 – 21]. However, progress in developing methods of comparable utility for determining shrinkage-stress has been slower. Part of the difficulty resides in the design of the specimen holder, which must become bonded in the process of measurement. It must be efficiently de-bonded to effect a subsequent measurement.
0109-5641/03/$ - see front matter q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved PII: S 0 1 0 9 - 5 6 4 1 ( 0 2 ) 0 0 1 2 3 - 9
2
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
movement of opposing cusps can be of the order of 20 mm [36 –42]. The greater the displacement of the bonded walls, the lower will be the effective shrinkage stress level sustained. While noting the above problems, and design challenges, we are not suggesting that previous designs are either flawed or in any way unsuitable. Nevertheless, we do contend that there is an important place for design innovation in measuring shrinkage-stress. Accordingly, the aims of this research were:
Fig. 1. Inter-relationship of network shrinkage phenomena.
The earliest work in this area appears to be that of Bowen [22,23] and of Hegdahl [6] who recorded development of shrinkage forces with a Universal Testing (Instron) machine (UTM). Major insights and developments have been achieved by the ACTA group of Davidson, Feilzer, de Gee and Alster [5,24 –30]. These workers have used a servo-hydraulic UTM in which sophisticated and careful procedures have been deployed with the aim of largely eliminating machine or system-compliance. Using this approach, firstly with self-cure composites and later with VLC composites, many important phenomena have been identified, including the effect of C-factor on the stress magnitudes. Broadly similar approaches have been reported by Ferracane [31], Bouchlicher [2] and Kunzelmann [32]. Different approaches to shrinkage phenomena have been made using strain-gages [33], photo-elastic and Moire methods [34], as well as Finite Element modeling [46]. There are, however, several practical disadvantages attached to the use of a large servo-hydraulic UTM for measuring shrinkage-stress kinetics. Firstly, the equipment is expensive and complex and may be required for a variety of research projects. If so, it is inconvenient to use this equipment for extensive measurements without intermission over several weeks. Secondly, the generality of this equipment entails construction of specialist attachments and couplings which tend to increase the machine-compliance requiring correction. Thirdly, in attempting to virtually eliminate compliance by servo-control, there is a fundamental limitation. For a measurement device with absolutely zero compliance, the recorded shrinkage-stress should approach an infinite value, as a corollary of Newton’s third law. In practice, all measurement systems for load (or stress) entail finite compliance. Load cells incorporate strain-gages which deform under load [35]. Furthermore, the hard dental tissues of cavity walls also exhibit compliance and relative
1. to develop a self-standing shrinkage-stress instrument, based on a rigid cantilever load-cell, capable of direct calibration, 2. to formulate a transparent protocol for determination of stress corresponding to a standardized and clinically appropriate system-compliance, 3. to evaluate the shrinkagestress kinetics of representative resin-composites under conventional irradiation regimes by Quartz Tungsten Halogen (QTH) light sources. The overall hypotheses to be tested were that the experimental procedure would (a) independently generate stress-magnitudes in broad correspondence to those perceived to be clinically meaningful by previous researchers and (b) prove to be effective and ergonomic to deploy in practice.
2. Materials and methods The Bioman shrinkage-stress instrument was designed and constructed at the University of Manchester (Fig. 2). Components were bolted to a 2 cm thick stainless steel baseplate. This was mounted vertically on a support frame. A cantilever load-cell of 500 kg capacity was fitted with a rigid integral clamp, at the free or compliant end of the cantilever, to securely hold a circular steel rod (10 mm diameter and 22 mm long) vertically and perpendicular to the load-cell axis. The lower end of the rod was sand-blasted to aid composite-retention and this formed one face of the specimen chamber. Across a gap, normally adjusted to either 0.8 or 1.2 mm, the counter-face consisted of a rigid glass plate, 3 mm in thickness. This was removable, before and after measurements, but was held rigidly relative to the base-plate in a special clamp during measurement. The surface of the glass-plate opposing the steel-rod was also sandblasted to aid composite retention. Composite paste was introduced between the plate and vertical rod to form an un-set specimen-disk of 10 mm diameter and, normally, either 0.8 or 1.2 mm thickness. The disc-radius orientation was thus horizontal so that the instrument could be used for fluid composite pastes as well as stiff pastes. These thicknesses correspond to different ratios of bonded to
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
3
Fig. 2. The Bioman shrinkage stress device: (a) load rod; (b) specimen (detached); (c) glass plate; (d) displacement-signal conditioning unit; (e) dataacquisition unit; (f) LVDT transducer; (g) cantilever cell; (h) light source; (i) load-cell amplifier.
unbonded surface areas, that is configuration factors (C-factor, C ¼ d/2h, where d and h are the diameter and thickness of the disk samples, respectively). C ¼ 6.25 for 0.8 mm thickness and C ¼ 4.16 for 1.2 mm thickness. The composite material was irradiated across its’ thickness dimension, rather than radially (Fig. 2). This corresponds to the orientation utilized by Condon and Ferracane [4]. During polymerization, stresses established through the material, and thus between the fixed glass plate and the rod, caused slight displacement of the free end of the loadcell. Along with the load-cell signal, this displacement was continually recorded to computer using an attached (LVDT) displacement transducer and a fast data-acquisition system. The displacement was measured as movement (in mm) between the load-cell end and the fixed glass-plate counter-face. The load-signal from the cantilever cell, which was amplified by a wide-range strain indicator (Model 3800, Vishay, Measurements Group, Rayleigh, NC, USA) was calibrated at periodic intervals, as described later. Prior to photo-polymerization, the system and composite-resin were maintained at a starting temperature of 23 8C. The curved optic of each light source utilized was replaced by a straight version of the same diameter,
typically 8 mm. The end of the optic was positioned at the desired distance, normally 0.5 mm, from the 3 mm thick glass-plate. Thereby full power from the light-source could be delivered to the composite, corresponding to optimal clinical irradiation conditions. Where temperature rises were produced by radiant energy and/or exotherm, these were permitted to occur, as may often obtain clinically. Four representative resin-composites materials were investigated (Table 1). Two groups (A and B) of specimens (n ¼ 3 – 5) of each composite were prepared in situ between the rod and glass plate, described above. Group A were 0.8 mm in thickness and Group B were 1.2 mm thick. For example, to produce the 0.8 mm specimens, typically 0.12 – 0.15 g of material were formed into the disk shape. The precise amount of unset material required for a 10 mm diameter, without excess, was established by trial closures, Table 1 Light-cured hybrid resin-composites investigated Materials
Code
Lot
Shade
Manufacturer
Clearfil AP-X Point4 P60 Z250
APX Point4 P60 Z250
584 910576 9AC 19991018
A3 A3 A3 A3
Kuraray, Osaka, Japan SDS-Kerr, CA, USA 3M, St Paul, MN, USA 3M, St Paul, MN, USA
4
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
in accordance with the following protocol: (i) the glass plate was firmly clamped horizontally; (ii) the steel rod was initially unclamped and using a 0.8 or 1.2 mm feeler gauge, the gap below the rod was set; (iii) then, while preventing any further vertical movement, the rod was clamped above the glass plate; (iv) the glass plate was then unclamped and removed; (v) using a template, a 10 mm diameter circle was drawn on the glass plate to identify the required specimen paste location directly below the rod; (vi) the composite paste was inserted on the glass, within the drawn circle; (vii) the glass plate was then re-positioned, raised and firmly clamped as before; (viii) this caused the paste to flow and adapt to the required specimen height, in the required location; (ix) any slight excess was removed manually, although with precise weighing this was rarely necessary. A period of 2 min was allowed for stabilisation, including any internal stress-relaxation, within each specimen, prior to irradiation. An Elipar II light source (3M-Espe, Seefeld, Germany) was used to irradiate the composites for 40 s photo-cure at 800 mW/cm2, as measured with a ‘Power Max 500A’ radiometer (PM500A, Molectron Detector Inc., Portland, OR, USA). The initial specimen temperature was 23 ^ 1 8C. Exactly 20 s prior to lamp-illumination, the data-recording
system was triggered for concurrent measurement of displacement and load, for the desired time period. This ranged from 10 min to 2 h, but in the present work data are presented for 0 – 30 min. Subsequently, the 20 s preirradiation baseline-data were deleted from file.
3. Device calibration procedures Load-calibration of the device was achieved by detaching the components opposing the cantilever end and, with the base-plate clamped vertically, attaching a series (n ¼ 8) of 5 kg calibration weights, via a rod-plus-retainer, in the axial-load direction (Fig. 3). This gave a strongly linear calibration plot (r , 0.999) of load-cell signal (mV) versus applied-load (N). During load-calibration, concurrent measurements were made of the end-deflection of the cantilever, in mm, for each applied load-increment. This was achieved using a device to clamp an LVDT displacement transducer onto the top of the base-plate with the LVDT-probe resting on the cantilever end, and in line with the axis of loading. When the applied loads were re-expressed as equivalent ‘applied-stress’ values, by dividing by the end-area of the 10 mm diameter
Fig. 3. Load-calibration arrangement, where some components are unbolted to permit application of successive 5 kg masses.
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
5
steel ‘load-rod’ (Fig. 2(a)), a plot of cantilever-displacement versus applied-stress was generated (Fig. 4). This showed a displacement of circa 6 mm/MPa. This means of measuring the cantilever displacement during load-calibration was different in detail, but generally comparable with the concurrent axial-displacement (DZ ) measurements made during routine measurement of shrinkage-stress.
4. Correction of stress data
Fig. 4. Calibration graph of cantilever-displacement versus stress, applied by means of standard weights (each of 5 kg). The load values were reexpressed as stresses equivalent to those acting on a circular specimen disk area of diameter 10 mm.
As noted in the introduction, stresses produced in tooth cavities by polymerising restoratives can result in substantial inward cuspal movement. This is analogous to compliance of many stress-measurement assemblies [35]. Review of the literature [36 – 42] indicated that total cuspal movements produced by the polymerisation of composite resin fillings could be of the order of 20 –25 mm. In order to establish an equal basis for comparison of materials, that
Fig. 5. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of ClearfilAPX, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
Fig. 6. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of Point-4, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
6
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
would have clinical relevance, it was decided to compare the actual displacement of the cantilever end, and thus of the disk specimen heights, with that occurring in tooth cavities. For a typical value of 2 MPa stress recorded as initial (raw) data, the end-displacement was circa 12 mm, as seen in the displacement-calibration plot (Fig. 4). This is the same order of magnitude as some of the dental cuspal deformations. The compliance (reciprocal stiffness) of the instrument is always the same (circa 6 mm/MPa) irrespective of a given disk-thickness of specimen. Nevertheless, we have considered it appropriate to treat the data as corresponding to a higher stiffness load cell (lower compliance), and have chosen to apply a constant multiplying Correction Factor of £ 4. This involves the reasonable assumption of Hooke’s law at each time interval. This then is equivalent to a more conservative (lower) estimate of ‘permitted’ cuspal displacement, equivalent to circa 3 mm. Hence the ‘raw’ shrinkage-stress values have been all multiplied by 4 to give the final shrinkage-stress values that are reported in the following tables.
Thus, if (say) the ‘raw’ stress was 1.5 MPa and the correction factor was 4, the corrected shrinkage-stress would be 6.0 MPa. These straightforward calculations were effected either with transform capabilities incorporated in standard graphical programs (Sigmaplot, www. spssscience.com; FigP, www.biosoft.com) or by use of custom routines written in Matlab (www.mathworks.com).
Fig. 7. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of P60, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
Fig. 8. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of Z250, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
5. Results During each measurement run, the shrinkage forces exerted by the polymerising composite were able to pull the steel rod downwards together with the cantilever end of the rather stiff load cell. This movement occurred relative to the fixed spatial coordinates of the glass plate and lower surface of the composite disk. The displacement (mm) values of the cantilever load-cell for each material with 0.8 and 1.2 mm sample thicknesses are presented in Figs. 5 –8. The mean values of the maximum
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
7
Table 2 Mean values of the maximum displacement (DZ, mm) of the Bioman cantilever beam load-cell, due to the shrinkage-stress of the restorative materials for two different disk-specimen thicknesses Materials
Maximum values of the displacement (mm) of the load beam-cell due to the shrinkage stress of the restorative materials Run #
Displacement (DZ ) (mm) Sample thickness 0.8 mm
1.2 mm
AP-X
1 2 3 Mean SD
7.68 7.66 7.89 7.74 0.13
7.90 7.70 7.50 7.70 0.20
Point-4
1 2 3 Mean SD
10.04 11.16 10.67 10.62 0.56
12.60 12.38 11.80 12.26 0.40
P-60
1 2 3 Mean SD
7.60 7.75 7.73 7.70 0.08
8.56 7.82 8.81 8.40 0.50
Z-250
1 2 3 Mean SD
6.93 6.62 6.96 6.84 0.19
8.73 9.17 9.48 9.13 0.37
displacement (DZ, mm) of the Bioman cantilever beam loadcell, due to the shrinkage-stress of the restorative materials for two different disk-specimen thicknesses are presented in Table 2. The stiffness of the system was apparent in that the maximum axial displacement was generally less than 12 mm (Figs. 4– 8; Table 2). Since this displacement occurred, however, the initial apparent stress recorded (i.e. load/area) was reduced relative to what it would have been in a system of greater stiffness. The corrected shrinkage-stress, as a function of time, for two different sample thicknesses of each material are shown in Figs. 9– 12, and the 30 min stress values are given in Table 3. The overall magnitudes of these stresses were between 5 and 8 MPa. These results were analysed by oneway ANOVA applied successively to each set of materials with a fixed specimen thickness. The data-sets for each material were also separately analysed, according to the two specimen thicknesses, by paired t-tests. The materials examined fell into either two or three homogenous sub-sets, depending upon the specimen thickness, according to the Scheffe´ procedure at the 0.05 level, as indicated in Table 3. The pair-wise comparisons for thickness effects on the shrinkage stress showed significant differences ( p , 0.05) for all materials, the maximum-stress
Fig. 9. Polymerization shrinkage stress kinetics for Clearfil-APX, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
values obtained with the greater thickness (lower C-factor) were significantly higher statistically. However the increases of stress with thickness were only moderate, in the range 11 – 15 percent.
6. Discussion The Bioman instrument and measurement methodology are based on a highly rigid cantilever load-cell, capable of direct calibration. The overall shrinkage-stress equipment was self-standing, using a standard PC and data-acquisition system. The equipment was compact, stable and portable. It proved to be effective and ergonomic to deploy in practice. At present, temperature control in the specimen chamber has not been added. This is more challenging design problem than with the ‘bonded-disk’ shrinkage-strain device, which incorporates this feature [21]. However, the instrument is highly appropriate for studying differing modes of irradiation, such as soft-start and LED sources.
8
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
Fig. 10. Polymerization shrinkage stress kinetics for P60, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
Fig. 11. Polymerization shrinkage stress kinetics for Point-4, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
The rapid response of the overall measurement-system made this suitable for determining stress-kinetics in all types of self-cured and photo-polymerising materials. The measurements of both system (cantilever) displacement and shrinkage-stress were fully reproducible such that three repeat measurements on a material proved sufficient, with coefficients of variation typically less than 3 percent. The magnitudes of shrinkage-stress obtained were of a similar order of magnitude to previous studies on composites [4,35,43 – 45], and appeared to be clinically reasonable. They may be compared in relation to the known variability of bond-strengths to dental hard tissues in cavity situations. A transparent protocol for determination of stress corresponding to a standardized and clinically appropriate system-compliance was devised. It is not claimed that this is fixed or absolute and thus incapable of revision. Rather, the basis of the stress correction—by a multiplying factor of x4 on the raw data—has been fully explained. Hence, if subsequently deemed appropriate, the correction factor may be adjusted.
It is also possible to combine shrinkage-stress/time ‘vectors’ with shrinkage-strain/time ‘vectors’ to yield the vector of modulus as a function of time; that is, the development profile of modulus during photopolymerisation. The studies of different specimen thicknesses showed that stress values higher by 11 – 15 percent were obtained for the thicker specimens. Two opposing factors are evidently at work in generating this net increase (Fig. 13). An increase in stress may be expected when a greater thickness of material is present to exert a pull on the cantilever arm. However, the greater thickness is also associated with a lower C-factor, which may provide for some slight radial stress-relief, diminishing the axial (measurable) stress. This latter tendency would be in agreement with previous studies of shrinkage-stress [30,35], and with previous work on geometry-dependence of shrinkage-strain [21]. Nevertheless, previous studies [30,35] on thicknesseffects on shrinkage-stress were especially concerned with very small adhesive or composite thicknesses, as occur
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
9
Table 3 Polymerization shrinkage-stress data for Clearfil-APX, Point-4, P60, and Z250, from 0 to 30 min at 23 8C. Multiple measurement runs are shown (n ¼ 3–5), following polymerization with the Elipar light unit at full intensity (800 mW/cm2). The raw-stress values have been multiplied by a correction-factor of four Materials
Fig. 12. Polymerization shrinkage stress kinetics for Z250, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
clinically in the cementation of inlays. A substantial variation of stress with thickness was more apparent at thicknesses , 300 mm. Compliance of the substrate materials was observed to have a rather complex effect upon the stress-variation with thickness [30] at these lower thickness ranges. Hence, in the present work, where composite materials were measured in greater thicknesses, more characteristic of restorative clinical applications, it is advantageous that the stress-levels measured are not too strongly dependent upon material thickness. Hence, values may be quoted (in MPa) for a reference thickness, such as 0.8, 1.0 or 1.2 mm. Although it is not the main focus of this publication, it is of great interest to compare trends, magnitudes and kinetics of shrinkage-stress with those of shrinkage-strain for the same materials [21]. Amongst the present composite materials, the larger stresses observed with Point-4e, may be attributable to a greater resin-monomer volume fraction in this product.
Corrected maximum polymerization shrinkage-stress, sample thickness 0.8 mm (calculated with CF ¼ 4.00)
1.2 mm (calculated with CF ¼ 4.00)
Independent sample t-test results
AP-X
1 2 3 Mean SD
4.86 4.82 4.95 4.88a 0.07
5.66 5.29 5.38 5.44d 0.19
p ¼ 0.026
Point-4
1 2 3 Mean SD
6.50 7.30 7.06 6.95b 0.41
7.50 8.06 7.79 7.78e 0.28
p ¼ 0.044
P-60
1 2 3 4 5 Mean SD
5.55 5.14 5.53 5.66 5.72 5.52c 0.23
6.19 5.73 5.84 6.80 7.10 6.33d 0.60
p ¼ 0.035
Z-250
1 2 3 4 5 Mean SD
5.29 5.67 5.21 5.39 5.78 5.47c 0.25
6.30 6.40 5.90 6.56 6.30 6.29d 0.24
p ¼ 0.001
Identical superscript numbers indicate homogenous sub-sets, for data corresponding to a given specimen thickness.
Fig. 13. Effects of specimen thickness upon stress-vectors. At lower thickness (a), the vectors are axially oriented. At greater thickness (b), there is a greater volume of material to exert an increased shrinkage-stress axially. However, where the C-factor is lower, this may be partially offset by some non-axial vectors, leading to some partial relief of stresses measured axially.
10
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
Finally, additional information and understanding can be generated from the data by further analysis of the kinetic profile of stress versus time, especially in relation to extrinsic or intrinsic soft-start effects and the use of different light sources, such as LEDs. Such treatments will be described in subsequent publications.
References [1] Eick JD, Welch FW. Polymerisation shrinkage of posterior composite resins and its possible influence on postoperative sensitivity. Quintessence Int 1986;17:103–11. [2] Bouchlicher MR, Vargas MA, Boyer DB. Effect of composite type, light intensity, configuration factor and laser polymerisation on polymerisation contraction forces. Am J Dent 1997;10:88–96. [3] Carvalho RM, Pereira JC, Yoshiyama M, Pashley DH. A review of polymerisation contraction: the influence of stress development versus stress relief. Oper Dent 1996;21:17 –24. [4] Condon JR, Ferracane JL. Reduction of composite contraction stress through non-bonded microfiller particles. Dent Mater 1998;14: 256–60. [5] Davidson CL, Feilzer AJ. Polymerisation shrinkage and polymerisation shrinkage stress in polymer-based restoratives. J Dent 1997;25: 435–40. [6] Hegdahl T, Gjerdet NR. Contraction stresses of composite resin filling materials. Acta Odontol Scand 1977;35:191–5. [7] Kinomoto Y, Torii M, Takeshige F, Ebisu S. Comparison of polymerization contraction stresses between self- and light-curing composites. J Dent Res 1999;27:383–9. [8] Labella R, Lambrechts P, Van Meerbeek B, Vanherle G. Polymerisation shrinkage and elasticity of flowable composites and filled adhesives. Dent Mater 1999;15:128–37. [9] Pananakis D, Watts DC. Incorporation of the heating effect of the light source in a non-isothermal model of a visible-light-cured resin composite. J Mater Sci 2000;35:4589–600. [10] Sakaguchi RL, Berge HX. Reduced light energy density decreases post-gel contraction while maintaining degree of conversion in composites. J Dent 1998;26:695 –700. [11] Sakaguchi RL, Berge HX. Effect of light intensity on polymerization contraction of posterior composite. J Dent Res 1997;76:74. (IADR abstracts). [12] Versluis A, Douglas WH, Cross M, Sakaguchi RL. Does an incremental filling technique reduce polymerization shrinkage stresses? J Dent Res 1996;75:871 –8. [13] Silikas N, Eliades G, Watts DC. Light intensity effects on resincomposite degree of conversion and shrinkage strain. Dent Mater 2000;16:292 –6. [14] Watts DC, Al Hindi A. Intrinsic soft-start polymerisation shrinkagekinetics in an acrylate-based resin-composite. Dent Mater 1999;15: 39–45. [15] Stansbury JW, Dickens B, Liu DW. Preparation and characterization of cyclopolymerizable resin formulations. J Dent Res 1995;74:1110 –5. [16] Feilzer AJ, Dooren LH, de Gee AJ, Davidson CL. Influence of light intensity on polymerization shrinkage and integrity of restorationcavity interface. Eur J Oral Sci 1995;103:322–6. [17] Sakaguchi RL, Ferracane JL. Stress transfer from polymerization shrinkage of a chemical-cured composite bonded to a pre-cast composite substrate. Dent Mater 1998;14:106–11. [18] de Gee AJ, Feilzer AJ, Davidson CL. True linear polymerization shrinkage of unfilled resins and composites determined with a linometer. Dent Mater 1993;9:11–14. [19] de Gee AJ, Davidson CL, Smith A. A modified dilatometer for continuous recording of volumetric polymerisation shrinkage of composite restorative materials. J Dent 1981;9:36–42.
[20] Watts DC, Cash AJ. Determination of polymerization kinetics in visible-light cured materials: methods development. Dent Mater 1991; 7:281–7. [21] Watts DC, Marouf AS. Optimal specimen geometry in bonded-disk shrinkage-strain measurements on light-cured biomaterials. Dent Mater 2000;16:447–51. [22] Bowen RL. Adhesive bonding of various materials to hard tooth tissues: Vl Forces developing in direct-filling materials during hardening. JADA 1967;74:439–45. [23] Bowen RL, Nemoto K, Rapson JE. Adhesive bonding of various materials to hard tooth tissues: forces developing in composite materials during hardening. JADA 1983;106:475 –7. [24] Davidson CL, de Gee AJ, Feilzer A. The competition between the composite-dentin bond strength and the polymerisation contraction stress. J Dent Res 1984;63:1396–9. [25] Davidson CL, de Gee AJ. Relaxation of polymerisation contraction stresses by flow in dental composites. J Dent Res 1984;63:146 –8. [26] Feilzer AJ, De Gee AJ, Davidson CL. Curing contraction of composites and glass-ionomer cements. J Pros Dent 1988;59: 297– 300. [27] Feilzer AJ, De Gee AJ, Davidson CL. Setting stress in composite resin in relation to configuration of the restoration. J Dent Res 1987;66: 1636–9. [28] Feilzer AJ, De Gee AJ, Davidson CL. Quantitative determination of stress reduction by flow in composite resin restorations. Dent Mater 1990;6:167–71. [29] Feilzer AJ, de Gee AJ, Davidson CL. Increased wall-to-wall curing contraction in thin bonded resin layers. J Dent Res 1989;68: 48– 50. [30] Alster D, Feilzer AJ, de Gee AJ, Davidson CL. Polymerization contraction stress in thin resin composite layers as a function of layer thickness. Dent Mater 1997;13:146–50. [31] Condon JR, Ferracane JL. Polymerisation contraction stress of commercial composites. J Dent Res 1998;77:639. Abstract no. [32] Kunzelmann K-H, Hickel R. Spannungsentwicklung durch polymerisationschrumpfung bei komposit-klebem. Dtsch Zahnartzl Z 1990; 45:699–700. [33] Sakaguchi RL, Versluis A, Douglas WH. Analysis of strain gage method for measurement of post-gel shrinkage in composite resins. Dent Mater 1997;13:233–9. [34] Kinomoto Y, Torii M. Photoelastic analysis of polymerisation contraction stress in resin composite restorations. J Dent 1998;26: 165–71. [35] Choi KK, Condon JR, Ferracane JL. The effects of adhesive thickness on polymerisation contraction stress of composite. J Dent Res 2000; 79:812–7. [36] McCullock AJ, Smith BGN. In vitro studies of cuspal movement produced by adhesive restorative materials. Br Dent J 1986;161: 405 –9. [37] Causton BE, Miller B, Sefton J. The deformation of cusps by bonded posterior composite restorations: an in vitro study. Br Dent J 1985; 159:397–400. [38] Pearson GJ, Hegarty SM. Cusp movement in molar teeth using dentine adhesives and composite filling materials. Biomaterials 1987; 8:473–6. [39] Suliman AA, Boyer DB, Lakes RS. Cusp movement in premolars resulting from composite polymerization shrinkage. Dent Mater 1993; 9:6–10. [40] Suliman AA, Boyer DB, Lakes RS. Interferometric measurements of cusp deformation of teeth restored with composites. J Dent Res 1993; 72:1532–6. [41] Suliman AA, Boyer DR, Lakes RS. Polymerisation shrinkage of composite resins: comparison with tooth deformation. J Prosthet Dent 1994;71:7–12. [42] Meredith N, Setchell W. In vitro measurement of cuspal strain and displacement in composite restored teeth. J Dent 1997;25: 331 –7.
D.C. Watts et al. / Dental Materials 19 (2003) 1–11 [43] Chen HY, Manhart J, Hickel R, Kunzelmann K-H. Polymerisation contraction stress in light-cured packable composite resins. Dent Mater 2001;17:253– 9. [44] Bouschlicher MR, Rueggeberg FA. Effect of ramped light intensity on polymerization force and conversion in a photoactivated composite. J Esthet Dent 2000;12:328–39.
11
[45] Feilzer AJ, de Gee AJ, Davidson CL. Setting stresses in composites for two different curing modes. Dent Mater 1993;9:2–5. [46] Ausiello P, Apicella A, Davidson CL. Effect of adhesive layer properties on stress distribution in composite restorations—a 3D finite element analysis. Dent Mater 2002;18: 295–303.
Photo-polymerization shrinkage-stress kinetics in resin-composites: methods development D.C. Wattsa,*, A.S. Marouf a,b, A.M. Al-Hindia a
University of Manchester Dental School, Higher Cambridge Street, Manchester M15 6FH, UK b Tishreen University Dental School, Lattakia, Syria Received 8 January 2001; revised 9 November 2001; accepted 06 December 2001
Abstract Objectives. Studies of free shrinkage-strain kinetics on restoratives have begun to multiply. However, there have been fewer investigations of the more difficult problem of concurrent stress-kinetic measurements. The aim was to outline design parameters for a new methodology for this problem, amenable especially to light-cured materials, and to present illustrative results for a range of restorative composites. Methods. Absolute values of stress measurable for a given material and geometry are dependent upon the stiffness of the measurement system. In an infinitely stiff system, the measured stress would also tend towards infinity. Real teeth and their cavities are not infinitely stiff; they have elastic and visco-elastic compliance. Consequently, it is important that some minimal, but essentially constant compliance be allowed, whatever the final or time-dependent modulus of the material may be. This goal has been realised by measurement of the timedevelopment, for a disk-geometry specimen (f ¼ 10, h < 1.0 mm) of stress (Sr), with a calibrated cantilever beam-geometry load cell. A novel specimen-holder design was used for this purpose, held in a rigid base assembly. Specimen thicknesses (or gap-widths) of 0.8 and 1.2 mm were specifically investigated on four representative resin-composites. Concurrent measurements were made of the enddisplacement of the cantilever load cell, relative to a lower glass plate retaining the specimen. Results. Load-calibration of the cantilever load cell gave an end-displacement per unit stress of circa 6 mm/MPa This compares with literature values for cuspal compliance or displacement of circa 20 mm. Re-normalisation of the stress-data was implemented. This was accomplished assuming Hooke’s law behavior at each instant and equivalent to a stiffer system, with a correction (multiplier) factor of 4 on the raw-stress values. For the materials examined, resultant maximum-stress levels determined were circa 5 – 8 MPa Stress-levels obtained at 1.2 mm thickness were slightly higher (11 – 15%) than the level of stress obtained at 0.8 mm thickness. This is attributable to the greater mass of material undergoing shrinkage at 1.2 mm, offset slightly by the different C-factors. Significance. The new device is a practical and self-contained system for rapid and accurate measurement of stress-kinetics in photopolymerising and also self-cure materials. q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved Keywords: Polymerisation shrinkage; Stress; Composite resin
1. Introduction Molecular densification during the polymerisation process of dental restoratives, and the macroscopic effects of shrinkage-strain and/or shrinkage-stress, continue to attract widespread international research interest [1 –14]. This is pursued at several levels. There is the need to characterize these properties of both candidate monomer molecules, synthesized with the aim of attaining reduced shrinkage materials [15], and also resin-composite materials – either commercially available or formulated experimentally [4]. In * Corresponding author. Tel.: þ 44-161-275-6749. E-mail address: [email protected] (D.C. Watts).
addition, different light-curing units and modes of operation require investigation with representative materials in the context of shrinkage phenomena [13,16,17]. These properties are strongly coupled and are controlled especially by the degree of conversion of the network [13] (Fig. 1). For more than a decade, several techniques have been developed and utilized for measurement of shrinkage-strain. The dilatometer, linometer and ‘bonded-disk’ methods have been widely employed [18 – 21]. However, progress in developing methods of comparable utility for determining shrinkage-stress has been slower. Part of the difficulty resides in the design of the specimen holder, which must become bonded in the process of measurement. It must be efficiently de-bonded to effect a subsequent measurement.
0109-5641/03/$ - see front matter q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved PII: S 0 1 0 9 - 5 6 4 1 ( 0 2 ) 0 0 1 2 3 - 9
2
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
movement of opposing cusps can be of the order of 20 mm [36 –42]. The greater the displacement of the bonded walls, the lower will be the effective shrinkage stress level sustained. While noting the above problems, and design challenges, we are not suggesting that previous designs are either flawed or in any way unsuitable. Nevertheless, we do contend that there is an important place for design innovation in measuring shrinkage-stress. Accordingly, the aims of this research were:
Fig. 1. Inter-relationship of network shrinkage phenomena.
The earliest work in this area appears to be that of Bowen [22,23] and of Hegdahl [6] who recorded development of shrinkage forces with a Universal Testing (Instron) machine (UTM). Major insights and developments have been achieved by the ACTA group of Davidson, Feilzer, de Gee and Alster [5,24 –30]. These workers have used a servo-hydraulic UTM in which sophisticated and careful procedures have been deployed with the aim of largely eliminating machine or system-compliance. Using this approach, firstly with self-cure composites and later with VLC composites, many important phenomena have been identified, including the effect of C-factor on the stress magnitudes. Broadly similar approaches have been reported by Ferracane [31], Bouchlicher [2] and Kunzelmann [32]. Different approaches to shrinkage phenomena have been made using strain-gages [33], photo-elastic and Moire methods [34], as well as Finite Element modeling [46]. There are, however, several practical disadvantages attached to the use of a large servo-hydraulic UTM for measuring shrinkage-stress kinetics. Firstly, the equipment is expensive and complex and may be required for a variety of research projects. If so, it is inconvenient to use this equipment for extensive measurements without intermission over several weeks. Secondly, the generality of this equipment entails construction of specialist attachments and couplings which tend to increase the machine-compliance requiring correction. Thirdly, in attempting to virtually eliminate compliance by servo-control, there is a fundamental limitation. For a measurement device with absolutely zero compliance, the recorded shrinkage-stress should approach an infinite value, as a corollary of Newton’s third law. In practice, all measurement systems for load (or stress) entail finite compliance. Load cells incorporate strain-gages which deform under load [35]. Furthermore, the hard dental tissues of cavity walls also exhibit compliance and relative
1. to develop a self-standing shrinkage-stress instrument, based on a rigid cantilever load-cell, capable of direct calibration, 2. to formulate a transparent protocol for determination of stress corresponding to a standardized and clinically appropriate system-compliance, 3. to evaluate the shrinkagestress kinetics of representative resin-composites under conventional irradiation regimes by Quartz Tungsten Halogen (QTH) light sources. The overall hypotheses to be tested were that the experimental procedure would (a) independently generate stress-magnitudes in broad correspondence to those perceived to be clinically meaningful by previous researchers and (b) prove to be effective and ergonomic to deploy in practice.
2. Materials and methods The Bioman shrinkage-stress instrument was designed and constructed at the University of Manchester (Fig. 2). Components were bolted to a 2 cm thick stainless steel baseplate. This was mounted vertically on a support frame. A cantilever load-cell of 500 kg capacity was fitted with a rigid integral clamp, at the free or compliant end of the cantilever, to securely hold a circular steel rod (10 mm diameter and 22 mm long) vertically and perpendicular to the load-cell axis. The lower end of the rod was sand-blasted to aid composite-retention and this formed one face of the specimen chamber. Across a gap, normally adjusted to either 0.8 or 1.2 mm, the counter-face consisted of a rigid glass plate, 3 mm in thickness. This was removable, before and after measurements, but was held rigidly relative to the base-plate in a special clamp during measurement. The surface of the glass-plate opposing the steel-rod was also sandblasted to aid composite retention. Composite paste was introduced between the plate and vertical rod to form an un-set specimen-disk of 10 mm diameter and, normally, either 0.8 or 1.2 mm thickness. The disc-radius orientation was thus horizontal so that the instrument could be used for fluid composite pastes as well as stiff pastes. These thicknesses correspond to different ratios of bonded to
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
3
Fig. 2. The Bioman shrinkage stress device: (a) load rod; (b) specimen (detached); (c) glass plate; (d) displacement-signal conditioning unit; (e) dataacquisition unit; (f) LVDT transducer; (g) cantilever cell; (h) light source; (i) load-cell amplifier.
unbonded surface areas, that is configuration factors (C-factor, C ¼ d/2h, where d and h are the diameter and thickness of the disk samples, respectively). C ¼ 6.25 for 0.8 mm thickness and C ¼ 4.16 for 1.2 mm thickness. The composite material was irradiated across its’ thickness dimension, rather than radially (Fig. 2). This corresponds to the orientation utilized by Condon and Ferracane [4]. During polymerization, stresses established through the material, and thus between the fixed glass plate and the rod, caused slight displacement of the free end of the loadcell. Along with the load-cell signal, this displacement was continually recorded to computer using an attached (LVDT) displacement transducer and a fast data-acquisition system. The displacement was measured as movement (in mm) between the load-cell end and the fixed glass-plate counter-face. The load-signal from the cantilever cell, which was amplified by a wide-range strain indicator (Model 3800, Vishay, Measurements Group, Rayleigh, NC, USA) was calibrated at periodic intervals, as described later. Prior to photo-polymerization, the system and composite-resin were maintained at a starting temperature of 23 8C. The curved optic of each light source utilized was replaced by a straight version of the same diameter,
typically 8 mm. The end of the optic was positioned at the desired distance, normally 0.5 mm, from the 3 mm thick glass-plate. Thereby full power from the light-source could be delivered to the composite, corresponding to optimal clinical irradiation conditions. Where temperature rises were produced by radiant energy and/or exotherm, these were permitted to occur, as may often obtain clinically. Four representative resin-composites materials were investigated (Table 1). Two groups (A and B) of specimens (n ¼ 3 – 5) of each composite were prepared in situ between the rod and glass plate, described above. Group A were 0.8 mm in thickness and Group B were 1.2 mm thick. For example, to produce the 0.8 mm specimens, typically 0.12 – 0.15 g of material were formed into the disk shape. The precise amount of unset material required for a 10 mm diameter, without excess, was established by trial closures, Table 1 Light-cured hybrid resin-composites investigated Materials
Code
Lot
Shade
Manufacturer
Clearfil AP-X Point4 P60 Z250
APX Point4 P60 Z250
584 910576 9AC 19991018
A3 A3 A3 A3
Kuraray, Osaka, Japan SDS-Kerr, CA, USA 3M, St Paul, MN, USA 3M, St Paul, MN, USA
4
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
in accordance with the following protocol: (i) the glass plate was firmly clamped horizontally; (ii) the steel rod was initially unclamped and using a 0.8 or 1.2 mm feeler gauge, the gap below the rod was set; (iii) then, while preventing any further vertical movement, the rod was clamped above the glass plate; (iv) the glass plate was then unclamped and removed; (v) using a template, a 10 mm diameter circle was drawn on the glass plate to identify the required specimen paste location directly below the rod; (vi) the composite paste was inserted on the glass, within the drawn circle; (vii) the glass plate was then re-positioned, raised and firmly clamped as before; (viii) this caused the paste to flow and adapt to the required specimen height, in the required location; (ix) any slight excess was removed manually, although with precise weighing this was rarely necessary. A period of 2 min was allowed for stabilisation, including any internal stress-relaxation, within each specimen, prior to irradiation. An Elipar II light source (3M-Espe, Seefeld, Germany) was used to irradiate the composites for 40 s photo-cure at 800 mW/cm2, as measured with a ‘Power Max 500A’ radiometer (PM500A, Molectron Detector Inc., Portland, OR, USA). The initial specimen temperature was 23 ^ 1 8C. Exactly 20 s prior to lamp-illumination, the data-recording
system was triggered for concurrent measurement of displacement and load, for the desired time period. This ranged from 10 min to 2 h, but in the present work data are presented for 0 – 30 min. Subsequently, the 20 s preirradiation baseline-data were deleted from file.
3. Device calibration procedures Load-calibration of the device was achieved by detaching the components opposing the cantilever end and, with the base-plate clamped vertically, attaching a series (n ¼ 8) of 5 kg calibration weights, via a rod-plus-retainer, in the axial-load direction (Fig. 3). This gave a strongly linear calibration plot (r , 0.999) of load-cell signal (mV) versus applied-load (N). During load-calibration, concurrent measurements were made of the end-deflection of the cantilever, in mm, for each applied load-increment. This was achieved using a device to clamp an LVDT displacement transducer onto the top of the base-plate with the LVDT-probe resting on the cantilever end, and in line with the axis of loading. When the applied loads were re-expressed as equivalent ‘applied-stress’ values, by dividing by the end-area of the 10 mm diameter
Fig. 3. Load-calibration arrangement, where some components are unbolted to permit application of successive 5 kg masses.
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
5
steel ‘load-rod’ (Fig. 2(a)), a plot of cantilever-displacement versus applied-stress was generated (Fig. 4). This showed a displacement of circa 6 mm/MPa. This means of measuring the cantilever displacement during load-calibration was different in detail, but generally comparable with the concurrent axial-displacement (DZ ) measurements made during routine measurement of shrinkage-stress.
4. Correction of stress data
Fig. 4. Calibration graph of cantilever-displacement versus stress, applied by means of standard weights (each of 5 kg). The load values were reexpressed as stresses equivalent to those acting on a circular specimen disk area of diameter 10 mm.
As noted in the introduction, stresses produced in tooth cavities by polymerising restoratives can result in substantial inward cuspal movement. This is analogous to compliance of many stress-measurement assemblies [35]. Review of the literature [36 – 42] indicated that total cuspal movements produced by the polymerisation of composite resin fillings could be of the order of 20 –25 mm. In order to establish an equal basis for comparison of materials, that
Fig. 5. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of ClearfilAPX, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
Fig. 6. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of Point-4, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
6
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
would have clinical relevance, it was decided to compare the actual displacement of the cantilever end, and thus of the disk specimen heights, with that occurring in tooth cavities. For a typical value of 2 MPa stress recorded as initial (raw) data, the end-displacement was circa 12 mm, as seen in the displacement-calibration plot (Fig. 4). This is the same order of magnitude as some of the dental cuspal deformations. The compliance (reciprocal stiffness) of the instrument is always the same (circa 6 mm/MPa) irrespective of a given disk-thickness of specimen. Nevertheless, we have considered it appropriate to treat the data as corresponding to a higher stiffness load cell (lower compliance), and have chosen to apply a constant multiplying Correction Factor of £ 4. This involves the reasonable assumption of Hooke’s law at each time interval. This then is equivalent to a more conservative (lower) estimate of ‘permitted’ cuspal displacement, equivalent to circa 3 mm. Hence the ‘raw’ shrinkage-stress values have been all multiplied by 4 to give the final shrinkage-stress values that are reported in the following tables.
Thus, if (say) the ‘raw’ stress was 1.5 MPa and the correction factor was 4, the corrected shrinkage-stress would be 6.0 MPa. These straightforward calculations were effected either with transform capabilities incorporated in standard graphical programs (Sigmaplot, www. spssscience.com; FigP, www.biosoft.com) or by use of custom routines written in Matlab (www.mathworks.com).
Fig. 7. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of P60, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
Fig. 8. Three measurement runs of the maximum displacement (DZ ) of the Bioman cantilever beam load-cell, due to the shrinkage stress of Z250, for two different sample thicknesses: (a) and (b) represent 0.8 and 1.2 mm high specimens, respectively.
5. Results During each measurement run, the shrinkage forces exerted by the polymerising composite were able to pull the steel rod downwards together with the cantilever end of the rather stiff load cell. This movement occurred relative to the fixed spatial coordinates of the glass plate and lower surface of the composite disk. The displacement (mm) values of the cantilever load-cell for each material with 0.8 and 1.2 mm sample thicknesses are presented in Figs. 5 –8. The mean values of the maximum
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
7
Table 2 Mean values of the maximum displacement (DZ, mm) of the Bioman cantilever beam load-cell, due to the shrinkage-stress of the restorative materials for two different disk-specimen thicknesses Materials
Maximum values of the displacement (mm) of the load beam-cell due to the shrinkage stress of the restorative materials Run #
Displacement (DZ ) (mm) Sample thickness 0.8 mm
1.2 mm
AP-X
1 2 3 Mean SD
7.68 7.66 7.89 7.74 0.13
7.90 7.70 7.50 7.70 0.20
Point-4
1 2 3 Mean SD
10.04 11.16 10.67 10.62 0.56
12.60 12.38 11.80 12.26 0.40
P-60
1 2 3 Mean SD
7.60 7.75 7.73 7.70 0.08
8.56 7.82 8.81 8.40 0.50
Z-250
1 2 3 Mean SD
6.93 6.62 6.96 6.84 0.19
8.73 9.17 9.48 9.13 0.37
displacement (DZ, mm) of the Bioman cantilever beam loadcell, due to the shrinkage-stress of the restorative materials for two different disk-specimen thicknesses are presented in Table 2. The stiffness of the system was apparent in that the maximum axial displacement was generally less than 12 mm (Figs. 4– 8; Table 2). Since this displacement occurred, however, the initial apparent stress recorded (i.e. load/area) was reduced relative to what it would have been in a system of greater stiffness. The corrected shrinkage-stress, as a function of time, for two different sample thicknesses of each material are shown in Figs. 9– 12, and the 30 min stress values are given in Table 3. The overall magnitudes of these stresses were between 5 and 8 MPa. These results were analysed by oneway ANOVA applied successively to each set of materials with a fixed specimen thickness. The data-sets for each material were also separately analysed, according to the two specimen thicknesses, by paired t-tests. The materials examined fell into either two or three homogenous sub-sets, depending upon the specimen thickness, according to the Scheffe´ procedure at the 0.05 level, as indicated in Table 3. The pair-wise comparisons for thickness effects on the shrinkage stress showed significant differences ( p , 0.05) for all materials, the maximum-stress
Fig. 9. Polymerization shrinkage stress kinetics for Clearfil-APX, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
values obtained with the greater thickness (lower C-factor) were significantly higher statistically. However the increases of stress with thickness were only moderate, in the range 11 – 15 percent.
6. Discussion The Bioman instrument and measurement methodology are based on a highly rigid cantilever load-cell, capable of direct calibration. The overall shrinkage-stress equipment was self-standing, using a standard PC and data-acquisition system. The equipment was compact, stable and portable. It proved to be effective and ergonomic to deploy in practice. At present, temperature control in the specimen chamber has not been added. This is more challenging design problem than with the ‘bonded-disk’ shrinkage-strain device, which incorporates this feature [21]. However, the instrument is highly appropriate for studying differing modes of irradiation, such as soft-start and LED sources.
8
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
Fig. 10. Polymerization shrinkage stress kinetics for P60, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
Fig. 11. Polymerization shrinkage stress kinetics for Point-4, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
The rapid response of the overall measurement-system made this suitable for determining stress-kinetics in all types of self-cured and photo-polymerising materials. The measurements of both system (cantilever) displacement and shrinkage-stress were fully reproducible such that three repeat measurements on a material proved sufficient, with coefficients of variation typically less than 3 percent. The magnitudes of shrinkage-stress obtained were of a similar order of magnitude to previous studies on composites [4,35,43 – 45], and appeared to be clinically reasonable. They may be compared in relation to the known variability of bond-strengths to dental hard tissues in cavity situations. A transparent protocol for determination of stress corresponding to a standardized and clinically appropriate system-compliance was devised. It is not claimed that this is fixed or absolute and thus incapable of revision. Rather, the basis of the stress correction—by a multiplying factor of x4 on the raw data—has been fully explained. Hence, if subsequently deemed appropriate, the correction factor may be adjusted.
It is also possible to combine shrinkage-stress/time ‘vectors’ with shrinkage-strain/time ‘vectors’ to yield the vector of modulus as a function of time; that is, the development profile of modulus during photopolymerisation. The studies of different specimen thicknesses showed that stress values higher by 11 – 15 percent were obtained for the thicker specimens. Two opposing factors are evidently at work in generating this net increase (Fig. 13). An increase in stress may be expected when a greater thickness of material is present to exert a pull on the cantilever arm. However, the greater thickness is also associated with a lower C-factor, which may provide for some slight radial stress-relief, diminishing the axial (measurable) stress. This latter tendency would be in agreement with previous studies of shrinkage-stress [30,35], and with previous work on geometry-dependence of shrinkage-strain [21]. Nevertheless, previous studies [30,35] on thicknesseffects on shrinkage-stress were especially concerned with very small adhesive or composite thicknesses, as occur
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
9
Table 3 Polymerization shrinkage-stress data for Clearfil-APX, Point-4, P60, and Z250, from 0 to 30 min at 23 8C. Multiple measurement runs are shown (n ¼ 3–5), following polymerization with the Elipar light unit at full intensity (800 mW/cm2). The raw-stress values have been multiplied by a correction-factor of four Materials
Fig. 12. Polymerization shrinkage stress kinetics for Z250, from 0 to 30 min at 23 8C. Three measurement runs are shown, following polymerization with the Elipar light unit at full intensity (800 mW/cm2).
clinically in the cementation of inlays. A substantial variation of stress with thickness was more apparent at thicknesses , 300 mm. Compliance of the substrate materials was observed to have a rather complex effect upon the stress-variation with thickness [30] at these lower thickness ranges. Hence, in the present work, where composite materials were measured in greater thicknesses, more characteristic of restorative clinical applications, it is advantageous that the stress-levels measured are not too strongly dependent upon material thickness. Hence, values may be quoted (in MPa) for a reference thickness, such as 0.8, 1.0 or 1.2 mm. Although it is not the main focus of this publication, it is of great interest to compare trends, magnitudes and kinetics of shrinkage-stress with those of shrinkage-strain for the same materials [21]. Amongst the present composite materials, the larger stresses observed with Point-4e, may be attributable to a greater resin-monomer volume fraction in this product.
Corrected maximum polymerization shrinkage-stress, sample thickness 0.8 mm (calculated with CF ¼ 4.00)
1.2 mm (calculated with CF ¼ 4.00)
Independent sample t-test results
AP-X
1 2 3 Mean SD
4.86 4.82 4.95 4.88a 0.07
5.66 5.29 5.38 5.44d 0.19
p ¼ 0.026
Point-4
1 2 3 Mean SD
6.50 7.30 7.06 6.95b 0.41
7.50 8.06 7.79 7.78e 0.28
p ¼ 0.044
P-60
1 2 3 4 5 Mean SD
5.55 5.14 5.53 5.66 5.72 5.52c 0.23
6.19 5.73 5.84 6.80 7.10 6.33d 0.60
p ¼ 0.035
Z-250
1 2 3 4 5 Mean SD
5.29 5.67 5.21 5.39 5.78 5.47c 0.25
6.30 6.40 5.90 6.56 6.30 6.29d 0.24
p ¼ 0.001
Identical superscript numbers indicate homogenous sub-sets, for data corresponding to a given specimen thickness.
Fig. 13. Effects of specimen thickness upon stress-vectors. At lower thickness (a), the vectors are axially oriented. At greater thickness (b), there is a greater volume of material to exert an increased shrinkage-stress axially. However, where the C-factor is lower, this may be partially offset by some non-axial vectors, leading to some partial relief of stresses measured axially.
10
D.C. Watts et al. / Dental Materials 19 (2003) 1–11
Finally, additional information and understanding can be generated from the data by further analysis of the kinetic profile of stress versus time, especially in relation to extrinsic or intrinsic soft-start effects and the use of different light sources, such as LEDs. Such treatments will be described in subsequent publications.
References [1] Eick JD, Welch FW. Polymerisation shrinkage of posterior composite resins and its possible influence on postoperative sensitivity. Quintessence Int 1986;17:103–11. [2] Bouchlicher MR, Vargas MA, Boyer DB. Effect of composite type, light intensity, configuration factor and laser polymerisation on polymerisation contraction forces. Am J Dent 1997;10:88–96. [3] Carvalho RM, Pereira JC, Yoshiyama M, Pashley DH. A review of polymerisation contraction: the influence of stress development versus stress relief. Oper Dent 1996;21:17 –24. [4] Condon JR, Ferracane JL. Reduction of composite contraction stress through non-bonded microfiller particles. Dent Mater 1998;14: 256–60. [5] Davidson CL, Feilzer AJ. Polymerisation shrinkage and polymerisation shrinkage stress in polymer-based restoratives. J Dent 1997;25: 435–40. [6] Hegdahl T, Gjerdet NR. Contraction stresses of composite resin filling materials. Acta Odontol Scand 1977;35:191–5. [7] Kinomoto Y, Torii M, Takeshige F, Ebisu S. Comparison of polymerization contraction stresses between self- and light-curing composites. J Dent Res 1999;27:383–9. [8] Labella R, Lambrechts P, Van Meerbeek B, Vanherle G. Polymerisation shrinkage and elasticity of flowable composites and filled adhesives. Dent Mater 1999;15:128–37. [9] Pananakis D, Watts DC. Incorporation of the heating effect of the light source in a non-isothermal model of a visible-light-cured resin composite. J Mater Sci 2000;35:4589–600. [10] Sakaguchi RL, Berge HX. Reduced light energy density decreases post-gel contraction while maintaining degree of conversion in composites. J Dent 1998;26:695 –700. [11] Sakaguchi RL, Berge HX. Effect of light intensity on polymerization contraction of posterior composite. J Dent Res 1997;76:74. (IADR abstracts). [12] Versluis A, Douglas WH, Cross M, Sakaguchi RL. Does an incremental filling technique reduce polymerization shrinkage stresses? J Dent Res 1996;75:871 –8. [13] Silikas N, Eliades G, Watts DC. Light intensity effects on resincomposite degree of conversion and shrinkage strain. Dent Mater 2000;16:292 –6. [14] Watts DC, Al Hindi A. Intrinsic soft-start polymerisation shrinkagekinetics in an acrylate-based resin-composite. Dent Mater 1999;15: 39–45. [15] Stansbury JW, Dickens B, Liu DW. Preparation and characterization of cyclopolymerizable resin formulations. J Dent Res 1995;74:1110 –5. [16] Feilzer AJ, Dooren LH, de Gee AJ, Davidson CL. Influence of light intensity on polymerization shrinkage and integrity of restorationcavity interface. Eur J Oral Sci 1995;103:322–6. [17] Sakaguchi RL, Ferracane JL. Stress transfer from polymerization shrinkage of a chemical-cured composite bonded to a pre-cast composite substrate. Dent Mater 1998;14:106–11. [18] de Gee AJ, Feilzer AJ, Davidson CL. True linear polymerization shrinkage of unfilled resins and composites determined with a linometer. Dent Mater 1993;9:11–14. [19] de Gee AJ, Davidson CL, Smith A. A modified dilatometer for continuous recording of volumetric polymerisation shrinkage of composite restorative materials. J Dent 1981;9:36–42.
[20] Watts DC, Cash AJ. Determination of polymerization kinetics in visible-light cured materials: methods development. Dent Mater 1991; 7:281–7. [21] Watts DC, Marouf AS. Optimal specimen geometry in bonded-disk shrinkage-strain measurements on light-cured biomaterials. Dent Mater 2000;16:447–51. [22] Bowen RL. Adhesive bonding of various materials to hard tooth tissues: Vl Forces developing in direct-filling materials during hardening. JADA 1967;74:439–45. [23] Bowen RL, Nemoto K, Rapson JE. Adhesive bonding of various materials to hard tooth tissues: forces developing in composite materials during hardening. JADA 1983;106:475 –7. [24] Davidson CL, de Gee AJ, Feilzer A. The competition between the composite-dentin bond strength and the polymerisation contraction stress. J Dent Res 1984;63:1396–9. [25] Davidson CL, de Gee AJ. Relaxation of polymerisation contraction stresses by flow in dental composites. J Dent Res 1984;63:146 –8. [26] Feilzer AJ, De Gee AJ, Davidson CL. Curing contraction of composites and glass-ionomer cements. J Pros Dent 1988;59: 297– 300. [27] Feilzer AJ, De Gee AJ, Davidson CL. Setting stress in composite resin in relation to configuration of the restoration. J Dent Res 1987;66: 1636–9. [28] Feilzer AJ, De Gee AJ, Davidson CL. Quantitative determination of stress reduction by flow in composite resin restorations. Dent Mater 1990;6:167–71. [29] Feilzer AJ, de Gee AJ, Davidson CL. Increased wall-to-wall curing contraction in thin bonded resin layers. J Dent Res 1989;68: 48– 50. [30] Alster D, Feilzer AJ, de Gee AJ, Davidson CL. Polymerization contraction stress in thin resin composite layers as a function of layer thickness. Dent Mater 1997;13:146–50. [31] Condon JR, Ferracane JL. Polymerisation contraction stress of commercial composites. J Dent Res 1998;77:639. Abstract no. [32] Kunzelmann K-H, Hickel R. Spannungsentwicklung durch polymerisationschrumpfung bei komposit-klebem. Dtsch Zahnartzl Z 1990; 45:699–700. [33] Sakaguchi RL, Versluis A, Douglas WH. Analysis of strain gage method for measurement of post-gel shrinkage in composite resins. Dent Mater 1997;13:233–9. [34] Kinomoto Y, Torii M. Photoelastic analysis of polymerisation contraction stress in resin composite restorations. J Dent 1998;26: 165–71. [35] Choi KK, Condon JR, Ferracane JL. The effects of adhesive thickness on polymerisation contraction stress of composite. J Dent Res 2000; 79:812–7. [36] McCullock AJ, Smith BGN. In vitro studies of cuspal movement produced by adhesive restorative materials. Br Dent J 1986;161: 405 –9. [37] Causton BE, Miller B, Sefton J. The deformation of cusps by bonded posterior composite restorations: an in vitro study. Br Dent J 1985; 159:397–400. [38] Pearson GJ, Hegarty SM. Cusp movement in molar teeth using dentine adhesives and composite filling materials. Biomaterials 1987; 8:473–6. [39] Suliman AA, Boyer DB, Lakes RS. Cusp movement in premolars resulting from composite polymerization shrinkage. Dent Mater 1993; 9:6–10. [40] Suliman AA, Boyer DB, Lakes RS. Interferometric measurements of cusp deformation of teeth restored with composites. J Dent Res 1993; 72:1532–6. [41] Suliman AA, Boyer DR, Lakes RS. Polymerisation shrinkage of composite resins: comparison with tooth deformation. J Prosthet Dent 1994;71:7–12. [42] Meredith N, Setchell W. In vitro measurement of cuspal strain and displacement in composite restored teeth. J Dent 1997;25: 331 –7.
D.C. Watts et al. / Dental Materials 19 (2003) 1–11 [43] Chen HY, Manhart J, Hickel R, Kunzelmann K-H. Polymerisation contraction stress in light-cured packable composite resins. Dent Mater 2001;17:253– 9. [44] Bouschlicher MR, Rueggeberg FA. Effect of ramped light intensity on polymerization force and conversion in a photoactivated composite. J Esthet Dent 2000;12:328–39.
11
[45] Feilzer AJ, de Gee AJ, Davidson CL. Setting stresses in composites for two different curing modes. Dent Mater 1993;9:2–5. [46] Ausiello P, Apicella A, Davidson CL. Effect of adhesive layer properties on stress distribution in composite restorations—a 3D finite element analysis. Dent Mater 2002;18: 295–303.