Dent Mater 14:2128, January, 1998

Surface energy characteristics of adhesive monomers

Erik Asmussen, Anne Peutzfeldt

Department of Dental Materials, University of Copenhagen, Copenhagen, DENMARK

ABSTRACT
Objectives. The aim of the investigation was to determine the surface free-energy components of potentially adhesive monomer mixtures. Methods. Four liquids with known components of surface free-energy were used as reference. Small drops of the liquids were placed on the polished surfaces of four types of solid (metal, porcelain, resin composite and hydrocarbon), and the contact angles were measured. By means of the fundamental equations for wetting, the three components of the surface free-energy of the four solids were calculated. Small drops of various monomeric mixtures were then placed on the four solid surfaces, and on the basis of the previously calculated components of surface free-energy of the solids, the surface energy characteristics of the monomeric mixtures were determined. The relationships between contact angles and composition, and between surface tension and composition, were studied by regression analyses. Comparisons between values were carried out by means of NeumanKeuls multiple range test at a level of statistical signicance of p = 0.05. Results. Statistically signicant differences between the monomeric mixtures as regards the wetting of the four solids were observed. These differences reected differences in the acid or base component of the surface free-energy of the monomers. In particular, monomeric mixtures containing HEMA, MAN or 4-META exhibited a signicant acid component of the surface free-energy. Signicance. Knowledge of the surface free-energy components of monomers throws light on the mechanisms associated with the adhesion of resin composites, including resin cements. A better understanding of the interfacial interactions may act as guide in a research aimed at developing resin materials of increased adhesion to metal, porcelain or resin composite. 1998 Academy of Dental Materials. Published by Elsevier Science Ltd

INTRODUCTION
In modern dentistry, increasing focus is directed towards adhering restorations. Adhering restorations imply that sound tooth structure may be saved because undercuts are abolished. Gap-free restorations may be another benet of an effective adherence between restoration and tooth. Resin composites will adhere to
PII: S0109-5641(98)00005-0

tooth structure following appropriate pretreatments of enamel and dentin (Zidan et al., 1980; McCabe and Rusby, 1994). Indirect restorations are potentially adhering when cemented with a resin cement. Such a cement will adhere to tooth structure in the same way as a resin composite, i.e., by micro-mechanical interlocking to etched enamel, and by attachment to dentin via the hybrid layer (Nakabayashi et al., 1982). Thus, the bonding to dentin is dependent on diffusion of resin monomers into the dentin surface. The adherence of the resin cement to the restoration differs in that it does not or only minimally rely on diffusion into the restoration surface, but stems from micro-mechanical interlocking and from adhesion in the physico-chemical sense of the word. The methods used to convey micromechanical attachment of a resin cement to the inner surfaces of restorations are numerous (Thompson et al., 1983; Saunders, 1987; van der Veen et al., 1988). Depending on the material of the restoration, the methods include etching, sandblasting and tin plating. The adhesion, on the other hand, is the result of a physico-chemical interaction across the interphase between the resin cement (adhesive) and the restoration (adherend = substrate). Some resin cements are known as adhesive because these interactions are of a signicant size, while other cements adhere mainly by micro-mechanical interlocking to the inner surfaces of the restoration (Tanaka et al., 1979, 1981, 1986, 1988). The interactions across the interphase between adhesive and adherend have been analyzed by Fowkes (1990). The thermodynamic work of adhesion WA between a solid surface (subscript s) and a liquid (subscript l) may be expressed as d 1a2 + 2gs+ gl 1a2 + 2gs gl+ 1a2 WA = gl (1 + cos ) = 2 gd s gl (1) In this expression g refers to the surface free-energy,
Dental Materials/January 1998 21

MATERIALS AND METHODS

TABLE 1: LIST OF MATERIALS CONSTITUTING THE SOLID SURFACES General. The three components of the surface free-energy of a liquid may be determined by measurements of the contact Remanium CD 65Co,28Cr,5Mo,2Sia 27967 Dentaurum, angle of a drop of the liquid Pfortzheim, Germany placed on at least three different solid surfaces of known Biodent Inlay Ceramic Feldspathic porcelain D2: 22 De Trey, surface energy characteristics. The Dreieich, Germany surface energy characteristics of a solid surface may be determined 406275 Kerr, Herculite XRV Resin matrix: BisGMA/TEGDMAb b by measurements of the contact Romulus, MI, USA Filler type: barium aluminum borosilica angles formed by at least three liquids of known surface freeParalm Mixture of polyalkenes 661B1 American Can Company, energy components. While the and parafn waxesa Greenwich, CT, USA surface free-energy components of solids are rarely described in BisGMA=bisphenol-A-glycidyl methacrylate; TEGDMA=triethyleneglycol dimethacrylate. the literature, these characteristics Source: amanufacturer; bLee et al. (1995). have been determined for several liquids. Thus, by use of four liquids with known surface and the superscripts d, + and to the dispersive, acid energy parameters, the contact angles on four solid and base component, respectively, of g. WA is also given surfaces were measured and the surface free-energy by W A = g l(1 + cos ), where is the contact angle of a parameters of these four surfaces were calculated. The measurements on solid surfaces may in turn lead to the drop of the liquid placed on the solid. The restoration of a given material has surface surface energy characteristics of other liquids, includ+ ing monomers of resin cements. energy characteristics dened by g s, gd s , gs and gs . A resin cement has surface energy characteristics given Solid surfaces. The materials forming the solid surfaces + used to establish the surface energy characteristics of by g l, gd l , gl and gl . An effective adhesion calls for a proper balance between the dispersive, base and acid the experimental liquids, are identied in Table 1. They surface free-energy values of the adhesive and the were a chromiumcobalt alloy, a feldspathic porcelain, an inlay material made of resin composite and adherend in order that WA be maximized. Earlier studies have demonstrated signicant differ- Paralm. The three former materials were used to ences between resin cements in their capability of produce small beams of dimensions 60 mm 6 mm adhering to solid surfaces (Asmussen et al., 1993; Flood 1.5 mm in accordance with the recommendations of et al., 1989; Atta et al., 1990). Scientically based the manufacturer. Before being used in the the surfaces were polished on explanations of the adhering qualities of various resin experiments, cements necessitate a determination of the surface free- carborundum paper #1000, rinsed with ethanol and energy characteristics of the cements. In an earlier dryed with a hair dryer. The fourth material, Paralm, study it was demonstrated that differences in is a material known as having zero polar component of adherence energy exibited by a series of experimental the surface free-energy. The lm was drawn gently over a resin cements could be explained, in part, by the polar small 20 mm 20 mm 10 mm box of Plexiglass to give a component of the surface free-energy of the cement smooth planar surface. monomers (Asmussen et al., 1995). The polar com- Liquids. The compounds used to produce the liquids of ponent of the surface free-energy g ab of a solid or a the investigation are listed in Table 2. The rst four liquid is dened as g ab = 2(g +g ) 1/2 (Fowkes, 1990). liquids are the reference liquids of known surface However, that study did not determine the individual energy components (Constanzo et al., 1990). The next values of the acid and base components of the polar part three compounds are methacrylate monomers and of the surface free-energy of the cements. common components of resin composites and bonding It was the aim of the present work to measure the agents. The propanal (propionic aldehyde = PAL) was three components of the surface free-energy of mono- included in the study because of its relatively polar mers that may be used in resin cements. It was nature (Asmussen and Uno, 1993) and because of its hypothesized that addition of presumably polar positive inuence on mechanical properties of resin compounds to conventional dental monomers would composites (Peutzfeldt and Asmussen, 1992). Maleic result in considerable changes in the acid and base acid anhydride (MAN) has potential use in the components of the surface free-energy of the mono- formulation of adhesive resin cements (Asmussen et meric mixtures. The information obtained may lead al., 1995). 4-META is an anhydride already used in to insight into the mechanisms of bonding to metal, adhesive resin cements (Takeyama et al., 1978; porcelain and resin composite, and may prove Watanabe et al., 1988). valuable in the formulation of new, adhesive resin Five series of liquids were used in the experiments. cements. In the rst series the liquids were mixtures of HEMA
Material Composition Batch no. Source
22 Asmussen & Peutzfeldt/Surface energy of monomers

to give the contact angle of that particular drop. For each liquid ve drops were examined, and the mean Material category Compound Acronym Batch no. Source value calculated from the Reference liquids Water Distilled ve determinations. Calculations. As mentioned Glycerol 944-1C13142393 Merck, above, the reference liquids Darmstadt, Germany were water, glycerol, ethylene glycol and 1-bromonaphEthylene glycol 1109153 Merck, thalene. The free surfaceDarmstadt, Germany energy values were taken from Constanzo et al. (1990). 1-Bromonaphthalene 070035-095 Aldrich, From Eq. (1) four equations Steinheim, Germany were set up, one for each of the reference liquids, and gd s, Methacrylate monomers 2-Hydroxyethyl HEMA, H 53217161 Merck-Schuchardt, gs+ and gs were calculated by methacrylate Hohenbrunn, Germany the method of least squares, and gl+ being g l, gd l , gl hm, Urethane UEDMA, U 16-204-59687-F2 Ro known as described earlier Darmstadt, Germany dimethacrylatea (Asmussen et al., 1995). For each liquid mixture, four Triethylene glycol TEGDMA, T 07304-066 Aldrich, similar equations were now dimethacrylate Steinheim, Germany set up, one for each of the solid surfaces, from which gd l, 4-Methacryloyloxyethyl 4-META Bayer, gl and gl+ were calculated, + trimellitate anhydride Dormangen, Germany the values of gd s , gs , gs and g l being known. Additives Propanal PAL 02805584 Merck-Schuchardt, Statistics. The means and Hohenbrunn, Germany standard deviations of the measured contact angles Maleic acid anhydride MAN 8249692 Merck-Schuchardt, (n = 5) and surface tensions Hohenbrunn Germany (n = 3) of the test liquids were calculated. Likewise, a Described by Ruyter (1985). means and standard + deviations of gd s , gs and gs were calculated. The and UEDMA, with the molar amounts of UEDMA given calculated values of gd , g and g + are not amenable to l l l in Table 3 together with the results. The second series common statistical methods, because of the inherent of liquids consisted of HEMA and UEDMA in the complexity of the method used to calculate these proportion 1/3 mol/mol, and of propanal in the molar indirectly obtained observations (Hald, 1952). Instead, amounts specied in Table 3. The third series consisted for each of the experimental liquids, and for one of of HEMA and UEDMA in the proportion 1/3 mol/mol, the four solid surfaces at a time, the surface tension and of maleic acid anhydride in the molar amounts and the contact angles used to calculate gd , g and g + , l l l specied in Table 3. The fourth series consisted of and gd , g + and g , were changed by one standard s s s TEGDMA and UEDMA in the proportion 1/3 mol/mol, deviation, and gd , g and g + were recalculated. l l l and of maleic acid anhydride in the molar amounts The difference between these values and the rst specied in Table 3. The fth series consisted of calculated set of values is an estimate of the variTEGDMA and UEDMA in the proportion 1/3 mol/mol, ability of the results obtained with the experimental and of 4-META in the molar amounts specied in Table 3. liquids. Measurements. Each of the test liquids was placed in a The relationships between contact angles and comtensiometer (model T50, Kru ss, Hamburg, Germany), position, and between surface tension and composition, and the surface tension was determined at 22C. For were studied by regression analyses (Hald, 1952). each liquid three determinations were carried out, Comparisons between values were carried out by and the mean value calculated. Small drops of the means of NeumanKeuls multiple range test at a level liquids were then placed on the four solid surfaces of statistical signicance of p = 0.05. specied in Table 1, and the contact angles were determined by means of a goniometer (model G10, RESULTS Kru ss, Hamburg, Germany). The measurements were carried out at an ambient room temperature of 22 Measurements. The results of the measurements are 1C. The advancing contact angles on the right- and presented in Tables 3 and 4. left-hand side of a drop were measured and averaged Table 4 gives the mean values and standard TABLE 2: LIST OF COMPONENTS OF THE LIQUIDS USED IN THE STUDY
Dental Materials/January 1998 23

TABLE 3: CONTACT ANGLES () AND SURFACE TENSION (mJ m 2). MEAN AND STANDARD DEVIATION (IN PARENTHESES)
Liquid mol% Remanium
1

Solid n=5 Biodent 19.5 0.5e 24.5 0.7h 25.6 0.6h 28.7 0.7i 33.8 1.0l 39.6 1.0n 31.4 0.9i 31.9 0.8i 31.8 0.8i 30.7 1.0f,g,h,i 29.9 0.6d,e,f,g,h 31.4 0.9l 24.8 0.6k 23.1 0.4h,i 23.0 0.5h,i 21.5 0.7f 29.4 3.1f 22.7 1.8e 23.0 3.0e 18.5 1.8d 17.8 2.3c,d 29.4 3.1e 27.6 1.6d,e 24.1 2.0b,c 25.8 1.6c,d Herculite 5.9 0.6a 10.6 0.7b 14.2 0.7c 20.5 0.9f 25.0 0.7h 31.7 1.2k 22.6 0.8b,c 23.6 0.7c 22.0 0.7a,b 20.6 1.1a 20.9 0.8a 22.6 0.8g 15.5 0.6c 16.6 0.7d 12.7 0.4b 11.3 0.7a 17.0 1.9c,d 14.6 2.0b,c 13.2 1.8b 13.5 1.5b 9.7 0.7a 17.0 1.9a 14.2 1.6a 14.0 2.0a 14.7 2.2a Paralm 58.7 0.8o 60.4 0.8p 59.2 0.8o 58.8 1.7o 60.5 1.0p 61.6 0.9p 62.0 1.0m 60.5 1.2k,l 62.2 0.9m 61.1 0.8l,m 59.5 0.5k 62.0 1.0n 61.4 0.8m,n 61.2 0.9m,n 60.2 0.8m 61.6 1.0n 62.5 1.4g 60.6 1.4g 60.5 1.8g 61.7 1.6g 60.8 2.3g 62.5 1.4f 60.4 1.7f 59.9 1.6f 59.2 2.1f 16.2 1.2d 22.0 0.6g 25.4 0.6h 28.0 0.9i 32.5 0.9k 37.2 0.9m 31.5 0.8i 29.5 0.7d,e,f,g 29.4 0.9d,e,f 28.4 0.9d 29.0 0.8d,e 31.5 0.8l 24.0 1.4i,k 19.2 0.7e 18.7 1.1e 19.8 1.2e 28.8 2.1f 22.9 1.9e 24.0 4.5e 24.9 1.9e 19.8 1.8d 28.8 2.1d,e 23.4 2.4b,c 21.1 2.3b 23.1 3.2b,c

gl (n = 3) 37.4 0.1a 38.5 0.1b 38.7 0.0c 39.4 0.1d 39.6 0.1e 39.9 0.1f 39.5 0.1d 39.2 0.1c 38.8 0.0b 39.5 0.1d 38.4 0.1a 39.5 0.1a 40.1 0.1b 39.7 0.1a 40.3 0.0b 39.6 0.1a 39.3 0.1b 39.0 0.1a 39.6 0.1c 39.9 0.1d 40.5 0.1e 39.3 0.1a 39.9 0.1b 40.2 0.1c 40.4 0.1c

H/U=0

H/U=20 H/U=40 H/U=60 H/U=80 H/U=100

2

H/U/PAL=0

H/U/PAL=5 H/U/PAL=10 H/U/PAL=15 H/U/PAL=20

3

H/U/MAN=0

H/U/MAN=10 H/U/MAN=20 H/U/MAN=30 H/U/MAN=40

4

T/U/MAN=0

T/U/MAN=10 T/U/MAN=20 T/U/MAN=30 T/U/MAN=40

5

T/U/4-META=0

T/U/4-META=5 T/U/4-META=10 T/U/4-META=15

1,2,3,4,5

In the ve series of measurements the numbers in the left column indicate the content of the last compound in the stated composition. 2H/U=1/ 3mol/mol. 3H/U=1/3mol/mol. 4T/U=1/3mol/mol. 5T/U=1/3mol/mol. The abbreviations used above are: H=HEMA=2-hydroxyethyl methacrylate; U=UEDMA=urethanedimethacrylate; T=TEGDMA=triethyleneglycol dimethacrylate; PAL=propanol; MAN=maleic acid anhydride; 4-META=4=methacryloyloxyethyl trimellatate anhydride. In each of the ve groups 1,2,3,4,5, values of contact angle and values of surface tension with the same superscript letter were not statistically different at p=0.05.

24 Asmussen & Peutzfeldt/Surface energy of monomers

TABLE 4: CONTACT ANGLES (). MEAN AND STANDARD DEVIATION (N=5)

Liquid Remanium Water Glycerol Ethylene glycol 26.71.6c 37.51.3e 31.23.1d Solid Biodent Herculite

TABLE 5: COEFFICIENTS OF CORRELATION: COMPOSITION OF LIQUIDS VERSUS CONTACT ANGLES AND SURFACE TENSION
Liquid Remanium H/U 0.995*** -0.826* -0.844* -0.775* -0.760 Contact angles on Biodent Herculite 0.981*** -0.792* -0.878* -0.936** -0.808 0.996*** -0.824* -0.917** -0.940** -0.663 gl Paralm 0.668* -0.627 -0.471 -0.424 -0.945* 0.964*** -0.631 0.184 0.904** 0.969*

Paralm

10.30.8a 49.51.5g 102.6;1.0k 25.11.7c 46.72.4f 92.01.2i H/U/PAL 22.00.8b 21.71.5b 81.81.4h H/U/MAN 20.51.6b 8.70.7a 25.70.7c T/U/MAN

1-Bromonaphthalene 22.10.9b

Values with the same superscript letter were not statistically different at p<0.05.

T/U/4-META

deviations of contact angles measured with the reference liquids on the solid surfaces. It appears from Table 4 that statistically signicant differences existed between the investigated liquids and solids with respect to contact angle, reecting the differences in surface energy parameters, as discussed below. Table 3 gives the mean values and standard deviations of the results obtained with the monomeric mixtures. Statistically signicant differences between the liquids were observed. Increasing the amount of UEDMA in HEMA/UEDMA mixtures resulted in increasing contact angles and surface tension. Addition of propionic aldehyde to a HEMA/UEDMA mixture tended to reduce contact angles and surface tension, but the inuence was small. A much stronger inuence on contact angles was exerted by the addition of maleic acid anhydride to HEMA/UEDMA or TEGDMA/UEDMA mixtures, where signicant reductions in contact angles were measured on Remanium, Biodent and Herculite. The same effect was observed with 4-META. Statistics. The results were subjected to linear regression analysis. For each solid surface the inuence of UEDMA, PAL, MAN, MAN and 4-META in the respective ve series of monomer mixtures was analyzed. The same analyses were carried out on the surface tensions. The coefcients of correlation are presented in Table 5 together with the levels of signicance. In most cases the coefcients were signicant at a level of signicance of p 0.05 or less. In some cases, however, the inuence of the variable was too small to give a signicant change in contact angle or surface tension. In the case of 4-META, in spite of relatively large (negative) coefcients of correlations, the number of degrees of freedom was insufcient to make the changes in contact angle signicant. This notwithstanding, in all cases the values on the regression line were calculated and used in the determination of the surface free-energy components characteristic of the monomeric mixtures. Calculations. The results of the calculations are presented in Table 6 and in Figs. 15. From Table 6 it can be seen that the surface free-energy components of the four solid surfaces differ with statistical

TABLE 6: SURFACE FREE-ENERGY COMPONENTS (mJ m 2). MEAN AND STANDARD DEVIATION
Remanium Ysd Ys+ Ys 39.8 0.3a 0.17 0.05b 53.8 1.4d Biodent 40.0 0.3a 0.54 0.06d 58.3 0.8c Herculite 43.5 0.2b Paralm 39.6 0.4a

0.32 0.08c 0.00 0.0a 27.1 1.6b 0.9 0.4a

For each of the three components of the surface free-energy, values with the same superscript letter were not statistically different at p=0.05.

signicance. In particular, considerable differences are present as regards the acid and base components. Figs. 15 give the results of the calculations for the ve series of liquids that were investigated. It can be seen that considerable changes in the components of surface free-energy resulted from the changes in composition. In general, relatively small changes in gd l were observed, with values in the range of 20.3 23.3 mJ m 2. Changes that were relatively more signicant were observed in the polar components of the mixtures, except for formulations containing PAL. In particular, HEMA and the anhydrides MAN and 4-META gave rise to signicant increases in the acid component gl+ of the monomer mixtures. As regards the precision of the calculated surface free-energy components, the estimations showed variations in the values presented in Figs. 15 that were at 2 2 most 0.1 mJ m 2 for gd l , 5.2 mJ m for gl , 0.4 mJ m for ab + 2 for gl , and in most cases gl and 0.08 mJ m considerably less.

DISCUSSION
The adherence between a resin cement and a restoration depends on many factors. For a given level of surface
Dental Materials/January 1998 25

Fig. 1. Surface free-energy components of HEMA/UEDMA mixtures in relation to the content of UEDMA.

Fig. 4. Surface free-energy components of TEGDMA/UEDMA/MAN mixtures in relation to the content of MAN. The ratio TEGDMA/UEDMA was 1/3 mol/mol.

Fig. 2. Surface free-energy components of HEMA/UEDMA/PAL mixtures in relation to the content of PAL. The ratio HEMA/UEDMA was 1/3 mol/mol.

Fig. 5. Surface free-energy components of TEGDMA/UEDMA/4-META mixtures in relation to the content of 4-META. The ratio TEGDMA/UEDMA was 1/3 mol/mol.

Fig. 3. Surface free-energy components of HEMA/UEDMA/MAN mixtures in relation to the content of MAN. The ratio HEMA/UEDMA was 1/3 mol/mol.

roughness, the adherence energy has been calculated to increase with the energy-dissipating properties of the adhesive and with the thermodynamic work of adhesion, WA (Gent and Schultz, 1972). In the present work only factors related to WA and to the interactions between adhesive and adherend were considered. The method. The results will, of course, depend on the extent to which Eq. (1) describes the interactions across the interphase in a correct manner. It is, however, obvious that this equation has its ab limitations. According to theory, g l = gd l + gl , but the d ab sum of the calculated values for gl and gl in all cases falls short of the measured g l. For this reason the + ab calculated gd l , gl , gl and gl can be considered only as relative values. Further, the maximum variation ab and gl+ presented above may seem in gd l , gl , gl somewhat large. But the variation is common to all values of a component of the surface free-energy, so that only the level, not the general shape of the curves in Figs. 15, is subject to variation.

26 Asmussen & Peutzfeldt/Surface energy of monomers

Fig. 6. The work of adhesion of TEGDMA/UEDMA/MAN mixtures in relation to the content of MAN. The ratio TEGDMA/UEDMA was 1/3 mol/mol. The adherends were: R, Remanium; B, Biodent; and H, Herculite.

The results. The results obtained with the solid surfaces agree with earlier data in that gd s does not vary much and is close to 40 mJ m 2 (Asmussen et al., 1995) and that the surfaces except Paralm have a relatively large base component gs and a relatively small acid component gs+ . The results obtained earlier with the in the range of 7.7 monomeric liquids found gab l 10.0 mJ m 2, whereas the present study found gab l in the range of 3.99.6 mJ m 2. This is a fair agreement, considering the differences in the composition of the liquids and in the method. An earlier study found HEMA to have higher fractional polarity than UEDMA (Asmussen and Uno, 1993), which is in agreement with the data presented in Fig. 1. On the other hand, the relatively high fractional polarity of PAL, determined earlier, is not reected in the present results in which rather a small decrease in gab l was observed (Fig. 2). No completely satisfactory explanation for this disparity can be given at present. It is possible that the content of PAL was too small for the effect to be observed, or that the polar substances PAL and HEMA were associated in the solution in such a manner that the polarity is obscured, or that PAL was associated with the NH group of the urethane. Anhydrides are known as compounds with a considerable polarity (Asmussen and Uno, 1993; Asmussen et al., 1995). This is in agreement with Fig. 4 but is less clear in Figs. 3, and 5, in which only the acid component gl+ increases while the base component gl and the overall polarity gab l are constant or decrease. Several tentative explanations may be offered to account for this behavior. In Fig. 3 the monomer contains HEMA, as opposed to the monomers in Figs. 4 and 5, having TEGDMA as diluent monomer. By reaction of the alcohol HEMA with the anhydride MAN, the polarity of HEMA is reduced and MAN is converted to an acid. This would cause a reduction in overall polarity, while increasing the acid component gl+ . In the case of Fig. 5, it is possible that the anhydride has, in part, been converted to the acid by ambient humidity.

However that may be, if we take a look at the potential for adhesion of the monomeric mixtures of Figs. 15, this potential is closely linked to the work of adhesion WA, as mentioned in Section 1. When WA is calculated for the ve series of monomeric mixtures placed on Remanium, Biodent or Herculite, and WA is plotted against composition of the mixtures, the three solid surfaces are found to behave in a similar manner, as evidenced by Fig. 6. This means that the three surfaces may be taken together in a discussion of WA of the monomeric mixtures. The calculations show that WA increases as UEDMA is replaced by HEMA. This is in agreement with an earlier study on the adherence to an NiCr alloy (Asmussen et al., 1995). The calculations also show that not much benet can be expected from addition of PAL to resin cements. On the other hand, the considerable adherence found in earlier studies with MAN-containing and in particular with 4-META-containing resin cements (Asmussen et al., 1995; Sadoun and Asmussen, 1994; Watanabe et al., 1988) as compared with conventional resin cements, is in corroborated by the calculations. As an example, the calculations with TEGDMA/UEDMA/MAN are presented in Fig. 6. The calculations with HEMA/UEDMA/MAN and TEGDMA/UEDMA/4-META give quite similar relationships. From a comparison between Figs. 35 and Fig. 6 it can be seen that in particular it is the acid component of the surface free-energy, gl+ , that accounts for the increase in WA. The reason is that the solid surfaces (excepting Paralm) have large gs and small gs+ . Accordingly, since gl+ and gs enter as a product in the expression for W A, even small increases in gl+ will affect WA in a positive way. These ndings suggest high values of gl+ as a route to increase the adhesiveness of resin cements to metal, porcelain or resin composite.

ACKNOWLEDGMENTS
The authors wish to acknowledge FUT-Calcin Fonden for nancial support to purchase the apparatus for measuring contact angles.
Received April 22, 1997 / Accepted December 19, 1997 Address correspondence and reprint requests to: E. Asmussen Department of Dental Materials School of Dentistry, Faculty of Health Sciences University of Copenhagen 20 Norre Alle DK-2200 Copenhagen N Denmark. Tel: 00 45 35326580; Fax: 00 45 35326505; E-mail: erik.asmussen@odont.leu.dk

REFERENCES
Asmussen E, Attal J-P, Degrange M (1993). Adherence of resin-based luting agents assessed by the energy of fracture. Acta Odontol Scand 51:235240.
Dental Materials/January 1998 27

Asmussen E, Uno S (1993). Solubility parameters, fractional polarities, and bond strengths of some intermediary resins used in dentistry. J Dent Res 72:558565. Asmussen E, Attal J-P, Degrange M (1995). Factors affecting the adherence energy of experimental resin cements bonded to a nickelchromium alloy. J Dent Res 74:715720. Atta MO, Smith BGN, Brown D (1990). Bond strengths of three chemical adhesive cements adhered to a nickelchromium alloy for direct bonded retainers. J Prosthet Dent 63:137143. Constanzo PM, Giese RF, Van Oss CJ (1990). Determination of the acidbase characteristics of clay mineral surfaces by contact angle measurements implications for the adsorption of organic solutes from aqueous media. J Adhesion Sci Technol 4:267275. Flood AM, Brockhurst P, Harcourt JK (1989). The bond strength of various adhesives used for Maryland bridges. Aust Dent J 34:449453. Fowkes FM (1990). Quantitative characterization of the acidbase properties of solvents, polymers, and inorganic surfaces. J Adhesion Sci Technol 4:669691. Gent AN, Schultz J (1972). Effect of wetting liquids on the strength of adhesion of viscoelastic materials. J Adhesion 3:281294. Hald A (1952). Statistical theory with engineering applications. New York: John Wiley and Sons, Inc.:52284. Lee S-Y, Greener EH, Menis DL (1995). Detection of leached moieties from dental composites in uids simulating food and saliva. Dent Mater 11:348353. McCabe JF, Rusby S (1994). Dentine bonding the effect of pre-curing the bonding resin. Br Dent J 176:333336. Nakabayashi N, Kojima K, Masuhara E (1982). The promotion of adhesion by the inltration of monomers into tooth substrates. J Biomed Mater Res 71:256273. Peutzfeldt A, Asmussen E (1992). Inuence of aldehydes on selected mechanical properties of resin composites. J Dent Res 71:15221524.

Ruyter IE (1985). Monomer systems and polymerization. In: Vanherle G, Smith DC, editors. Posterior composite resin dental restorative materials. The Netherlands: Peter Szulc Publishing Co.:10935. Sadoun M, Asmussen E, (1994. Bonding of resin cements to an aluminous ceramic: a new surface treatment. Dent Mater 10:185189. Saunders WP (1987). The effect of fatigue impact forces upon the retention of various designs of resin retained bridgework. Dent Mater 3:8589. Takeyama M, Kashibuchi S, Nakabayashi N, Masuhara E (1978). Studies on dental self-curing resins (17). Adhesion of PMMA with bovine enamel or dental alloys. J Jpn Soc Dent Apparatus Mater 19:179185. Tanaka T, Atsuta M, Uchiyama Y, Kawashima I (1979). Pitting corrosion for retaining acrylic resin facings. J Prosthet Dent 42:282291. Tanaka T, Nagata K, Takeyama M, Atsuta M, Nakabayashi N, Masuhara E (1981). 4-META opaque resin a new resin strongly adhesive to nickel chromium alloy. J Dent Res 60:16971706. Tanaka T, Fujiyama E, Shimazu H, Takaki A, Atsuta M (1986). Surface treatment of nonprecious alloys for adhesion-xed partial dentures. J Prosthet Dent 55:456462. Tanaka T, Atsuta M, Nakabayashi N, Masuhara E (1988). Surface treatment of gold alloys for adhesion. J Prosthet Dent 60:271279. Thompson VP, del Castillo E, Livaditis GJ (1983). Resin bonded retainers. Part I: Resin bond to electrolytically etched nonprecious alloys. J Prosthet Dent 50:771779. van der Veen JH, Bronsdijk AE, Slagter AP, Poel ACM, Arends J (1988). Tensile bond strength of Comspan resin to six differently treated metal surfaces. Dent Mater 4:272277. Watanabe F, Powers JM, Lorey RE (1988). In vitro bonding of prosthodontic adhesives to dental alloys. J Dent Res 67:479483. Zidan O, Asmussen E, Jorgensen KD (1980). Correlation between tensile and bond strength of composite resins. Scand J Dent Res 88:348351

28 Asmussen & Peutzfeldt/Surface energy of monomers