Fig. 24.3 A mixing device that can be used for mixing a number of materials including dental cements. Mixing is achieved by rapid shaking of a capsule. Typically glass ionomer cements as shown in Fig. 24.2 are mixed on this sort of instrument and mixing times may vary between 5 and 15 seconds.

Fig. 24.4 A rotary mixing device which can be used as an alternative to the shaking device shown in Fig. 24.3. In this instrument the capsule is rapidly rotated and centrifuged and the aim is to attempt to produce a mix which contains lower levels of porosity.

mixing by rotating and centrifuging the fluid cement in order to eliminate as much air as possible from the material. Such a rotary mixer is shown in Fig. 24.4. Cements of a less fluid (higher viscosity) nature are less prone to ‘froth’ formation and are therefore likely to have lower levels of porosity following mechanical mixing. The levels of porosity trapped inside a viscous cement is similar for hand mixing and mechanical mixing whilst for fluid cements hand mixing results in lower porosity than mechanical mixing.

Fig. 24.5 Structural formula of (a) polyacrylic acid and

(b) its cross-linking through calcium and aluminium ions.

24.3 Setting reaction

The structure of polyacrylic acid is shown in Fig. 24.5a. It consists of repeating units derived from acrylic acid with reactive carboxylic acid groups at alternate carbon atoms along the polymer chain. Polymaleic acid has a similar structure except that there are acid groups at every carbon atom on the polymer chain. Hence, for a given chain length there are twice as many carboxylic acid groups in polymaleic acid compared with polyacrylic acid. The setting reaction involves the formation of a salt through reaction of the acid groups with cations released from the surface of the glass. The nature of the cross-linked polyalkenoate salt is illustrated in Fig. 24.5b.

On mixing the powder and liquid or powder and water the acid slowly degrades the outer layers of the glass particles releasing Ca2+ and Al3+ ions. During the early stages of setting, Ca2+ is released more rapidly and is primarily responsible for reacting with the polyacid to form a reaction product akin to that shown in Fig. 24.5. Al3+ is released more slowly and becomes involved in setting at a later stage, often referred to as a secondary reaction stage. The set material consists of unreacted glass cores embedded in matrix of cross-linked polyacid. The stages of setting are illustrated in Fig. 24.6. The matrix region is composed of the salt reaction product shown in Fig. 24.5b. The second stage of the setting reaction involves the incorporation of significant quantities of aluminium in the matrix structure and results

in a marked maturation of the physical properties of the material. Prior to this stage, the materials remain very weak and soluble. In order to ensure that the reaction proceeds to full maturity it is essential that the setting cement is protected from excessive moisture contamination since the presence of disproportionate quantities of water at this stage can interfere with salt formation.

The presence of tartaric acid plays a significant part in controlling the setting characteristics of the material. It helps to break down the surface layers of the glass particles, rapidly liberating aluminium ions with which it undergoes complex formation. Hence the aluminium ions are not immediately available for reaction with the polyacid so the working time of the cement is maintained. The initial onset of setting is further inhibited by the tartaric acid preventing unwinding and ionization of the polyacid chains. When the concentration of solubilized aluminium reaches a certain level the second stage of the setting reaction proceeds rapidly. The tartaric acid aids complex formation between the polyacid and the trivalent aluminium ions by overcoming steric hindrance problems which may occur when an aluminium ion attempts salt formation with three carboxylic acid groups. Hence many of the aluminium salt links consist of an aluminium ion bound to two carboxylate groups and one tartrate group. This mechanism is supported by the fact that there is very little unbound tartaric acid left in the set cement. The release of fluoride ions from the glass particles results in the matrix phase of the set material becoming a reservoir for fluoride. After setting the matrix is able to release this fluoride

into the surrounding environment or to absorb fluoride from the surroundings when the ambient fluoride concentration is high (e.g. from a fluoride containing toothpaste). In addition to the potential therapeutic effects of the fluoride concentrated in the matrix phase, its presence is also thought to contribute towards optimizing the setting characteristics by maintaining workability for a longer period followed by a relatively sharp increase in viscosity.

24.4 Properties

Some of the property requirements of glass ionomer cements are embodied in the ISO Standard for dental water-based cements (ISO 9917). These requirements are given in Table 24.2 along with those for the glass ionomer luting materials and cavity lining/base materials. Since glass ionomers and composites may be considered for similar applications – both being tooth coloured restorative materials – their properties are compared qualitatively in Table 25.1.

Like many dental cements the properties of glass ionomers are critically dependent upon the powder/liquid ratio. Unfortunately hand mixing at optimal powder/liquid ratios may result in a dry and apparently crumbly mix which dentists do not like. Hence there is a tendency for dentists to add too much liquid to give a wetter consistency with a deleterious effect on the physical properties of the material. This problem is surmounted by the use of encapsulation and mechanical mixing.

The powder/liquid ratio should be high in order to optimize strength and solubility, but there

should be sufficient free polyacid available to form a bond with tooth substance. The materials are often difficult to mix at the ratios recommended (typically around 3 : 1 by weight). The powder/ water materials tend to be easier to handle than the powder/liquid products. Aqueous solutions of polyacids should not be stored in a refrigerator since this may initiate crystallization. The variability of material properties with powder/liquid ratio and the difficulties of mixing by hand suggest there are definite advantages to be gained by using the materials in encapsulated form. Here the proportions are fixed by the manufacturer and mixing takes only a few seconds in an electrically-powered mixer of the type available in most dental surgeries.

The setting reaction is rather protracted despite a fairly rapid initial hardening. The material must be protected from moisture contamination during the first hour, otherwise strength and solubility are adversely affected. Hence it is necessary to varnish the surface of the filling immediately after initial hardening. The varnishes used consist of a water resistant resin dissolved in a volatile solvent such as ether or ethylacetate. These varnishes can be expected to afford protection to the glass ionomer for a variable period of time, from a few seconds to an hour or more depending on how quickly they become dislodged. A more long-term protective effect can be achieved by using a resinbonding agent or fissure sealant of the type described in Chapter 23.

One of the most important properties of these materials is their ability to adhere to both enamel and dentine, although the precise mechanism of bonding is still somewhat unclear. One theory is that polyacid molecules chelate with calcium at the tooth surface (Fig. 24.7). Support for this mechanism stems from the fact that the formation

Fig. 24.7 Diagram illustrating one possible mechanism of bonding to tooth substance for glass ionomer cements.

of the interfacial calcium polyalkenoate salt would involve a reaction similar to that thought to occur during the setting of the cement. Also, the significantly greater bond strength achieved with enamel than with dentine suggests that it is the calcium of tooth substance which is involved in bond formation. Another theory is that the outer layers of the apatite on the tooth surface become solubilized in the presence of acid. As more apatite dissolves and as the cement begins to set the pH begins to rise. This may cause reprecipitation of a complex mixture of calcium phosphate (from the apatite) and calcium salts of the polyacid onto the tooth surface. Thus the polyacid chains would be bound to a reprecipitated layer on the tooth surface. This mechanism could operate on enamel or dentine surfaces and could thus also be supported by bond strength data. It has been suggested that in the case of dentine there may also be some bonding between carboxylic acid groups of the cement and reactive groups within collagen, either by hydrogen bonding or by metallic ion bridging.

Various surface treatments have been used in an attempt to improve the strength and reliability of the bond between glass ionomer cements and dentine. One feature of dentine is the presence of a smear layer, as mentioned in Chapter 23. Many

suggested surface treatments attempt to remove the smear layer or dissolve and reprecipitate it in order to give better binding to the underlying dentine. Citric acid solutions have been advocated for ‘cavity cleansing’ and their use will almost certainly lead to the disruption of the smear layer. The use of citric acid produces no improvement in bond strength however, probably due to the fact that the dentine surface is partially demineralized and starved of calcium required for bond formation. Treatment of the surface with a solution of polyacrylic acid has been found to be more effective in improving bond strengths. The mode of action is unclear but probably involves dissolution and reprecipitation of the smear layer. Another attempt to improve bond strengths to dentine involved artificially increasing the mineral content of the dentine surface by applying solutions from which calcium phosphate or calcium sulphate could crystallize to give a surface layer which is potentially more reactive towards polyacids.

There are limits to which the bond strength of glass ionomer cements to tooth substance can be improved. The tensile strength of the cement itself is only around 12 MPa and during tensile bond strength testing with enamel failure often occurs cohesively within the cement at a stress of around 5 MPa. This suggests that there is a zone of stress concentration within the cement which is set up as a result of bonding to the tooth. Hence it would appear that there is only limited scope for improving the bond strength of the cements to dentine (normally measured at around 1–3 MPa). Certainly there seems little scope for increasing the bond strength to values approaching that for resins to acid-etched enamel (16–20 MPa).

Most authorities consider glass ionomers to have acceptable biocompatibility despite their acidic nature. Attitudes towards the tolerance of acids by the dental pulp are changing rapidly as outlined in Chapter 23. However, the cements based on polyacids have traditionally been considered less harmful than those based on phosphoric acid for two reasons. First, the acids used are weaker acids than phosphoric acid. Second, the polyacid chains are large and immobile, their mobility being further restricted by their affinity for calcium ions in the tooth to which the material is applied. Pulpal studies indicate that glass ionomer cements cause a mild inflammatory response which is normally resolved within 30 days. The response is moderated according to the

thickness of residual dentine. Only in very deep cavities having a thin residual layer of dentine is it considered necessary to use a cavity lining. In these cases, the lining of choice is normally one of the calcium hydroxide materials. The amount of calcium hydroxide material used is limited to that required to just cover any dentine which is considered close to the pulp. Covering large areas of the dentine with calcium hydroxide cement would negate any beneficial effect which could be gained by creating adhesion between the glass ionomer and dentine. These views will almost certainly be modified as a greater understanding of acid tolerance by the pulp is obtained.

The raw materials from which the glasses used in the products are derived sometimes contain small quantities of arsenic and lead. In appreciation of the potential harmful effects of such heavy metals the ISO Standard allows only 2 parts per million of acid soluble arsenic and 100 parts per million of acid soluble lead in the cements.

The thermal diffusivity value for glass ionomer cements is close to that for dentine. Hence the material has an adequate thermal insulating effect on the pulp and helps to protect it from thermal trauma. Although the setting reaction for the materials is exothermic it is considered that the temperature rise created for an average sized cavity (about 2–5ºC) would not be great enough to cause serious damage.

The extent to which a material expands and contracts when subjected to hot and cold stimuli is characterised by the value of coefficient of thermal expansion (Section 2.6). This is thought to be one of the parameters which affects the quality of the seal and bond between a restorative material and the tooth. Ideally, values for tooth and material should match as closely as possible. It is difficult to measure the coefficient of thermal expansion for a sample of glass-ionomer because of the presence of water within the structure of the set cement. On heating a sample of the cement under dry conditions the sample undergoes marked shrinkage as the cement desiccates. Under moist conditions the expected expansion on heating is compensated by an accelerated flow of water from within the cement into the surrounding wet environment. On cooling, the expected contraction is compensated by water uptake. Hence, the lack of any noticeable change in dimensions over a range of normal intra-oral temperatures under moist

conditions may be considered as a type of ‘smart’ behaviour being exhibited by the cement.

The matrix phase of a set glass ionomer cement contains significant amounts of fluoride ions which are fairly mobile since they are not involved in salt formation. The mobile fluoride ions readily diffuse to the surface of the cement where they may be washed away with saliva or undergo reactions with the surrounding tooth substance. Fluoride ions may replace hydroxy groups in the apatite structure and this change renders the apatite more resistant to acid attack. The presence of glass ionomer cements is likely therefore to reduce the chance of caries developing in the surrounding tooth substance. The cement can be considered as applying a long term topical fluoridation effect on the tooth substance with which it is in contact.

As stated earlier, not only can the matrix phase of the cement gradually release fluoride but it can also absorb fluoride from an aqueous medium which has a high fluoride concentration. Hence the level of fluoride in the cement can be ‘topped up’ as it absorbs ions released from toothpastes, mouthwashes, and drinking water. The amount of fluoride release required in order to give a beneficial therapeutic effect is not known – a fact which makes product optimization difficult. Also, in the UK, manufacturers rarely claim a therapeutic effect for glass-ionomer products, for to do so may result in the materials being classified as pharmaceutical products.

The glass ionomers are relatively brittle, having a flexural strength of only 15–20 MPa (cf >70 MPa for composites), and cannot be considered suitable as general-purpose filling materials for permanent teeth but more to answer a specific need for certain applications. The brittleness of the material, for example, would preclude the use of these products for restoring fractured incisal edges. Cavity designs for glass ionomer restorations should ideally avoid the production of thin sections of the material such as would occur in a filling with a knife-edge margin.

The tests required in the ISO Standard are unable to predict these problems as only compressive strength is measured (Table 24.2). Brittle materials often have relatively high values of compressive strength, but fail at much lower stress in tension or when subjected to a flexural stress. The difference between the compressive strength value for restorative cements (130 MPa minimum) and luting or lining/base cements (70 MPa minimum)

is mostly due to the different powder/liquid ratios used. Whereas a ratio of about 3 : 1 by weight is normal for a restorative cement the ratio for luting and lining/base cements is closer to 1.5 : 1. This marked effect of powder/liquid ratio on mechanical properties highlights the need to achieve the ratio recommended by the manufacturer.

In addition to powder/liquid ratio and the extent to which the setting reaction is completed, the factor which most affects strength is porosity. Porosity is incorporated during mixing, as described earlier. Care should be taken during hand mixing to minimize trapped air. During mechanical mixing of encapsulated materials, air bubbles may be formed by ‘frothing’ of fluid cements and this may produce levels of porosity well in excess of levels produced by hand mixing. In more viscous cements, commonly used for restorative purposes, frothing and bubble formation is less common. Alternative mixing regimes, as described earlier, are available to reduce porosity.

The glass ionomers have relatively poor abrasion resistance which results not only in a change of anatomical form but also in considerable surface roughening. Roughening may also be caused by polishing, leaving hard glass cores protruding from the softer matrix material.

Glass ionomers, like their close relatives the silicates, are susceptible to acid erosion. This is an inherent property in materials which are effectively inorganic salts formed by an acid–base reaction. Factors which affect the rate of erosion include the pH of the eroding medium (e.g. saliva, plaque or drinks) and the maturity of the cement at the time it contacts the acid. The worst possible conditions are when a freshly placed cement is bathed in a highly acidic fluid such as fruit juice. To reduce the risk of acid erosion it is essential that cements are protected with varnish or resin during setting. The situation is further improved if the material has a short setting time. Also, the cement should be used at the correct powder/ liquid ratio, which is pre-set in the case of encapsulated materials.

The processes of acid erosion and wear are somewhat synergistic. The surface of a cement which is softened and undermined by acid may be more readily broken down by wear.

In the ISO Standard (ISO 9917) the resistance to acid erosion is determined using the jet test. This involves preparing a sample of the cement in a mould with one surface of the cement exposed