Dental abrasion as a cutting process rsfs.royalsocietypublishing.org

Research Cite this article: Lucas PW et al. 2016 Dental abrasion as a cutting process. Interface Focus 6: 20160008. http://dx.doi.org/10.1098/rsfs.2016.0008 One contribution of 14 to a theme issue ‘Cutting science in biology and engineering’. Subject Areas: biomechanics, biomaterials Keywords: biomechanics, cutting, dental function, dental wear, mammalian diets Author for correspondence: Peter W. Lucas e-mail: [email protected]

Peter W. Lucas1, Mark Wagner3, Khaled Al-Fadhalah4, Abdulwahab S. Almusallam5, Shaji Michael1, Lidia A. Thai6, David S. Strait7, Michael V. Swain1, Adam van Casteren8, Waleed M. Renno2, Ali Shekeban6, Swapna M. Philip1, Sreeja Saji1 and Anthony G. Atkins9 1 Department of Bioclinical Sciences, Faculty of Dentistry, and 2Department of Anatomy, Faculty of Medicine, Kuwait University, PO Box 24923, Safat 11310, Kuwait 3 Department of Mechanical Engineering, School of Engineering and Applied Science, Science and Engineering Hall, 800 22nd St NW, Washington, DC 20052, USA 4 Department of Mechanical Engineering, 5Department of Chemical Engineering, and 6Nanotechnology Research Facility, College of Engineering and Petroleum, Kuwait University, PO Box 5969, Safat 13060, Kuwait 7 Department of Anthropology, Washington University in St Louis, Campus Box 1114, One Brookings Drive, St Louis, MO 63130-4899, USA 8 Max Planck Weizmann Center for Integrated Archaeology and Anthropology, Max Planck Institute for Evolutionary Anthropology, Deutscher Platz 6, 04103 Leipzig, Germany 9 School of Construction Management and Engineering, University of Reading, Whiteknights, PO Box 219, Reading RG6 6AW, UK

A mammalian tooth is abraded when a sliding contact between a particle and the tooth surface leads to an immediate loss of tooth tissue. Over time, these contacts can lead to wear serious enough to impair the oral processing of food. Both anatomical and physiological mechanisms have evolved in mammals to try to prevent wear, indicating its evolutionary importance, but it is still an established survival threat. Here we consider that many wear marks result from a cutting action whereby the contacting tip(s) of such wear particles acts akin to a tool tip. Recent theoretical developments show that it is possible to estimate the toughness of abraded materials via cutting tests. Here, we report experiments intended to establish the wear resistance of enamel in terms of its toughness and how friction varies. Imaging via atomic force microscopy (AFM) was used to assess the damage involved. Damage ranged from pure plastic deformation to fracture with and without lateral microcracks. Grooves cut with a Berkovich diamond were the most consistent, suggesting that the toughness of enamel in cutting is 244 J m22, which is very high. Friction was higher in the presence of a polyphenolic compound, indicating that this could increase wear potential.

1. Introduction The mouth of mammals is adapted to ingest food items and then fracture them repeatedly in order to reduce their particle sizes. The advantage of this oral processing, which is particularly important in species that feed on chemically protected plant tissue, is that the food surface increase provided by this comminution escalates the rate of nutrient liberation via enzymes later in the gut passage. This helps mammals to service their high metabolic rates [1]. It follows that anything that lowers the efficiency of the mouth in this endeavour has the potential to jeopardize the overall health and survival of an individual. Very much included in such health threats is damage to the teeth via fractures at a wide variety of scales (figure 1a). Loss of tooth tissue is vital to avoid because the teeth of most mammals grow for a limited period and are not replaced in adults [2– 4]. Circumstantial evidence from several mammalian species strongly suggests that the loss of critical features on the working surfaces of the teeth is deleterious to tooth function and to the overall health and survival of afflicted individuals [5– 10]. It is thus logical to suppose that natural selection has acted to minimize threats to the integrity of the working surface of a tooth.

& 2016 The Author(s) Published by the Royal Society. All rights reserved.

(a)

(b) 30

2

enamel margin crack

dentine pulp

toughness (MPa m0.5)

radial crack chipping

natural mineralized composites (e.g. bone, nacre)

20

10

0

0.4 0.2 crack extension (mm)

0.6

Figure 1. (a) Schematic diagram of categories of fracture that may damage a ‘bunodont’ molar tooth crown (shown in longitudinal section to display the tooth interior). (b) The dependence of toughness on crack length in biological mineralized tissues (adapted from [11]). Toughness is here depicted as Kc. Otherwise in this paper, toughness is given as R, where Kc2  ER, where E is the elastic modulus. (Online version in colour.) Most fractures in the type of molar tooth crown found in modern humans and extinct hominins, i.e. one with a lowrelief on the working surface and in which the enamel is thick (termed a ‘bunodont’ crown), appear to start in the enamel (figure 1a). This is the tooth tissue that surfaces the crown in the mouth and makes critical contacts with food. The tissue consists of elongate crystals of hydroxyapatite, 40–70 nm in cross-sectional dimensions, which occupy more than 90% of the tissue volume. These crystals are clumped into long multi-crystalline cylinders called prisms or rods. Prisms, 3–5 mm wide, are separated from each other by a 100 nm gap containing a protein gel that also extends into the 1–2 nm gaps between crystals. The prisms run from close to the junction with the underlying dentine towards the enamel surface. However, in many mammals, their course is wavy. The path of neighbouring prisms is slightly out of phase with each other leading to a structure that looks as though prisms cross each other. This apparent interweaving is called ‘decussation’. Simply put, the high crystal content of enamel provides it with stiffness and hardness, while the remnant protein fragments that survive into the mature tissue [12], coupled with decussation [13], give it toughness. Mechanisms by which this toughening is achieved are still being worked out, but the protein content is crucial [12]. Although enamel is classically thought to be incapable of repair, recent studies suggest that this is possible via movement of the protein gel into cracks so as to heal them [14,15]. Figure 1a shows categories of fracture that may damage a bunodont molar tooth crown. A crack that runs from the outside working tooth surface towards the inside (i.e. towards the junction with dentine) can produce chipping, the crack being deflected back to the surface by decussation of the enamel prisms [13,16]. Inside-to-outside (radial) cracks are very common in modern human teeth and can run straight to the surface [3]. Margin cracks, also common in humans in the form of ‘abfractions’, seem to result from contacts with soft foods that smother the crown. When forces build up sufficiently, margin cracks grow as a result of the internal hoop stresses generated by the dentine–pulp complex [17]. Both radial and margin cracks may actually initiate from preexisting faults in enamel called ‘tufts’. These wavy strands are in-built flaws passing one-third of the way into the enamel in unerupted (unused) teeth [18]. Present in enormous numbers [19], they form a ‘forest’ of pre-cracks that compete

against each other to extend when forces increase. In the process, such competition acts to shield neighbouring tufts from extending through the enamel, so providing damage tolerance to the growth of large cracks [20]. Despite this, ‘through cracks’ called lamellae, which arise from extensions of tufts, are seen in the enamel of erupted teeth [21], increasing with age, but not in the quantities that might be expected without some protective mechanism. Possibly as a result of crack healing and fatigue tolerance provided by tufts, large-scale fractures are rare in teeth compared with the quantity of microscopic (abrasive) fractures that lead to wear. The working surfaces of teeth make thousands of contacts with ingested food particles every day [22] and the heavy wear that can result is common in mammalian teeth, particularly in plant-eaters like ungulates [23,24] and hominins [25–28]. Although plant foods are much less hard than enamel, these food items are generally not clean, being contaminated by a surface coating of dust or grit. The composition of this natural pollution is variable [29], but soil derivatives are probably dominant. Much of the Earth’s crust consists of silicates with a hardness exceeding that of enamel [30]. In certain conditions where the angularity or sharpness (to be defined later) of dust and grit particles is sufficient, contacts with a tooth surface can severely damage it at minute loads, essentially by cutting the enamel [31]. This paper deals with wear—the smallest scale of damage— which leaves a textural fabric on the surface of enamel that has been the subject of an enormous body of research [32–40]. The aim of most of it has been to document diet from an analysis of enamel surface texture [41–50]. Until recently, the mechanics underlying the cause of wear have been neglected [31,51]. However, before continuing, it is worth noting that there must be protective mechanisms against tooth wear, just as there are for other types of fracture. An important source of protection is physiological. Evidence in the human from both isolated particle presentation tests [52,53] and chewing experiments [54] show that minute hard particles can be detected between teeth at correspondingly tiny loads [55–57]. The neural centre in the brain that receives this information is also known [58]. This allows potentially abrasive mouthfuls to be quickly detected and discarded. Another line of defence is enamel toughness, because the higher the toughness, the sharper wear particles have to be before they become capable of abrasion [31].

Interface Focus 6: 20160008

enamel and artificial composites

rsfs.royalsocietypublishing.org

abrasive wear

2. Material and methods Longitudinal sections of a resin-embedded orangutan molar tooth, used in previous papers on this subject [31,84] were employed here. The tests were made in mid-enamel, closer to the working surface than to the enamel– dentine junction. These surfaces have been characterized in terms of nanoindentation with a Berkovich indenter to depths of about 1 mm, and possess an average hardness of 5 GPa and an elastic modulus of 100 GPa [31]. The properties of enamel vary with scale [85], so even smaller contact areas of this surface were scanned via tapping with bimodal nanomechanical spectroscopy. These showed that the modulus was broadly distributed around a mode of 75 GPa [86]. This is consistent with results on human teeth [85]. The tooth surface was re-polished between tests, which were variously conducted air dry, with a salivary coating or with saliva plus a 0.1 M solution of epigallocatechin gallate, a common polyphenolic standard. Three types of cutting experiments were made on the enamel of this tooth. Each had in common that they employed a load cell capable for registering forces in two directions, the cutting force (Fc) and the transverse reaction force, also called the thrust (Ft), simultaneously. By applying Amontons/Coulomb analysis, friction is the ratio of the sliding force (F) divided by the normal force (N) on the rake face of the contacting tip where the coefficient of friction

m¼

F ¼ tanl: N

ð2:1Þ

3

Interface Focus 6: 20160008

angulation of the cutting tip becomes important. Assuming the particle to be rigid compared to the material being cut, then a deeper groove can displace material in one of two ways. A ‘standing wave’ may be created ahead of the particle tip via a process of plastic material displacement that we call ‘prowing’ here [31,65], often called ‘ploughing’ by others. Material builds on either side of the groove as ridges. Alternatively, a ribbon or chips of material are fractured away. The first possibility can be called ‘rubbing’ in contrast to the other, which is abrasion [65]. It needs to be shown that fracture has indeed taken place and that the cut does not just consist of displaced material [71,73]. Complicating factors include the presence of microcracks on either side of the cut, which leads to a larger fracture surface than can be accounted for by the application of formulae restricted to cutting along the track of the blade. In addition, there is a need to ‘subtract’ friction. This is particularly important when conventional ‘Coulomb– Amontons’ estimates tend to show that the coefficient of friction varies with groove depth [71]. However, some theoretical approaches to abrasive wear have ignored such complications, leaning heavily on theory from static tests [74], and these have recently been applied to dental abrasion studies [75]. Here, we examine whether the behaviour of enamel in cutting tests accords with predictions from such cutting theory and whether a toughness estimate for the tissue can be established. Given the importance of an understanding of friction to cutting, we also make a preliminary investigation into the effect of plant polyphenolic compounds on abrasion. There is evidence that interaction between plant polyphenolics and the proline-rich proteins in saliva result in a significant decrease in lubrication and an increase in friction [76]. Friction is an extremely important perceived aspect of plant food texture [77,78]. An increase in friction affects the ease of movement of food around the mouth [79–82] and during swallowing [83], and it is possible too that the polyphenolic–salivary interaction increases the potential for tooth wear [76].

rsfs.royalsocietypublishing.org

A key issue here is shown in figure 1b where biological mineralized tissues such as bone and nacre are distinguished by low resistance to very small cracks [11]. Resistance increases until a plateau at around 0.5 mm crack length where toughness can be more than 10 times that of the smallest cracks. If these ‘R curves’ were found in tooth enamel, then its toughness would not defend it against abrasion because these wear events are very small scale. Potential wear particles themselves can be less than 100 mm in diameter. Features that typically form the basis of dental microwear analysis tend to fall in a size range of one to tens of micrometres in width, with the length of scratches being greater by perhaps an order of magnitude or so, with the depth generally unknown. A recent comprehensive survey of dental microwear in primates sampled tooth surfaces in rectangular areas, 276  204 mm in dimensions, sufficient to reveal tens or hundreds of individual microwear features in each image [59]. At this scale, both bone and nacre would be abraded very rapidly. The key then to anticipating why enamel might behave differently to bone and nacre is to understand that the latter tissues are not exposed to abrasion. Bone is part of an endoskeleton, shielded from surface damage by the soft tissues that cover it. As such, fracture matters less to the integrity of its support than fragmentation—the physical separation of a bone (as an organ) into two parts. Thus, low resistance to microcracking is not so important to bone. Only one type of bone is mounted on the exterior of mammals, and that is antler, which is far tougher than other bone types particularly with respect to small cracks [60]. Tooth enamel is a specialized remnant tissue of a vertebrate exoskeleton that only ‘migrated’ later in evolution to the mouth [61]. Teeth are exoskeletal elements and are by virtue of their function exposed to abrasion, i.e. to the loss of very small pieces of tissue from their working surfaces by correspondingly small cracks. Mollusc shell is also exoskeletal, but not all mollusc species suffer heavy abrasion. When they do, they tend to possess other structural variants rather than nacre on their shell surface. Even a homogeneous shell, often found in mollusc species that get heavily abraded, seems to perform better in tests than nacre, which is a tissue that is rarely exposed anyway [62]. The basic aim of this paper is to establish the toughness of enamel at the scale of wear events. To do this, we employ cutting tests, arguing that, as did [31], a sliding contact of a wear particle on a tooth surface that leads to immediate loss of tooth tissue is akin to a cutting event. The estimation of fracture toughness from cracks produced via static indentation is well established [63] and has been used to examine tooth enamel many times previously. These estimates are uniformly low because the crack tip takes its own course [64]. By contrast, crack paths during a cutting event are dictated by the invasive particle tip and thus toughness values relevant to cutting must be expected to differ from those in static tests. A key advance in the theory of cutting has been the inclusion of a surface term in the form of toughness to account for the separation of material via fracture [65]. Since then, there has been considerable debate in recent years as to whether sliding contacts (scratching) tests can actually be used as a method to estimate the fracture toughness of both ‘brittle’ materials [66–68] and ‘ductile’ polymers [69,70]. Theory shows clearly that the depth of the cut is a critical parameter [71]. At small enough depths, cutting with any particle geometry is liable just to move material plastically in the form of rubbing because of a brittle–ductile transition [72]. At larger depths, the

(a)

(b)

4

spring attaches to micrometer mount maintains probe contact with carriage

support plate

micrometer (anchored to track) controls a blunt probe pushing on carriage

ball-screw

(c)

Fc Ft

support plate

digital microscope

tungsten carbide cutter

tooth specimen

micrometer probe

carriage

(e) (d)

top view

tooth section

tip geometry

side view

vickers indenter

1 mm

20 mm

1 mm

Figure 2. (a) Tester design equipped with a multiaxial load cell (not shown). Panels (b,c) show detail of the positioning of the tooth specimen and the cutting apparatus. (d ) SEM images of the tungsten carbide blades modelled after [70], except that, viewed from above, the blade has a 228 included angle rather than 908. (e) Vickers indenter, tilted 88 to the horizontal, at the start of a test. (Online version in colour.) If the rake angle of the tip is a, then in cutting, Ft ¼ Fc tanðl  aÞ,

ð2:2Þ

where a is measured from the vertical to the plane of the specimen. (An alternative to the rake angle is the attack angle b, which is given in degrees by (90 þ a) [31]). By measuring Fc and Ft at different depths of cut, we could estimate tanl. For the purpose of finding the toughness though, we first applied a friction correction factor [65]    sinlsinf Qshear ¼ 1  , ð2:3Þ cosðl  aÞcosðf  aÞ where f the shear plane angle is given by p cotf ¼ tanðl  aÞ + ½1 þ tan2 ðl  aÞ þ Zftanðl  aÞ þ tanag: ð2:4Þ The term Z ¼ R=ty t [87] includes R, the enamel toughness during cutting (in J m22), which we aimed to evaluate, while ty is the shear yield stress. The force Fc to cut a groove of depth t [71] is      1 tand 2Rt þ , ð2:5Þ Fc ¼ ty t2 cosða  fÞsinf Qshear cosd where d is the semi-point angle of the tip. By plotting Fc/t versus t, it is possible to use equation (2.5) to estimate ty from the slope

of the plot and the toughness R from the intercept. First though, the shear plane angle f must be estimated from iterating equations (2.3) and (2.4) because f depends on Z, a term involving both R and ty, so requiring the relevant plot of Fc/t versus t to help determine the best-fit value of f via a process of minimizing Fc [71]. Two of the experimental sets used a small purpose-built portable mechanical tester (figure 2a). Cutting was displacementcontrolled with a linear variable differential transformer (LVDT) accurate to +2 mm. The tester was fitted with a Nano 17 multiaxial load cell (ATI Industrial Automation, Apex, NC, USA) configured with high gain via a USB-2408 24-bit AD converter (Measurement Computing Corp., Norton, MA, USA). Cutting tracks were made along the side of the tooth crown, which could be 10 mm in length. One set of experiments was run with a tungsten carbide blade design (shown in figure 2c,d) fashioned after [70], but differing via a small 228 included angle. In practice, contact was made at forces of 0.01 –0.2 N in these experiments. This meant that small asperities on the cutting tips, which were far sharper than the 10 mm tip radius shown in figure 2d, cut the grooves in the enamel. As a result of this, no calculations were made from these experiments. However, operated at high b (i.e. high 90 þ a) it was possible to try to detect groove depths associated with transitions between cutting tracks that involved either prowing or fracture. The second experiments employed the same tester fitted with a Vickers indenter, 4 mm in diameter, also made of tungsten

Interface Focus 6: 20160008

LVDT monitors displacement

carriage running on track (low-friction)

rsfs.royalsocietypublishing.org

hand-cranked with steering wheel

24-bit ADC fits here

Figure 3 shows several tracks imaged by AFM that were produced by the cutting tip shown in figure 2d. The type and extent of damage differed. The profile shows that track 2, which is the shallowest at 140 nm, has a clean path associated with prowing. Track 1, 180 nm deep, shows debris formation associated with abrasion, while track 3 (up to 650 nm deep) shows abrasion and lateral microcracking, causing lateral pitting up to 1 mm in depth. Nine such tracks were analysed. Below 180 nm, these tracks looked clear of debris. Between 180 and 400 nm, the tracks showed debris to the lateral sides of the groove. Deeper cuts than 400 nm showed lateral pitting beside the grooves, some of which were deeper than the grooves themselves (as in figure 3). Experiments with the Berkovich tip in the nanoindenter under constant Ft loads produced relatively stable displacements in most tests (figure 4). The cutting force Fc was consistently lower than Ft. A tendency for the enamel to fracture was increasingly evident above 1000 mN as indicated by abrupt drops in Fc, e.g. the arrowed locations for Ft of 4000 mN in figure 4. However, cross-comparison with imaging data before and after washing in figure 5 indicates that all grooves showed fracture, possibly within prowing fields.

2.5 0 2.0

(µm

)

30

(mm)

1.5

60

1.0 90 2.65µm

0.5

1

0

2

30

(µm)

depth (µm)

1

60

3 0

90 3

2

1.0 0.5 0 –0.5 –1.0 –1.5 0

10

20

30 40 50 displacement (µm)

60

70

80

Figure 3. The effect of groove depth with large attack angles with the cutter shown in figure 2d on damage to enamel. (Online version in colour.) Prowing was supported by the persistence of ridges besides the grooves after washing and also by the persistence of the final position of the prow itself at the end of the experimental run (expanded view at top right of figure 5). For the Berkovich tests, the cutting force, normalized to cutting depth in line with equation (2.5), i.e. Fc/t, is shown plotted against t in figure 6. There was a linear relationship with a slope of 0.039 mN nm21 nm21 and an intercept of 1.4 mN nm21. The Ft/Fc ratio was relatively constant across all these experiments, averaging 2.74 (s.d. 0.21). From equation (2.2), given a ¼ 265.38 for a Berkovich tip, l was estimated as 4.78, giving a value for the Coulomb –Amontons coefficient of friction m ¼ 0.082. For an upright Berkovich tip, d ¼ 2 a, and from the process of minimization of Fc, f was found to be 108. Thus, Qshear ¼ 0.836. From the data in figure 6, this leads to an overall estimate for ty and R of enamel in cutting as 1.48 GPa and 244 J m22, respectively. Lastly, the effect of adding polyphenolics to saliva on the cutting of enamel was considered with the Vickers indenter (figure 7). Forces fluctuated, but via equation (2.2), we estimated that the frictional coefficient m ¼ 0.12 for enamel when covered with saliva against a tungsten carbide tip, rose on average to m ¼ 0.27 after the epigallocatechin gallate was added.

4. Discussion Our analysis adds weight to the value of using cutting tests to investigate the toughness of materials. It is surprising, however, that an analysis with ‘dead-weight’ loading, i.e. one where Ft is fixed and the displacement allowed to vary, proved tractable in this regard. It is probably only because the displacement was very stable (figure 5) that the results were so significant (figure 6). The value obtained for enamel toughness, at R ¼ 244 J m22, is surprisingly high— more than 20 times the value of R obtained from smallscale indentation fracture toughness tests. Assuming E ¼ 75 GPa [86], then this equates to a Kc of 4.66 MPa m0.5, which is a value achieved only occasionally in human

5

Interface Focus 6: 20160008

3. Results

feature height indicator

rsfs.royalsocietypublishing.org

carbide. Two of the faces were separated by a 10 mm line contact. In the experiments, the indenter was driven into the enamel with the line contact edge-on. In both the above experiments, the procedure was to advance the cutting tip towards the tooth section until a small force was registered indicating contact. The displacement was then recorded and the micrometer zeroed. The tip was then moved away from the specimen and down below the lower edge of the resin. The micrometer was then returned to the zero position and a small advance added to allow the tip to cut through the specimen when moved upwards (the position shown in figure 2e). At the start of the experiment, cutting at 12 mm min21, the indenter tip passed first across a known length of embedding resin before entering the enamel. The micrometer that provided depth control had an accuracy of only +1 mm, but via trial and error, it was found possible to produce cutting tracks of submicrometre depth. In experiments where epigallocatechin gallate was added, half the test was run with a coating of (whole unstimulated) human saliva added to the face of the tooth via a 100 ml micropipette. The test was then stopped for about 5 s while 100 ml of epigallocatechin gallate was added. Within a further 2 s of that action, the test was restarted. A third series of experiments was run on a Hysitron Ubi700 (Minneapolis, MN, USA) nanoindenter, equipped with a twoaxis load cell. The cutting tip was that of a Berkovich diamond with a tip sharpness of about 10 nm. The system of loading was equivalent to ‘dead-weight’ vertical loads (i.e. constant Ft) that were varied between 250 and 4000 mN. Experiments at each force were repeated once. The estimation of the cutting force, Fc, was only available for horizontal displacements of 16 mm, so confining the length of cut to this limit. In the experiments, the Berkovich indenter was driven basically face-forwards into the enamel, with a slight deviation of 5–108, being unable to set to greater precision. The displacement at the start of the experiment was taken as the groove depth t. Post-test surfaces were examined in an atomic force microscope set in tapping mode (Agilent 5500 atomic force microscopy (AFM), Santa Clara, CA, USA). Images of the Berkovich tests were taken post-test before and after washing with a jet of deionized water.

Ft = 250 µN

0

10

–40

20

–80

6

–120

30

–160

40

–200 50

–240

(c)

(d)

Interface Focus 6: 20160008

1750

100

1500

50

cutting force Fc (µN)

Ft = 4000 µN

(b)

0

rsfs.royalsocietypublishing.org

tip displacement (nm)

(a)

0

1000

–50 500

–100 –150

0

–200 –500

–250 0

10

20

30

40

50

0

10

20

30

40

50

time (s)

time (s)

Figure 4. Panels (a,b) show plots of the tip displacements under load, while (c,d) indicate cutting forces (Fc) for a Berkovich tip making an enamel groove. The displacement is taken as equal to groove depth t in calculations, both plotted against time. (a,c) Plots for Ft ¼ 250 mN (the lowest), while (b,d) are for Ft ¼ 4000 mN (the largest). (Online version in colour.)

er

aft

0

100 nm –100

10

µm

5

e for 3 be 2

0

5

1 25

0µ

N 50

0

µN 10

00

µN 20

00

t

µN 00

µN

on

00

µN

height/depth (nm)

m

di

2

1

0

of

ti rec

40

100 75 50 25 0 –25 –50 –75 –100

en

m

e ov

30

20 40 displacement (µm)

60

0

3

20 40 displacement (µm)

60

0

20 40 displacement (µm)

60

Figure 5. The effect of the Berkovich tip on enamel. The indenter has moved face-first into the enamel at different fixed vertical (Ft) forces (indicated on the image at the start of the run). The images were made both post-test (orange) and after washing of the surface with deionized water (light blue). Profiles were constructed (1) early, (2) in the middle and (3) at the end of the 16 mm long cuts. These indicate prowing (clearest at higher resolution at top right), but disappearance of material after washing is also diagnostic of fracture. Reduced groove depths compared to the displacement under load in figure 4 reflect approximately 50% elastic recovery at the deepest part of the groove. (Online version in colour.) enamel at much larger crack lengths when cracks are run from outside the crown towards the enamel–dentine junction [64,75]. Values of approximately 0.8–1 MPa m0.5 are obtained

more usually from indentations [87]. The crucial difference lies in the crack paths. When cracks run freely, they take a route that avoids all the crystal –protein interfaces that are

r2 = 0.989

6

3

1.4 0 50

100 t (nm)

150

200

saliva

saliva + polyphenolic

0.25

force (N)

0.20 0.15

Fc

0.10 0.05

0

Ft 1

2 3 displacement (mm)

4

Figure 7. The rise in force produced via the addition of a polyphenolic (0.1 M epigallocatechin gallate) to a salivary coating on the enamel surface. (Online version in colour.)

encountered during abrasive wear when a cutting tip forces the crack to cross such obstacles. For the present tests, the extensive ‘plastic’ pile-up, as a result of prowing, about the scratches indicates that considerable plastic energy is also dissipated during cutting. In sum, the results here seem to indicate clearly that high toughness in enamel at very small scale helps to protect against abrasion. It is almost as if enamel has adapted to resist abrasion-induced cutting, i.e. it shows a reverse R-curve response, especially under highly constrained shear loading, thereby being better able to resist cracking than under tensile loading. This is very different behaviour to a tissue like bone. The estimate obtained here for the shear yield stress, at 1.48 GPa, also exceeds published estimates. For example, assuming the Tresca yield criterion, then a value of ty ¼ 0.83 GPa, i.e. one-sixth of the hardness obtained in nanoindentation tests [31], would be expected. Similar estimates can be extrapolated from measurements of the compressive yield stress of human enamel at the same scale as our tests [88]. This suggests that our toughness estimate is also too high. However, it would in any case be far above that obtained from standard ‘free-cracking’ indentation tests. The sum effect for a mammal facing tooth wear is that the high toughness of enamel at this small scale slows the wear rate in that rubbing via blunt contacts requires multiple contacts to remove tissue while abrasion is immediate. Thus, the avoidance of abrasion prolongs tooth life.

7

Interface Focus 6: 20160008

Figure 6. Fc/t plotted against t for the Berkovich tests shown in figures 4 and 5. (Online version in colour.)

Biologists and materials scientists have approached the mechanical behaviour of tooth tissues from very different perspectives. The former have tended to focus on tooth function and the factors that disturb this, such as tooth wear. By contrast, materials scientists seem to have been more concerned with how mineralized tissues as a whole, made from apparently unpromising structural components, succeed in surviving so long without falling apart. Damage control is where the two approaches should coincide, but often do not. The explanations for the difference between bone and enamel, from a biological perspective, relate to their performance as organs. A simple reflection of this is their variability. Bone tissue is, in any particular location in a healthy adult body, a very similar tissue in any mammalian species. By contrast, enamel is very variable, reflecting its need to adapt to diet [2]. It is highly likely that the value for toughness that we have established here will vary between species. The nature of the structural differences between species and how they coincide with diet is currently an open-ended question. There is a need to catalogue variation in the mechanical properties of enamel at different scales within mammalian taxa. Combining these variations with selection pressures, be these environmental, dietary or otherwise, that may have facilitated the functional optimization of the material could broaden understanding as to how enamel reacts evolutionarily to ensure its lifelong survival and efficient food processing. Previously, it has been stated that the transition between rubbing and abrasion, clearly demonstrated in cutting of more malleable materials such as metals, and now with a satisfactory theoretical basis [65,71], does not apply to a crystalline composite such as enamel [89]. The data presented here suggest that it is possible to apply this analysis, even though both prowing and fracture may coexist. There will be a transition between almost completely plastic behaviour and that involving fracture (figure 3). We do not yet know how sharp this transition is—it may well depend on enamel orientation, for example—but we believe that these experiments help cement the need to include fracture mechanics and cutting when considering wear events in enamel and promote a departure from more simplistic models that tend to ignore these factors. Many analyses of cutting assume that, all else being equal, groove depth (for which we may imply ‘wear rate’) is inversely proportional to hardness, or that wear resistance (the reciprocal of wear rate) is directly proportional to hardness [74]. This presumes that the force perpendicular to the surface, Ft, is borne by the horizontal projection of the indenter area and the horizontal (cutting) force, Fc, is endured by the vertical projection of the indentation. However, this extension of static indentation to translation needs to include fracture toughness to account for the cost of tissue loss: ‘scratch hardness’ and ‘indentation hardness’ are thus different phenomena [71]. The importance of plant polyphenolic compounds, or tannins, in ecology is great since they are among the most common chemical defences of plants. Many of the proteins in enamel bind to them [90], the resulting complex being precipitated. Our results at very small scale add to the general conclusion that this reaction increases friction against the enamel. Further research should help to establish if this increases enamel wear rates. Our findings help elucidate the relationship between macroand microscopic wear. We demonstrate that not all microscopic

rsfs.royalsocietypublishing.org

Fc /t (mN nm–1)

9

important variable in understanding the relationship between micro- and macroscopic wear. All other things being equal, deeper grooves are more likely to represent abrasive scratches than shallow grooves. A complication, of course, is that groove depth is also a function of the mechanical properties of the particles creating the grooves, and these may be quite variable. An interesting avenue of future research would be to determine whether or not groove shape can be used to discriminate rubbing versus abrasive marks that were formed on tooth surfaces during normal feeding behaviours.

Facilities Projects GD 02/11, GE 01/07 and also grant DB 01/12.

Acknowledgements. Many thanks to Muhammed Aleem and Aateka Vahora for technical assistance.

References 1.

Lucas PW. 2004 Dental functional morphology. Cambridge, UK: Cambridge University Press. 2. Lucas PW, Constantino P, Wood BA, Lawn BR. 2008 Dental enamel as a dietary indicator in mammals. BioEssays 30, 374– 385. (doi:10.1002/bies.20729) 3. Lee JJ-W, Constantino P, Lucas PW, Lawn BR. 2011 Fracture in teeth—a diagnostic for inferring tooth function and diet. Biol. Rev. 86, 959–974. (doi:10. 1111/j.1469-185X.2011.00181.x) 4. Lucas PW, van Casteren A. 2015 The wear and tear of teeth. Med. Princ. Pract. 24(Suppl. 1), 3 –13. (doi:10.1159/000367976) 5. Lanyon JM, Sanson GD. 1986 Koala (Phascolarctos cinereus) dentition and nutrition. II. Implications of tooth wear in nutrition. J. Zool. 209, 169– 181. (doi:10.1111/j.1469-7998.1986.tb03573.x) 6. Pahl LI. 1987 Survival, age determination and population age structure of the common ringtail possum, Pseudocheirus peregrinus, in a Eucalyptus woodland and a Leptospermum thicket in southern Victoria. Aust. J. Zool. 35, 625–639. (doi:10.1071/ ZO9870625) 7. Logan M, Sanson GD. 2002 The effect of tooth wear on the feeding behaviour of free-ranging koalas (Phascolarctos cinereus, Goldfuss). J. Zool. 256, 63 –69. (doi:10.1017/S0952836902000080) 8. DeGusta D, Everett MA, Milton K. 2003 Natural selection on molar size in a wild population of howler monkeys (Alouatta palliata). Proc. R. Soc. Lond. B 270, S15–S17. (doi:10.1098/rsbl.2003.0001) 9. King SJ, Arrigo-Nelson SJ, Pochron ST, Semprebon GM, Godfrey LR, Wright PC, Jernvall J. 2005 Dental senescence in a long-lived primate links infant survival to rainfall. Proc. Natl Acad. Sci. USA 102, 16 579–16 583. (doi:10.1073/pnas. 0508377102) 10. Cuozzo FP, Sauther ML. 2006 Severe wear and tooth loss in wild ring-tailed lemurs (Lemur catta): a function of feeding ecology, dental structure, and individual life history. J. Hum. Evol. 51, 490–505. (doi:10.1016/j.jhevol.2006.07.001)

11. Munch E, Launey ME, Alsem DH, Saiz E, Tomsia AP, Ritchie RO. 2008 Tough, bio-inspired hybrid materials. Science 322, 1516– 1520. (doi:10.1126/ science.1164865) 12. He L, Swain MV. 2007 Contact induced deformation of enamel. Appl. Phys. Lett. 90, 171916. (doi:10. 1063/1.2450649) 13. Ziscovici C, Lucas PW, Constantino PJ, Bromage TG, van Casteren A. 2014 Sea otter dental enamel is highly resistant to chipping due to its microstructure. Biol. Lett. 10, 20140484. (doi:10. 1098/rsbl.2014.0484) 14. Myoung S, Lee J, Constantino P, Lucas P, Chai H, Lawn B. 2009 Morphology and fracture of enamel. J. Biomech. 42, 1927– 1951. (doi:10.1016/j. jbiomech.2009.05.013) 15. Rivera C, Arola D, Ossa A. 2013 Indentation damage and crack repair in human enamel. J. Mech. Behav. Biomed. Mater. 21, 178–184. (doi:10.1016/j. jmbbm.2013.02.020) 16. Constantino PJ, Lee JJ-W, Chai H, Zipfel B, Ziscovici C, Lawn BR, Lucas PW. 2010 Tooth chipping can reveal the diet and bite forces of fossil hominins. Biol. Lett. 6, 826 –829. (doi:10.1098/rsbl.2010.0304) 17. Chai H, Lee JJW, Kwon JY, Lucas PW, Lawn BR. 2009 A simple model for enamel fracture from margin cracks. Acta Biomater. 5, 1663– 1667. (doi:10.1016/ j.actbio.2008.11.007) 18. Sognnaes RF. 1949 The organic elements of the enamel: II. The organic framework of the internal part of enamel, with special regard to the organic basis for the so-called tufts and Schreger’s bands. J. Dent. Res. 28, 549–557. (doi:10.1177/00220345490280060401) 19. Amizuka N et al. 2005 Ultrastructural images of enamel tufts in human permanent teeth. J. Oral Biosci. 47, 33 –41. (doi:10.1016/S13490079(05)80006-4) 20. Chai H, Lee JJ-W, Constantino P, Lucas PW, Lawn BR. 2009 Remarkable resilience of teeth. Proc. Natl Acad. Sci. USA 106, 7289– 7293. (doi:10.1073/pnas. 0902466106)

21. Sognnaes RF. 1950 The organic elements of enamel: IV. The gross morphology and the histological relationship of the lamellae to the organic framework of the enamel. J. Dent. Res. 29, 260–269. (doi:10.1177/00220345500290030201) 22. Organ C, Nunn CL, Machanda Z, Wrangham RW. 2011 Phylogenetic rate shifts in feeding time during the evolution of Homo. Proc. Natl Acad. Sci. USA 108, 14 555–14 559. (doi:10.1073/pnas. 1107806108) 23. Fortelius M, Solounias N. 2000 Functional characterization of ungulate molars using the abrasion– attrition wear gradient: a new method for reconstructing paleodiets. Am. Mus. Novitates 3301, 1–36. (doi:10.1206/0003-0082(2000) 301,0001:FCOUMU.2.0.CO;2) 24. Damuth J, Janis CM. 2011 On the relationship between hypsodonty and feeding ecology in ungulate animals, and its utility in palaeoecology. Biol. Rev. 86, 733 –758. (doi:10.1111/j.1469-185X. 2011.00176.x) 25. Bermudez de Castro JM, Martino´n-Torres N, Sarmiento S, Lozano M, Arsuaga JL, Carbonell E. 2003 Rates of anterior tooth wear in Middle Pleistocene hominins from Sima de los Huesos (Sierra de Atapuerca, Spain). Proc. Natl Acad. Sci. USA 100, 11 992–11 996. (doi:10.1073/pnas. 2034879100) 26. Kay RF. 1981 The nut-crackers—a new theory of the adaptations of the Ramapithecinae. Am. J. Phys. Anthropol. 55, 141– 151. (doi:10.1002/ajpa. 1330550202) 27. Kay RF. 1985 Dental evidence for the diet of Australopithecus. Ann. Rev. Anthropol. 14, 315–341. (doi:10.1146/annurev.an.14.100185.001531) 28. Dean MC, Jones ME, Pilley JR. 1992 The natural history of tooth wear, continuous eruption and periodontal disease in wild shot great apes. J. Hum. Evol. 22, 23–39. (doi:10.1016/0047-2484(92) 90027-7) 29. Engelbrecht JP, McDonald EV, Gillies JA, Gertler AW. 2008 Department of Defense Enhanced Particulate

Interface Focus 6: 20160008

Competing interests. We declare we have no competing interests. Funding. We acknowledge support from Kuwait University General

8

rsfs.royalsocietypublishing.org

marks on tooth surfaces are necessarily abrasive: enamel deformed plastically along the shallowest grooves produced by the tungsten carbide blade (in figure 3), even though the blade was angulated and harder than enamel. It follows that particles like phytoliths, which are both softer and lower in modulus than enamel [84], or that lack sufficient angulation, might also produce such rubbing marks [31]. The lower modulus may also lead to elastic blunting due to the strain prior to the yield point, but we calculate from published values of the hardness and modulus [84] that the strain at yield will be less than 4% and thus unlikely to change much about the contact geometry. Phytoliths are abundant in some plant tissues, so researchers must not assume that all marks on herbivore teeth are abrasive. Similar results were found with the Berkovich indenter, which has a lower attack angle than the blade: extensive prowing was observed, but microcracking within a prowing field was also observed. Groove depth emerges as an

31.

32.

34.

35. 36.

37.

38.

39.

40.

41.

42.

43.

44.

60.

61.

62.

63.

64.

65.

66.

67.

68.

69.

70.

71. 72. 73.

74. 75.

76.

Am. J. Phys. Anthropol. 147, 551–579. (doi:10. 1002/ajpa.22007) Launey ME, Chen P-Y, McKittrick J, Ritchie RO. 2010 Mechanistic aspects of fracture and R-curve behavior in elk antler bone. Acta Biomater. 6, 1505–1514. (doi:10.1016/j.actbio.2009.11.026) Qu Q, Haitina T, Zhu M, Ahlberg PE. 2015 New genomic and fossil data illuminate the origin of enamel. Nature 526, 108 –111. (doi:10.1038/ nature15259) Currey JD. 1980 Mechanical properties of mollusc shell. In The mechanical properties of biological materials (eds JFV Vincent, JD Currey), pp. 75 –97. Cambridge, UK: Cambridge University Press. Anstis GR, Chantikal P, Marshall DB, Lawn BR. 1981 Evaluation of indentation techniques for measuring fracture toughness: I. Direct crack measurements. J. Am. Ceram. Soc. 64, 533–538. (doi:10.1111/j. 1151-2916.1981.tb10320.x) Bajaj D, Arola DD. 2009 On the R-curve behavior of human tooth enamel. Biomaterials 30, 4037–4046. (doi:10.1016/j.biomaterials.2009.04.017) Atkins AG, Liu JH. 2007 Toughness and the transition between cutting and rubbing in abrasive contacts. Wear 262, 146–159. (doi:10.1016/j.wear. 2006.04.002) Akono A-T, Ulm F-J. 2011 Scratch test model for the determination of fracture toughness. Eng. Fract. Mech. 78, 334 –342. (doi:10.1016/j.engfracmech. 2010.09.017) Ulm F-J, James S. 2011 The scratch test for strength and fracture toughness determination of oil well cements cured at high temperature and pressure. Cement Concrete Res. 41, 942 –946. (doi:10.1016/j. cemconres.2011.04.014) Lin J-S, Zhou Z. 2013 Can scratch tests give fracture toughness? Eng. Fract. Mech. 109, 161 –168. (doi:10.1016/j.engfracmech.2013.06.002) Wang H, Chang L, Williams JG. 2015 On the toughness measurement for ductile polymers by orthogonal cutting. Eng. Fract. Mech. 149, 276– 286. (doi:10.1016/j.engfracmech.2015.06.067) Blackman BRK, Hoult T, Patel Y, Steininger H, Williams JG. 2016 Steady-state scratch testing of polymers. Polym. Test. 49, 38 –45. (doi:10.1016/j. polymertesting.2015.11.002) Atkins AG. 2009 The science and engineering of cutting. Amsterdam, The Netherlands: Elsevier. Atkins AG, Mai Y-W. 1985 Elastic and plastic fracture. Chichester, UK: Ellis Horwood. Atkins T. 2009 Toughness and processes of material removal. Wear 267, 1764–1771. (doi:10.1016/j. wear.2009.04.010) Archard JF. 1953 Contact and rubbing of flat surfaces. J. Appl. Phys. 24, 981–988. (doi:10.1063/1.1721448) Erickson GM, Krick BA, Hamilton M, Bourne GR, Norell MA, Lilliodden E, Sawyer WG. 2012 complex dental structure and wear biomechanics in hadrosaurid dinosaurs. Science 338, 98 –101. (doi:10.1126/science.1224495) Prinz JF, Lucas PW. 2000 Saliva tannin interactions. J. Oral Rehabil. 27, 991 –994. (doi:10.1046/j.13652842.2000.00578.x)

9

Interface Focus 6: 20160008

33.

45. Walker A, Teaford MF. 1989 Inferences from quantitative analysis of dental microwear. Folia Primatol. 53, 177 –189. (doi:10.1159/000156415) 46. Mainland IL. 2003 Dental microwear in grazing and browsing Gotland sheep (Ovis aries) and its implications for dietary reconstruction. J. Archaeol. Sci. 30, 1513–1527. (doi:10.1016/S0305-4403(03) 00055-4) 47. Scott RS, Ungar PS, Bergstrom TS, Brown CA, Grine FE, Teaford MF, Walker A. 2005 Dental microwear texture analysis shows within-species diet variability in fossil hominins. Nature 436, 693 –695. (doi:10. 1038/nature03822) 48. Scott RS, Ungar PS, Bergstrom TS, Brown CA, Childs BE, Teaford MF, Walker A. 2006 Dental microwear texture analysis: technical considerations. J. Hum. Evol. 51, 339–349. (doi:10.1016/j.jhevol.2006.04.006) 49. Ungar PS, Grine FE, Teaford MF. 2008 Dental microwear and diet of the Plio-Pleistocene hominin Paranthropus boisei. PLoS ONE 3, 101371. (doi:10. 1371/annotation/195120f0-18ee-4730-9bd60d6effd68fcf ) 50. Merceron G, Escarguel G, Angibault J-M, VerheydenTixier H. 2010 Can dental microwear textures record inter-individual dietary variations? PLoS ONE 5, e9542. (doi:10.1371/journal.pone.0009542) 51. Constantino P, Borrero-Lopez O, Pajares A, Lawn BR. 2016 Simulation of enamel wear for reconstruction of diet and feeding behavior in fossil animals: a micromechanics approach. BioEssays 30, 89– 99. (doi:10.1002/bies.201500094) 52. Imai E, Hatae K, Shimada A. 1995 Oral perception of grittiness: effect of particle size and concentration of the dispersed particles and their dispersion medium. J. Texture Stud. 26, 561–576. (doi:10. 1111/j.1745-4603.1995.tb00804.x) 53. Imai E, Shimichi Y, Maruyama I, Inoue A, Ogawa S, Hatae K, Shimada A. 1997 Perception of grittiness in an oil-in-water emulsion. J. Texture Stud. 28, 257–272. (doi:10.1111/j.1745-4603.1997.tb00116.x) 54. Prinz JF. 2004 Abrasives in foods and their effect on intra-oral processing: a two-colour chewing gum study. J. Oral Rehabil. 31, 968–974. (doi:10.1111/j. 1365-2842.2004.01328.x) 55. Trulsson M, Johansson RS. 1994 Encoding of amplitude and rate of forces applied to the teeth by human periodontal mechanoreceptive afferents. J. Neurophysiol. 72, 1734–1744. 56. Trulsson M. 2006 Sensory-motor function of human periodontal mechanoreceptors. J. Oral Rehabil. 33, 262 –273. (doi:10.1111/j.1365-2842.2006.01629.x) 57. Trulsson M, Essick GK. 2010 Sensations evoked by microstimulation of single mechanoreceptive afferents innervating the human face and mouth. J. Neurophysiol. 103, 1741–1747. (doi:10.1152/jn. 01146.2009) 58. Kadohisa M, Verhagen JV, Rolls ET. 2005 The primate amygdala: neuronal representations of the viscosity, fat texture, temperature, grittiness and taste of foods. Neurosci. 132, 33– 48. (doi:10.1016/ j.neuroscience.2004.12.005) 59. Scott RS, Teaford MF, Ungar PS. 2012 Dental microwear texture and anthropoid diets.

rsfs.royalsocietypublishing.org

30.

Matter Surveillance Program final report. Reno, NV: Desert Research Institute. Riede F, Wheeler JM. 2009 Testing the ‘Laacher See hypothesis’: tephra as dental abrasive. J. Archaeol. Sci. 39, 2384–2391. (doi:10.1016/j.jas.2009.06.020) Lucas PW et al. 2013 Mechanisms and causes of wear in tooth enamel: implications for hominin diets. J. R. Soc. Interface 10, 20120923. (doi:10. 1098/rsif.2012.0923) Walker PL. 1976 Wear striations on the incisors of cercopithecoid monkeys as an index of diet and habitat preference. Am. J. Phys. Anthropol. 45, 299–307. (doi:10.1002/ajpa.1330450215) Teaford MF, Walker A. 1984 Quantitative differences in dental microwear between primate species with different diets and a comment on the presumed diet of Sivapithecus. Am. J. Phys. Anthropol. 64, 191–200. (doi:10.1002/ajpa.1330640213) Grine FE. 1986 Dental evidence for dietary differences in Australopithecus and Paranthropus: a quantitative analysis of permanent molar microwear. J. Hum. Evol. 15, 783–822. (doi:10. 1016/S0047-2484(86)80010-0) Teaford MF. 1988 A review of dental microwear and diet in modern mammals. Scan. Microsc. 2, 1149–1166. Teaford MF, Oyen OJ. 1989 In vivo and in vitro turnover in dental microwear. Am. J. Phys. Anthropol. 80, 447 –460. (doi:10.1002/ajpa. 1330800405) Teaford MF, Glander KE. 1991 Dental microwear in live, wild-trapped Alouatta palliata from Costa-Rica. Am. J. Phys. Anthropol. 85, 313–319. (doi:10.1002/ ajpa.1330850310) Gugel IL, Grupe G, Kenzelmann K-H. 2001 Simulation of dental microwear: characteristic traces by opal phytoliths give clues to ancient human dietary behavior. Am. J. Phys. Anthropol. 114, 124–138. (doi:10.1002/1096-8644(200102)114: 2,124::AID-AJPA1012.3.0.CO;2-S) Rodrigues HG, Merceron G, Viriot L. 2009 Dental microwear patterns of extant and extinct Muridae (Rodentia, Mammalia): ecological implications. Naturwissenschaften 96, 537 –542. (doi:10.1007/ s00114-008-0501-x) Schulz E, Piotrowski V, Clauss M, Mau M, Merceron G, Kaiser TM. 2013 Dietary abrasiveness is associated with variability of microwear and dental surface texture in rabbits. PLoS ONE 8, e56167. (doi:10.1371/journal.pone.0056167) Walker A, Hoeck H, Perez L. 1978 Microwear of mammalian teeth as an indicator of diet. Science 201, 908–910. (doi:10.1126/science.684415) Covert HH, Kay RF. 1981 Dental microwear and diet: implications for determining the feeding behaviors of extinct primates, with a comment on the dietary pattern of Sivapithecus. Am. J. Phys. Anthropol. 55, 331–336. (doi:10.1002/ajpa.1330550307) Kay RF. 1987 Analysis of primate dental microwear using image processing techniques. Scan. Microsc. 1, 657–662. Grine FE, Kay RF. 1988 Early hominid diets from quantitative image analysis of dental microwear. Nature 333, 765 –768. (doi:10.1038/333765a0)

82.

83.

84.

86.

87.

88.

89.

90.

from an evolutionary perspective. Evol. Dev. 18, 54– 61. (doi:10.1111/ede.12169) Bajaj D, Park S, Quinn GD, Arola D. 2010 Fracture processes and mechanisms of crack growth resistance in human enamel. JOM 62, 76 –82. (doi:10.1007/s11837-010-0113-8) Ang SF, Scholz T, Klocke A, Schneider GA. 2009 Determination of the elastic/plastic transition of human enamel by nanoindentation. Dent. Mater. 25, 1403–1410. (doi:10.1016/j.dental.2009. 06.014) Xia J, Zheng J, Huang D, Tian ZR, Chen L, Zhou Z, Ungar PS, Qian L. 2015 New model to explain tooth wear with implications for microwear formation and diet reconstruction. Proc. Natl Acad. Sci. USA 112, 10 669 –10 672. (doi:10.1073/pnas.1509491112) Bennick A. 2002 Interaction of plant polyphenols with salivary proteins. Crit. Rev. Oral Biol. Med. 13, 184–196. (doi:10.1177/154411130201300208)

10

Interface Focus 6: 20160008

85.

Hydrocoll. 23, 1984–1992. (doi:10.1016/j.foodhyd. 2009.03.001) van Aken GA. 2010 Modelling texture perception by soft epithelial surfaces. Soft Matter 6, 826–834. (doi:10.1039/b916708k) Prinz JF, Lucas PW. 1997 An optimization model for mastication and swallowing in mammals. Proc. R. Soc. Lond. B 264, 1715–1721. (doi:10. 1098/rspb.1997.0238) Lucas PW et al. 2014 The role of dust, grit and phytoliths in tooth wear. Ann. Zool. Fenn. 51, 143 –152. (doi:10.5735/086.051.0215) Yilmaz ED, Schneider GA, Swain MV. 2015 Influence of structural hierarchy on the fracture behaviour of tooth enamel. Phil. Trans. R. Soc. A 373, 20140130. (doi:10.1098/rsta.2014.0130) Lucas PW, Philip SM, Al-Qeoud D, Al-Draihim N, Saji S, van Casteren A. 2016 Structure and scale of the mechanics of mammalian dental enamel viewed

rsfs.royalsocietypublishing.org

77. Breslin BAS, Gilmore MM, Beauchamp GK, Green BG. 1993 Psychophysical evidence that oral astringency is a tactile sensation. Chem. Senses 18, 405 –417. (doi:10.1093/chemse/18.4.405) 78. Le Bourvellec C, Renard CMGC. 2012 Interactions between polyphenols and macromolecules: quantification methods and mechanisms. Crit. Rev. Food Sci. Nutr. 52, 213– 248. (doi:10.1080/ 10408398.2010.499808) 79. de Wijk RA, Prinz JF. 2005 The role of friction in perceived oral texture. Food Qual. Pref. 16, 121– 129. (doi:10.1016/j.foodqual.2004.03.002) 80. de Wijk RA, Prinz JF. 2006 Mechanisms underlying the role of friction in oral texture. J. Texture Stud. 37, 413–427. (doi:10.1111/j.1745-4603.2006. 00060.x) 81. Rosetti D, Bongaerts JHH, Wantling E, Stokes JR, Williamson A-M. 2009 Astringency of tea catechins: more than an oral lubrication tactile percept. Food