Journal of Oral Rehabilitation 2006 33; 26–30
Hardness of enamel exposed to Coca-Cola
and artificial
saliva
H. DEVLIN*, M. A. BASSIOUNY† & D. BOSTON†
†
UK and Department of Restorative Dentistry, Temple University School of Dentistry, Philadelphia, PA, USA
SUMMARY The purpose of this study was to deter-
mine the rate of change in indentation hardness of
enamel in permanent teeth exposed to Coca-Cola
.
In a further experiment, the ability of a commer-
cially available artificial saliva to remineralize
enamel treated with Coca-Cola
was tested. Ten
enamel specimens were randomly chosen to be
treated with Coca-Cola
(experimental groups)
and seven with water (control group). The fluids
were applied for 1, 2, 3 h and overnight (15 h),
washed off with a few drops of water and the moist
enamel indentation hardness tested after each
interval. With Coca-Cola
treatment, the mean
enamel hardness was 92Æ6% (s.d. ¼ 7Æ9) of the ori-
ginal baseline hardness after 1 h, 93Æ25%
(s.d. ¼ 10Æ15) after 2 h, 85Æ7% (s.d. ¼ 12Æ03) after 3 h
and 80Æ3% after 15 h. The mean indentation hard-
Introduction
It has been shown previously that acidic beverages
(Coca-Cola
) cause a softening of enamel during long-
term exposure for an hour (1). However, the time
course of further softening with further exposure is not
known, despite the almost continuous ingestion of
these types of drinks by some patients. Johansson (2)
examined Saudi military inductees (n ¼ 95) and found
the average soft drink consumption to be 247 L year)1,
with erosion intensified by holding the drink in the
mouth before swallowing it.
Tooth erosion is highly prevalent, especially amongst
children. A recent study from the UK of 12–14-year-old
children (3) showed that erosion affected 56Æ3% of
12-year-old children and 64Æ1% of children aged
14 years. Arnadottir et al. (4) found that tooth erosion
ª 2006 Blackwell Publishing Ltd
*The Dental School, University of Manchester, Manchester,
ness of control specimens treated with water was
108Æ7% (s.d. ¼ 16Æ09) of the original hardness after
1 h, 99Æ09% (s.d. ¼ 18Æ98) after 2 h, 98Æ97% (s.d. ¼
11Æ24) after 3 h and 98Æ42% (s.d. ¼ 22Æ78) after 15 h.
In a separate experiment, the hardness of 9 enamel
specimens was tested, as previously described,
before and after treatment with Coca-Cola
over-
night and again after application of artificial saliva
for 3 min. Coca-Cola
reduced the mean indenta-
tion hardness of enamel in the teeth, but the
hardness was partially restored with artificial saliva
(Salivart
) and increased by 18% from the demin-
eralized enamel hardness.
KEYWORDS: artificial saliva, hardness, enamel erosion
Accepted for publication 17 April 2005
was found in 21Æ6% of 15-year-old children in Iceland.
Those patients with xerostomia are also prone to tooth
erosion because the absence of saliva means that
buffering of acidic drinks and remineralization of the
demineralized enamel does not take place. Reduced
salivation can occur in anorexia nervosa (5) and
primary Sjogren’s syndrome (6) and is a common
side-effect of many drugs.
Where salivary flow is much reduced, tooth erosion
is a common clinical problem. Many of these xeros-
tomic patients may consume large quantities of soft
drinks as a method of relieving their oral discomfort
and the absence of salivary flow will mean that the
unbuffered beverage is in contact with the enamel far
longer than normal. Artificial saliva will remineralize
early enamel lesions (7). Commercially available oral
moisturizers such as Salivart
contain calcium
26
 REDUCED ENAMEL HARDNESS AND ARTIFICIAL SALIVA
chloride and potassium phosphate and other mineral
salts, which may assist in the remineralization of the
enamel.
Recent developments in instrument technology
have allowed more accurate measurement of enamel
hardness, and shown that considerable intra-tooth
variation occurs across the enamel surface (8). The
situation is further complicated by the absence of
absolute standards of hardness, i.e. that hardness
values refer to a given load applied under particular
circumstances. This results in great difficulty in repro-
ducing experimental results where hardness is the
outcome variable.
The purpose of this study was to determine the rate of
change in indentation hardness and indentation mod-
ulus of enamel exposed to Coca-Cola
in permanent
teeth and to determine the effect on hardness of
applying an artificial saliva (Salivart
) to demineralized
enamel.
Materials and methods
Experimental application of Coca-Cola
only
Seventeen permanent teeth were extracted from adults
who gave their informed consent. The teeth were
disinfected for 24 h with Cetylcide II* which contains
no alcohol, bleach, iodophor, phenol, or glutaralde-
hyde, and stored in deionized, distilled water at room
temperature. The specimens were sectioned randomly
from caries-free permanent teeth and embedded in
epoxy resin cylinders. The exposed facial section of the
enamel specimens was ground flat and polished wet
with 600-grit silicon carbide paper. Ten specimens were
randomly chosen to be treated with Coca-Cola
†
(experimental groups) and seven with water (control
group). The pH of the Coca-Cola
solution was 2Æ48
(expiry date 2005) and was measured using a model
1Q150 pH meter‡. The pH of the distilled water used in
the control group was 6Æ40. Three hundred microlitres
of either fluid was left for 1, 2, 3 h and overnight
(15 h), washed off with a few drops of water and the
enamel tested for hardness after each interval. The
enamel was tested moist, as allowing it to dry out would
increase the hardness (9).
*Cetylite Industries Inc., Pennsauken, NJ, USA.
† §
Coca-Cola Co., Atlanta, GA, USA.
‡ ¶
IQ Scientific Instruments, Inc., San Diego, CA, USA.
ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 26–30
27
Indentation methodology
A CSM
Instruments hardness tester§ with a fine
load range of 0Æ05 to 10 N was used. This sat on a
Vibraplane Model 9100/9200 vibration isolation table¶
to eliminate slight environmental tremors which
might interfere with the indentation tests. Using this
micro-indentation system, a diamond Vickers indenter
(tip radius 50 nm) was forced into the enamel
specimen with an increasing load for a known
duration.
Depth adjustment was performed using a contact
load detection of 0Æ01 N. During testing, the enamel
specimen was loaded at 1 N min)1 and unloaded at
1 N min)1. After reaching a maximum load of 0Æ5 N,
with a zero dwell time, the load was reduced to
complete relaxation. The position of the indenter
relative to the enamel surface was monitored continu-
ously with a differential capacitive sensor with a
100 lm range. At the start of the experiment and on
each occasion after application of Coca-Cola
or water,
12 indentations were recorded by moving the enamel
sample 50 lm between indentations. Subsequent
indentations following application of test or control
solutions were made parallel to and 50 lm from the
previous series of indentations. The data acquisition
was automatic with real-time display of applied force
and indentation depth.
Indentation hardness (HIT) was calculated from the
indentation curve with load plotted against depth
(Fig. 1). The following parameters were used: elastic
modulus of the indenter (1141 GPa), Poisson’s ratio of
the indenter (0Æ07), and Poisson’s ratio of the enamel
sample (0Æ25) (8).
Earlier technology required the visual measurement
of permanent plastic indentations, but our methodo-
logy measured total (including elastic) deformation.
The depth resolution of the indentation was 0Æ3 nm,
with a maximum indenter range of 200 lm. Indenta-
tion depth was measured using a Vickers (V-D44)
indenter as described by Oliver and Pharr (10). The
enamel specimen was moved to a previously defined
position and the indenter located to within 100 lm.
The final approach by the measurement head of the
indenter was made, and following contact, indentation
depth recorded continuously.
Micro Photonics Inc., Irvine, CA, USA.
Kinetic Systems Inc., Boston, MA, USA.
28 H . D E V L I N et al.

Load-displacement curve behave as a perfectly elastic material, as indicated by

the hysteresis in the curve. Indentation of enamel
Load (N)
causes an elastic and a non-recoverable plastic compo-
0·5
nent of deformation recovery.
For the experimental group the baseline measure-
ments for mean indentation hardness (HIT) was
0·3
4Æ48 GPa (s.d. ¼ 0Æ64). Mean HIT for the control group
was 4Æ32 GPa (s.d. ¼ 0Æ69). There was no significant
difference between baseline indentation hardness
0·1 hp b values of experimental and control groups (P ¼ 0Æ922,
0 using Mann–Whitney U-test).
0 1 2 3 To allow comparison across the different specimens,
Displacement (µm)
the hardness at 1, 2, 3 h and overnight (15 h) were
Fig. 1. Typical load-displacement curve of enamel. A maximum expressed as percentages of hardness at baseline, prior
load of 0Æ5 N was applied and the displacement continuously to any treatment (Fig. 2). With Coca-Cola
treatment,
recorded. the mean enamel hardness was 92Æ6% (s.d. ¼ 7Æ9) of
the original baseline hardness after 1 h, 93Æ25%
A calibration hardness standard (HIT ¼ 9Æ987 GPa) was (s.d. ¼ 10Æ15) after 2 h, 85Æ7% (s.d. ¼ 12Æ03) after
used at the beginning and end of the experiment. The 3 h and 80Æ3% after 15 h. The mean indentation
indentation hardness (HIT) had a measurement error at hardness of control specimens treated with water
the start of the experiment of 0Æ06%, and at the end of was 108Æ7% (s.d. ¼ 16Æ09) of the original hardness
1Æ1%. after 1 h, 99Æ09% (s.d. ¼ 18Æ98) after 2 h, 98Æ97%
(s.d. ¼ 11Æ24) after 3 h and 98Æ42% (s.d. ¼ 22Æ78)
after 15 h.
Experimental application of Coca-Cola
and Salivart

A further nine facial enamel specimens were sectioned

and the hardness measured as previously described. The
enamel specimens were demineralized with Coca- 1 hour
Cola
applied for 16Æ5 h and the enamel hardness exposure
re-tested. Salivart
,** was applied to the demineralized 110 2 hour
enamel for 3 min and the hardness re-assessed in an exposure
adjacent region parallel to and 50 lm from the last 3 hour
series of indentations. The re-mineralizing period of exposure
15 hour
3 min was chosen as representative of the time that the 100
exposure
low-viscosity aqueous solution might be expected to
remain in the mouth of a xerostomic patient (11).
Salivart
, an artificial saliva, contains potassium chlor-
90
ide, sodium chloride, potassium phosphate, calcium
chloride and magnesium chloride, but does not contain
fluoride. The pH of the Salivart
was measured as 6Æ48. Mean hardness

80

Results

Experimental application of Coca-Cola
only

Water Coca-cola®
Figure 1 shows a typical load displacement curve
Treatment
resulting from the indentation. Enamel does not
Fig. 2. Percentage mean indentation hardness (95% CI) for the
**Gebauer Co., Cleveland, OH, USA. water- and Coca-Cola
-treated specimens.

ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 26–30

 REDUCED ENAMEL HARDNESS AND ARTIFICIAL SALIVA
Experimental application of Coca-Cola
and artificial saliva
(Salivart
)
In a separate experiment, the mean hardness of nine
enamel specimens was determined (mean hardness
HIT ¼ 4Æ451 GPa, s.d. ¼ 0Æ59). This baseline enamel
hardness was comparable with that described in the
previous experiment. The specimens were then
demineralized in Coca-Cola
overnight and the mean
hardness HITwas reduced to 2Æ048 GPa (s.d. ¼ 1Æ279).
After a 3-min application an artificial saliva (Saliv-
art
), the mean hardness increased to 2Æ417 GPa
(s.d. ¼ 1Æ438), a mean increase of 18%. However,
the change in enamel hardness produced by the
Salivart
was highly variable, as seen by the large
standard deviation. It is easier to demonstrate large
changes in remineralization after profound deminer-
alization, and further experiments are necessary to test
the remineralization of the enamel after shorter
durations of Coca-Cola
treatment.
Discussion
The baseline indentation hardness (HIT) of untreated
enamel in our experiments was similar to that previ-
ously published by other researchers. Barbour et al. (12)
found that the median hardness for their untreated
enamel samples was 4Æ83 GPa (s.d. ¼ 0Æ66). Willems
et al. (13) found average values of enamel hardness was
3Æ39 GPa (s.d. ¼ 0Æ18). Cuy et al. (8) measured enamel
hardness which varied from 2Æ7 to 6Æ4 GPa, and there
was similar large variations in modulus from 47 to
120 GPa. They attributed the variation to be dependent
on the degree of mineralization of the enamel, local
variations resulting from enamel rods, tufts and
increased porosity near the enamel–dentin junction,
etc. Barbour et al. (12) found that the baseline hardness
of enamel in their study was 4Æ74
0Æ14 GPa. In our
study, we used indentation sites 50 lm apart in an
attempt to reduce the intra-tooth variation in mechan-
ical properties, but which would also avoid any inter-
action between indentation sites.
In our study, there was a reduction in indentation
hardness after static application of acidulated beverage
to enamel. There was also an increased standard
deviation in indentation hardness with acid exposure
over time, which may reflect differences in the rate of
dissolution of the enamel among the specimens.
Disparity and heterogeneity in the orientation and
ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 26–30
29
packing of the enamel prisms may affect the rate of
dissolution.
Gedalia et al. (1) exposed enamel blocks in the mouth
to Coca Cola
,and measured a 14Æ3–17Æ2% reduction
in Vickers hardness number (VHN) after 1 h. To limit
the effect of further applications of acidic beverage,
calcium and other ions may accumulate on the enamel
surface through diffusion of ions through the enamel.
Application of phosphoric acid solutions (a constituent
of Coca-Cola
) causes an increase in diffusion through
enamel (14). Those studies, which have shown rapid
enamel dissolution with acidic solutions, have used
rapid agitation [500 r.p.m., in the study by Barbour
et al. (12)] where a surface layer was prevented from
forming. Clinically, prolonged static contact of teeth
with acidic beverages may be much less harmful than
swishing the fluid around the mouth prior to swallow-
ing. There may be a buffering action provided by the
phosphate in the Coca-Cola
, which slowed the disso-
lution of the enamel in our experiments. Titrateable
acidity of a solution may therefore be a better measure
of erosion potential of a fluid than pH only (15).
Mahoney et al.’s (15) study was the first to use
computerized microhardness to measure the erosive
potential of various test beverages on the enamel of
primary teeth. In their study, a brief 10-min exposure
to a solution of pH 2Æ72 caused a reduction in enamel
hardness of 47% (from 4Æ7
0Æ3 GPa). They recorded a
much greater effect on the enamel hardness than in our
study, despite a similar pH of test solution and the
shorter duration of exposure. Primary teeth are more
prone to erosion than permanent teeth (16) because of
the disordered crystal structure in deciduous enamel
(17) and the increased porosity (18). Deciduous tooth
enamel has been shown to be softer than that of
permanent teeth (19). Younger children have a lower
salivary flow rate than adults (20), which further
increases their susceptibility to tooth erosion.
Many previous studies have pointed to the resistance
to erosion shown by enamel surfaces bathed by saliva,
e.g. the lingual surfaces of lower incisors. Enamel
demineralized by acidulated beverages has a significant
propensity to reharden following exposure to saliva in
the mouth (7) or to remineralizing solutions (21, 22).
However, our study has shown only partial recovery of
hardness values following application of artificial saliva,
which may be related to a less than optimum effect of
the remineralizing solution. Manufacturers should
consider including an optimum concentration of
30 H . D E V L I N et al.
fluoride in artificial saliva to improve the rehardening
of demineralized enamel. Tucker et al. (23) suggested
that differences in the distribution of carbonate and
magnesium in enamel could contribute to acid suscep-
tibility. Koulourides et al. (24) also found a large
variation in the rehardening process with a failure of
enamel re-hardening if the initial demineralization
process had progressed too far (i.e. below 150 KHN).
No such phenomenon could be detected in our data
with Salivart
causing very similar re-hardening in
samples demineralized below 2 GPa as above.
We conclude that applied acidulated beverage
reduced the indentation hardness of enamel. Although
this is an expected finding, the erosion was shown to
occur slowly over many hours. Knowing the time scale
involved allowed us to demonstrate a partial reminer-
alization using an artificial oral moisturizer, Salivart
.
References
1. Gedalia I, Dakuar A, Shapira L, Lewinstein I, Goultschin J,
Rahamim E. Enamel softening with Coca-Cola and rehard-
ening with milk or saliva. Am J Dent. 1991;4:120–122.
2. Johansson AK. On dental erosion and associated factors. Swed
Dent J Suppl. 2002;156:1–77.
3. Dugmore CR, Rock WP. The progression of tooth erosion in a
cohort of adolescents of mixed ethnicity. Int J Paediatr Dent.
2003;13:295–303.
4. Arnadottir IB, Saemundsson SR, Holbrook WP. Dental erosion
in Icelandic teenagers in relation to dietary and lifestyle
factors. Acta Odontol Scand. 2003;61:25–28.
5. Montecchi PP, Custureri V, Polimeni A, Cordaro M, Cosat L,
Marinucci S, Montecchi F. Oral manifestations in a group of
young patients with anorexia nervosa. Eat Weight Disord.
2003;8:164–167.
6. Wall GC, Magarity ML, Jundt JW. Pharmacotherapy of
xerostomia in primary Sjogren’s Syndrome. Pharmacothera-
py. 2002;22:621–629.
7. Amaechi BT, Higham SM. In vitro remineralisation of eroded
enamel lesions by saliva. J Dent. 2001;9:371–376.
8. Cuy JL, Mann AB, Livi KJ, Teaford MF, Weihs TP. Nanoin-
dentation mapping of the mechanical properties of human
molar tooth enamel. Arch Oral Biol. 2002;47:281–291.
9. Staines M, Robinson WH, Hood JAA. Spherical indentation of
tooth enamel. J Mater Sci. 1981;16:2551–2556.
10. Oliver WC, Pharr GM. An improved technique for determin-
ing hardness and elastic modulus using load and displacement
sensing indentation experiments. J Mater Res. 1992;7:1564–
1583.
ª 2006 Blackwell Publishing Ltd, Journal of Oral Rehabilitation 33; 26–30
11. Vissink A, ‘s-Gravenmade EJ, Gelhard TBFM, Panders AK,
Franken MH. Rehardening properties of mucin- or CMC-
containing saliva substitutes on softened human enamel.
Effects of sorbitol, xylitol and increasing viscosity. Caries Res.
1985;19:212–218.
12. Barbour ME, Parker DM, Allen GC, Jandt KD. Human enamel
dissolution in citric acid as a function of pH in the range 2Æ30<
or ¼pH< or ¼6Æ30 – a nanoindentation study. Eur J Oral Sci.
2003;111:258–262.
13. Willems G, Celis JP, Lambrechts P, Braem M, Vanherle G.
Hardness and Young’s modulus determined by nanoindenta-
tion technique of filler particles of dental restorative materials
compared with human enamel. J Biomed Mater Res.
1993;27:747–755.
14. Kuhar M, Cevc P, Schara M, Funduk N. Enhanced permeab-
ility of acid-etched or ground dental enamel. J Prosthet Dent.
1997;77:578–582.
15. Mahoney E, Beattie J, Swain M, Kilpatrick N. Preliminary
in vitro assessment of erosive potential using the ultra-micro-
indentation system. Caries Res. 2003;37:218–224.
16. Hunter ML, West NX, Hughes JA, Newcombe RG, Addy M.
Erosion of deciduous and permanent dental hard tissue in the
oral environment. J Dent. 2000;28:257–263.
17. Skaleri U, Ravnik Č, Cevc P, Schara M. Microcrystal arrange-
ment in human deciduous dental enamel studied by electron
paramagnetic resonance. Caries Res. 1982;16:47–50.
18. Silness J, Hegdahl T, Gustavsen F. Area of the organic-
inorganic interface of dental enamel. Acta Odontol Scand.
1973;31:123–129.
19. Johansson AK, Sorvari R, Birkhed D, Meurman JH. Dental
erosion in deciduous teeth–an in vivo and in vitro study.
J Dent. 2001;29:333–340.
20. Crossner CG, Hase JC, Birkhed D. Oral sugar clearance in
children compared with adults. Caries Research. 1991;25:201–
206
21. Kim JW, Jang KT, Lee SH, Kim CC, Hahn SH, Garcia-Godoy F.
In vivo rehardening of enamel eroded by a cola drink. Asd C J
Dent Child. 2001;68:122–124, 142.
22. Shannon IL, McCrary BR, Starcke EN. A saliva substitute for
use by xerostomia patients undergoing radiotherapy to the
head and neck. Oral Surg Oral Med Oral Pathol. 1977;44:656–
661.
23. Tucker K, Adams M, Shaw L, Smith AJ. Human enamel as a
substrate for in vitro dissolution studies: Influence of tooth
surfaces and morphology. Caries Res. 1998;32:135–140.
24. Koulourides T, Cueto H, Pigman W. Rehardening of softened
enamel surfaces of human teeth by solutions of calcium
phosphate. Nature 1961;189:226–227.
Correspondence: Dr Hugh Devlin, The Dental School, University of
Manchester, Higher Cambridge St., Manchester M15 6FH, UK.
E-mail: hugh.devlin@manchester.ac.uk