Examination by X-ray Photoelectron Spectroscopy of the Adsorption
of Chlorhexidine on Hydroxyapatite
R.N.S. SODHI, H.A. GRAD, and D.C. SMITH
Centre for Biomaterials, University of Toronto, 170 College Street, Toronto, Ontario, Canada M5S lA1
X-ray photoelectron spectroscopy (XPS) was used for determination of
the effects of chlorhexidine (CLX) solutions (0.2% and 1% solutions of
the digluconate salt) on the elemental composition of hydroxyapatite
surfaces. So that the nature ofthe adsorbed species after they were
washed with water could be identified, comparisons were made with
reference spectra for CHX obtained from a CHX digluconate film and
CHX dichloride powder. The XPS results clearly indicated the
retention of CHX moieties, which could be ascertained from the
spectra by the presence of N and Cl, features unique to CHX. The
spectral envelopes were virtually identical to those obtained from the
reference spectra. High-resolution C ls spectra also gave support for
the retention of CHX; however, the spectra differed from those of the
CIX digluconate film in that no feature attributable to the C-OH of
the gluconate anion was present, consistent with the view that the
CHX cation remains behind to form an electrostatic bond with the
phosphate groups of the hydroxyapatite. The N:C1 ratio for the
washed samples was found to be higher than that for the reference
samples and may be indicative of partial decomposition ofthe CHX.
Decomposition was also seen to be induced by x-ray exposure. While
the high-resolution spectra presented here do not directly address the
controversy on the mechanism for the anti-plaque efficacy of CHX,
they do provide the necessary basis for the application ofXPS to future
in vitro studies on the retention of CHX to dental surfaces.
J Dent Res 71(8):1493-1497, August, 1992
Introduction.
Chlorhexidine (C22H38N10C12), the structure ofwhich is shown in Fig.
1, is a bisbiguanide antiseptic which has a broad spectrum of
antimicrobial activity. It is an accepted therapeutic agent for the
control of plaque-related periodontal diseases, where it is generally
administered in the form of a 0. 12-0.2% solution of its digluconate
salt (C22H30N1OC12-2C 6H207) The anti-plaque efficacy of chlorhexi-
dine (CHX) at these concentrations has been attributed to its
bactericidal effect on the microbiota of dental plaque along with its
ability to bind to either salivary proteins and/or tooth enamel. More
specifically, two types ofmechanisms have been postulated for this
effect. The first is that a reservoir of CHX is bound to oral surfaces,
and active drug is slowly released into the surrounding medium,
thereby producing a prolonged bacteriostatic environment in the
mouth (Rolla et al., 1970; Gjermo et al., 1975; Gjermo, 1989). A
second theory postulates a more site-specific effect, and it is pro-
posed that after a rinse or topical application, CHX has an immedi-
ate, short bactericidal effect, followed by a prolonged bacteriostatic
effect due to adsorption onto the tooth enamel (Jenkins et al., 1988).
In both cases, however, it is recognized that the retention of CHX
upon oral surfaces plays an important role.
CHX has been shown to adsorb to tooth enamel and hydroxyapa-
tite, by both indirect (Rolla et al., 1970; Nordbo, 1972; Emilson et al.,
1973) and, more recently, direct means (Perdoketal., 1989; Busscher
et al., 1990). However, the nature ofthe adsorbed components after
a rinse with CHX has not been fully investigated. The purpose of
this study was to determine the effects ofsolutions of CX (0.2% and
Received for publication September 26, 1991
Accepted for publication March 24, 1992
Downloaded from jdr.sagepub.com by guest on July 9, 2011 For personal use only. No other uses without permission.
1% solutions of the digluconate salt) on the surface composition of
hydroxyapatite. Hydroxyapatite was chosen for our initial studies,
since it is a major component of tooth enamel as well as a material
used for dental implants (Lemons, 1990). Further, retention ofCHX
on implant material may allow for its use in a pre-treatment
procedure for hydroxyapatite and other materials. X-ray photoelec-
tron spectroscopy (XPS) was used for study of the nature of the
adsorbed species. XPS provides a convenient means of obtaining
detailed characterization of the treated surface (Andrade, 1985;
Briggs and Seah, 1990). It gives a quantitative elemental surface-
analysis to a depth of about 10 nm, and, unlike other surface-
analytical techniques, such as Auger electron spectroscopy, is one of
the few techniques capable of providing detailed chemical informa-
tion on that surface. Thus, the XPS spectra obtained in this study
allowed the nature of the species adsorbed to the hydroxyapatite
surface to be identified and also provided reference spectra as a
basis for future studies of the effect of a CHX rinse on the surface
composition of other implant materials.
Materials and methods.
Materials.-Sintered pure hydroxyapatite coupons (10 x 10 mm, 1
mm thick, Durapatite®, Sterling-Winthrop) were cleaned by sonica-
tion in detergent (2% Decon, 1 h) and thoroughly rinsed in double-
distilled, de-ionized water (DDDW). Duplicate cleaned coupons
were given either a CHX wash (0.2% or 1%, Hibitane® Gluconate,
Ayerst) or a DDDW wash. The latter treatments constituted a
control. All coupons then received a DDDW wash (two one-minute
rinses with shaking at 370C) and were allowed to dry overnight in
a laminar-flow hood. The samples were then wrapped in Al foil until
analyzed by XPS. Reference spectra for CHX were obtained from
two types of samples, namely, (i) hydroxyapatite samples washed
with a 1% solution of CHX digluconate and allowed to dry by
evaporation without a DDDW rinse, or (ii) CHX dihydrochloride
powder stuck onto sticky-sided Cu tape.
XPS analysis.-The XPS spectra were obtained on a Leybold
MAX 200 XPS system. Unmonochromatized Mg Ka x-ray radiation
was used as the excitation source. The source was run at 12 kV and 25
mA. Features in the resultant spectra due to excitation from the
weaker x-ray satellite lines which are also present in the
unmonochromatic source were subtracted by use of an algorithm
supplied with the instrument and based upon a program by van
Attekum and Trooster (1977). Atomic percentages of the elements
TABLE 1
ATOMIC % OF C, Cl, N, AND 0 FOR CHLORHEXIDINE
DIGLUCONATE FILMS (TWO RUNS) DETERMINED BY XPS,
IN COMPARISON WITH THE THEORETICAL VALUES
C Cl N 0
Theoretical 56.7 3.3 16.7 23.3
Experiment (1) 59.4 3.3 15.7 21.6
(2) 58.8 3.2 15.8 22.2
1493
1494 SODHI et al. J Dent Res August 1992
NH NH NH NH
/\ 1 11 1 11
Cl - -NNHCNHCN(CH2) NHCNHCNH - Cl

z 40

20- _ _
CHLORHEXIDINE
400 200 400 200

Fig. 1-Structure of chlorhexidine. L b d

present were derived from spectra run in a low-resolution mode (pass
energy = 192 eV) which were normalized to unit transmission of the
electron spectrometer by means of a routine provided by the manu-
facturer (Berresheimetal., 1991). The sensitivityfactors used (O Is =
0.78, N Is = 0.54, Ca 2p = 1.98, C Is = 0.34, Cl 2p = 1.08, and P 2p =
0.61) were empirically derived by Leybold for the normalized spectra.
For further information on the nature of the surface to be obtained,
the spectral regions of interest were also run in a high-resolution
mode (pass energy = 48 eV). Binding energies and peak areas were
obtained by use of the routines provided with the spectrometer. The
energy scale ofthe spectrometer was calibrated to the Ag 3d5,2 and Cu
2P3,2 peaks at 368.3 eV and 932.7 eV, respectively (Anthony and Seah,
1984; Briggs and Seah, 1990), and the binding energy scale was then
shifted to place the C Is feature present at 285.0 eV. Large-area
analysis was performed (2 x 4 mm or 4 x 7 mm) so that exposure of the
samples to the x-ray would be minimized while sufficient signal-to-
noise ratios could be obtained for the spectral features. The effect ofx-
ray exposure on CIX was checked on the CHX film by the source
being left on for 15 min in between the collection of high-resolution
spectra (O Is, N Is, C Is, and Cl 2p), which were obtained in
approximately three min. Examination of the high-resolution Cl 2p
spectra from the rinsed samples indicated minimal decomposition.
Results.
Fig. 2 shows survey spectra of hydroxyapatite and the CIX
digluconate film, both ofwhich constitute controls. The hydroxyapa-
80-
40-
40 2.
2 4 0 0
BINDING ENERGY (eV)
Fig. 2-Survey spectra of hydroxyapatite surfaces: (a) Hydroxyapatite
control; (b) + 1 % CHX digluconate, not rinsed (constitutes control); (c) + 0.2 %
CHX digluconate, rinsed; and (d) + 1% CHX digluconate, rinsed.
tite spectrum (Fig. 2a) shows the expected features (0, Ca, P)
associated with the hydroxyapatite, along with some unavoidable C
contamination. No features due to Ca and P are present in the
spectrum of the film (Fig. 2b), and the experimentally determined
atomic percentages of 0, N, C, and Cl are in good agreement with
those expected from the stoichiometryofCHXdigluconate (Table 1).
Therefore, it is concluded that this spectrum is representative of the
CHX digluconate salt. Also shown (Figs. 2c and 2d) are the spectra
of hydroxyapatite samples which had been washed with CHX
digluconate and rinsed. The rinsed samples show features from
both hydroxyapatite and CHX, as is evident from the Ca/P and N/Cl
peaks. Thus, it is clear that, while much of the CHX had been
removed by rinsing, some remained.
More information on the chemical nature of the surface can be
obtained from high-resolution spectra. Fig. 3 shows reference
spectra for CHXfrom (a) the digluconate film and (b) CHX dichloride
powder. The N is spectral envelope is identical for both compounds

TABLE 2
MEAN ELEMENTAL RATIOS [±( )] OBTAINED FROM THE XPS SPECTRA FOR THIS WORK (a), IN COMPARISON
WITH THE THEORETICAL RATIOS (b) AND THOSE OF PERDOK et al. (1989) ON DENTAL ENAMEL
Ca/P O/P N/P N/Cl C/N
(a) This work:
Hydroxyapatite 1.4 (1) 4.5 (2) -
CHX digluconate film - - - 4.8(i) 3.8 (1)
0.2% solution 1.4 (1) 4.3 (2) 0.53 (3) 6.1(4) 5.6 (6)
1.0% solution 1.3 (1) 4.5 (3) 1.1 (1) 6.6 (4) 3.9 (3)
(b) Theoretical:
Hydroxyapatite 1.67 4.33 - - -

CHX - - - 5 2.2
CHX digluconate - - 5 3.4
(c) Perdok et al. (1989):
Dental Enamel 1.3 4.2 - -

Dental Enamel + Peridex® 1.3 4.3 0.6 - 7.1

Downloaded from jdr.sagepub.com by guest on July 9, 2011 For personal use only. No other uses without permission.
Vol. 71 No. 8 X-RAY PHOTOELECTRON STUDIES ON CHLORHEXIDINE 1495

a A Ols a Nt
350o
CI" CF-

535 530 400 395 390

.b 0)b

CO 200 0
0
I-

0. 100-,
IU
'-
400 395 39~O a a
I-
z
a C is a 012n 290 205 280 205 200 195

300' BINDING ENERGY (eV)

I-
z 200- Fig. 4-Effect of x-ray exposure on CHX digluconate: (a) initial high-
resolution C is and C1 2p spectra; and (b) after 150 min ofx-ray exposure (Mg
i00-
202 KI, 300 watts). Formation of Cl- ion in addition to covalent Cl (Cl0) is clearly
2 90 285 205 200 195 seen from the Cl 2P3,2 peaks.
b 2b
300- sample is seen to be virtually identical and superimposable on that
200. 60, of the hydroxyapatite and has a binding energy of 531.1 eV, which
40 can be attributed to the phosphate. Table 2 summarizes elemental
100.
20- ratios found for these surfaces as well as for the controls.
290 285 205 200 195

BINDING ENERGY (eV) Discussion.

Fig. 3-High-resolution reference spectra of (a) 1% CHX film and (b)The aim of the present study was to determine the effect of a CHX
CHX dichloride powder. Deconvolution of CHX dichloride Cl 2p spectrum digluconate solution on hydroxyapatite and to identify the nature of
clearly shows the 2p spin-orbit pairs for Cl and CL. the adsorbed species after a wash with water. The XPS results
clearly showed retention of CHX moieties which could be ascer-
tained from the spectra by the presence of N is and Cl 2p features
and has a peak maximum at 399.9 eV. Chlorine is also evident, with unique to the CHX. The spectral envelopes and binding energies
the covalently-bonded C1 2p spin-orbit pair (Cl 2P312 = 200.5 eV) obtained in the high-resolution spectra were virtually identical to
common to both compounds. The dichloride also shows a peak due those obtained in the reference spectrafromboththe CHXdigluconate
to the Cl- ion (Cl 2p3, = 197.8 eV), which, from the deconvolution of film and the CLIX dichloride (once the contribution of Cl- was taken
the spectrum (Fig. 3), was found to be of an intensity equal to that into account). However, the N:Cl ratio on the washed samples
ofthe covalent Cl, as would be expected. The N:Cl ratio obtained for (Table 2) did differ from that obtained from the reference spectra
the dichloride, as for the digluconate, was also found to be in and that which would be expected from stoichiometry. This point
excellent agreement with that expected from stoichiometry. No will be discussed later.
attempt has been made to deconvolute the C is spectra, which will While the C is feature is not entirely unique to the CHX, in that
consist of contributions from C in several types of bonding situa- contribution from carbon contamination may also be present, the
tions. The C is spectral envelope of the digluconate is seen to be spectral envelope also supports the notion that CHX was retained.
different from that of the dichloride, due to the presence of the C-OH In particular, the intensity of the high binding-energy shoulder at
bonds in the former, which would appear at 1.5 eV above the main 288.4 eV arising from the imide carbon unique to the CHX moiety
hydrocarbon (C-C) peak at 285.0 eV. Also within this region, remained relatively constant in the normalized spectra. The C is
contributions from C-Cl and C-N wouldbe expected. A high binding- spectral envelope in the washed samples was more similar to that
energy shoulder (288.4 eV) is common to both spectra, and this is of the CHX dichloride spectrum as opposed to that for the CHX
attributed to the -NH-C(=NH)-NH- imide carbon. The 0 is spec- digluconate film, in that the spectrum did not show an obvious
trum for the digluconate shows a single broad peak at 532.4 eV. feature attributable to C-OH bonds which would be apparent if the
The effects of x-rays on these samples are illustrated in Fig. 4, gluconate anion had been present. Thus, it appears that the rinsing
which shows the high-resolution C is and Cl 2p spectra obtained with water washed away the gluconate anion, while some of the
after a total exposure of 150 min to the x-rays. A conversion of the CLX cation remained behind to form a strong electrostatic bond
covalent Cl to an ionic Cl- with exposure is clearly demonstrated. with the phosphate group of the hydroxyapatite. This is consistent
The digluconate also showed loss of oxygen. These effects were seen with previous observations of CHX binding to other anionic surfaces
to increase with exposure, as demonstrated in the plots shown in such as phosphate groups in lipopolysaccharides and carboxylic
Fig. 5. Interestingly, the absolute intensity of the N is feature was groups in proteins (Gjermo, 1989). Further indication for the
seen to increase with x-ray exposure (Fig. 5). removal of the gluconate anion due to water rinsing is given in the
Finally, Fig. 6 shows the high-resolution 0 is, N is, C is, and Cl O is spectra (Fig. 7), which is basically that of the phosphate.
2p spectraforthe samples which werewashedwith CHXdigluconate As mentioned above, the ratio of N:C1 for the washed samples
and rinsed. The relative intensities have been normalized to the N was higher than that found in the controls and that expected from
is feature. The spectra of the CHX digluconate film are also shown, stoichiometry (Table 2). This difference can be explained by partial
for reference, plotted on the same scale. Fig. 7 shows the 0 is decomposition of CHX, leading to the loss of Cl or Cl-containing
spectra for the sample rinsed with 1% CHX, superimposed on those species, which were then washed away. For example, decomposi-
ofthe hydroxyapatite control. The scale has been normalized to the tion of CLX digluconate by heat/light has been noted, leading to the
P 2p peak for these samples. The 0 is spectrum for the rinsed formation of p-chloroaniline (Kohlbecker, 1989). Loss of p-
Downloaded from jdr.sagepub.com by guest on July 9, 2011 For personal use only. No other uses without permission.
1496 SODHI et al. J Dent Res August 1992

c
a1)
NIs
a)

A N -

.0 *0
0)
T Cl
m

a-
0)
I-

20 40 60 80 100 120 140 160 395

X-ray Exposure Time ( min) I-

en
Cl 2p
Fig. 5-Change in the absolute intensity of the N Is, 0 Is, and the C1 2p z
(covalent) features ofthe CHX film with x-ray exposure. The value for Cl was
extracted from the total Cl present by means of the supplied curve-fitting -1
routines.
-J
LU

chloroaniline from the surface would remove one N atom with every
Cl atom. Thus, partial decomposition of CHX, leading to the loss of
p-chloroaniline, with the rest of the molecule remaining bound,
would increase the observed N:Cl ratio. If this amounted to 20% of
the total amount ofadsorbed CHX, then an increase in the N:Cl ratio
from 5:1 to 6:1 would be observed. The ratio of C:N was also seen to 195
be somewhat higher than that which would be expected for CHX. BINDING ENERGY(eV)
This was especially so for the 0.2%-rinsed sample. This increase can
be accounted for either from pick-up of hydrocarbon contamination Fig. 6-High-resolution spectra of sample washed with (a) 0.2% CHLX
from the atmosphere after drying and prior to XPS analysis or from digluconate solution and rinsed, and (b) 1% CHX digluconate solution and
re-exposure of C contamination that was originally present on the rinsed. Also shown are spectra for the CHX digluconate control (c). The
hydroxyapatite surface. A large increase in the amount of C relative relative intensities of the spectra have been normalized to the N is feature.
to Cl has been noted for CHX on tooth enamel (Perdok et al., 1989).
This was attributed to the adsorption of hydrocarbon additives onto
the treated tooth enamel from the commercially-available CHX
used, as well as from contamination.
-9 ._. Decomposition was also observed to be induced by the x-ray
exposure. The phenomena observed here probably arose from
dissociative electron attachment onto the chlorine (Zhou and White,
-9
1990), the low-energy electrons required for this being produced by
the bremsstrahlung radiation inherent in the use of an
unmonochromatized x-ray. The released Cl- is then able to attack
(: the gluconate site or bind to the phosphate surface. Thus, care must
be taken that the spectra are obtained in as short a time as possible
I- so that this decomposition will be minimized. Examination of the
high-resolution Cl 2p spectra (Fig. 6) obtained on the rinsed samples
indicated that this was indeed the case.
(n
I-

The removal of the gluconate anion, which was apparent from

z the decrease in 0 Is intensity as well as from decrease in the 'C-OH'
w region ofthe C Is spectrum, from the surface would also explain the
z increase in the absolute intensity of the nitrogen feature which was
observed, and indicates that the anion was on the outer surface
w above the cation.
More coverage ofthe hydroxyapatite was obtained with the more
-i concentrated solution of CHX, as can be seen in the reduction of the
J normalized 0 Is peak intensity for the sample rinsed with 1% CHX
(Fig. 6). Some idea of the relative surface coverage can be obtained
from the N/P ratio (Table 2). If it is assumed that the remaining
536 534 532 530 528 CIX forms a uniform overlayer, a rough estimate of the thickness
(d) can be made by means of the relationship (Andrade, 1985)
BINDING ENERGY (eV)
Fig. 7-0 is spectra for i% CHX washed and rinsed sample ( ) IN I'
= . exp(-d/I esinO)
I-exp(-d/X *sinO)
compared with that for the hydroxyapatite (HA) control (...... ). The relative
intensities have been normalized to the P 2p feature of these samples. where IP and IN are the measured intensities (corrected by the
Downloaded from jdr.sagepub.com by guest on July 9, 2011 For personal use only. No other uses without permission.
Vol. 71 No. 8 X-RAY PHOTOELECTRON STUDIES ON CHLORHEXIDINE
appropriate sensitivity factors) of P and N found in the rinsed
samples; IPO and INO are the concentrations of P and N found in
hydroxyapatite and CHX, respectively, which can be related to the
measured intensities (corrected by the appropriate sensitivity fac-
tor) of the control spectra obtained for hydroxyapatite and the CHX
digluconate film; Xp and XN are the inelastic mean free-paths for the
appropriate photoelectrons in the overlay, and 0 is the photoelec-
tron takeoff angle (= 90°).
Aratio of0.70 was obtained for IPO/INO, while values of 3.5 nm and
2.8 nm were obtained for XAP and XN) respectively, by use ofthe Ashley
formula (Ashley, 1980; Andrade, 1985), with 2.4 as the M/pn value
(Busscheret al., 1990). Use ofthese values, along withthe measured
'AN ratios (Table 2), gave an estimate for overlayer thickness of 0.9
nm and 1.8 nm for the samples which were subjected to solutions of
0.2% and 1% CHX followed by DDDW, respectively. It must be
emphasized, however, that these estimates are very rough because
of the number of assumptions which have had to be made. In
particular, the model assumes a smooth surface-morphology and a
uniform adsorbed overlayer, neither of which is necessarily correct.
Indeed, angle-resolved measurements, which would have allowed
for a more accurate estimate ofthe amount and type of coverage (i.e.,
patchy vs. uniform overlayers), were not performed because of
surface roughness. Thus, firm conclusions cannot be drawn with
regard to coverage from these data. Work is currently under way on
the addition of CHLX to polished metal surfaces which would allow
meaningful angle-resolved XPS to be performed, thereby allowing
for a more reliable estimate of type of coverage. A further source of
error would arise from the estimate of IN0 from the CHX digluconate
film, the intensity of which was seen to increase as the gluconate
anion was lost.
In spite ofthe above reservations, it is of interest to compare the
obtained coverage with that of Busscheret al. (1990), who examined
the coverage of a commercial CHX rinse (Peridex®-0. 12% solution of
CIX) on dental enamel. They also assumed a uniform coverage and
obtained an estimate for film thickness of 0.8 nm, which is in fair
agreement with the estimated coverages obtained here with the
stronger solutions of CHX. It must be noted, however, that the C is
signal was used in their work for estimation of the film thickness,
which would not be unique to CHX and thus lead to an overestima-
tion of the CHX film thickness. Perdok et al. (1989), who also
examined Peridee on dental enamel, reported a C/N ratio of 7.1,
which was considerably higher than that which would be expected
from CHX (see Table 2), and this was mainly attributed to the co-
adsorption of hydrocarbon additives in Peridex® as well as from
contamination.
Comparison ofthe elemental ratios obtained here on hydroxyapa-
tite with those on dental enamel (Perdok et al., 1989), which are
shown in Table 2, suggest that the adsorption of CHX was very
comparable on the two substrates. Further, the Ca/P and O/P ratios
obtained on both substrata were basically the same. Thus, as noted
for the ground and polished dental enamel (Perdoket al., 1989), one
sees the expected O/P ratio, but the Ca/P ratio is indicative of a Ca-
deficient surface. The large deficiency ofCa at the surface observed
by XPS was attributed to the possibility of excess phosphate ions
being present during the final polishing step on the dental enamel
and, presumably, would be applicable to hydroxyapatite.
Downloaded from jdr.sagepub.com by guest on July 9, 2011 For personal use only. No other uses without permission.
1497
In summary, the work presented here shows that CHX was
retained on a surface of hydroxyapatite after being rinsed with
water and supports the notion that it formed a strong electrostatic
bond with the surface. While this work does not directly address the
controversy with regard to the mechanism for the anti-plaque
efficacy of CHX (Jenkins et al., 1988; Gjermo, 1989), the high-
resolution spectra provide the necessary basis for the application of
XPS to future in vitro studies on the retention of CHX to dental
surfaces and its interaction with saliva.
REFEIRENCES
Andrade JD (1985). X-ray photoelectron spectroscopy. In: Andrade JD,
editor. Surface and interfacial aspects of biomedical polymers. Vol. 1.
New York: Plenum Press, 105-195.
Anthony MT, Seah MP (1984). XPS: energy calibration of electron spectrom-
eters. 1-An absolute, traceable energy calibration and the provision of
atomic reference line energies. Surf InterfAnal 6:95-106.
AshleyJC (1980). Inelastic interactions oflow-energyelectrons with organic
solids: simple formulae. IEEE Trans Nucl Sci NS-27:1454-1458.
Berresheim K, Mattern-Klossen M, Wilmers M (1991). A standard form of
spectraforquantitative ESCA-analysis. FreseniusJAnal Chem 341:121-
124.
Briggs D, Seah MP, editors (1990). Practical surface analysis. 2nd ed. Vol.
1. Chichester (England): John Wiley and Sons.
Busscher HJ, van der Mei HC, Genet MJ, Perdok JF, Rouxhet PG (1990).
XPS determination ofthe thickness of adsorbed mouthrinse components
on dental enamel. Surf InterfAnal 15:344-346.
Emilson CG, Ericson TH, Heyden G, Magnusson BC (1973). Uptake of
chlorhexidine to hydroxyapatite. J Periodont Res 8 (Suppl 12):17-21.
Gjermo P(1989). Chlorhexidine andrelatedcompounds. JDentRes 68 (Spec
Iss):1602-1608.
Gjermo P, Bonesvoll P, Hjeljord LG, Rolla G (1975). Influence of variation
of pH of chlorhexidine mouth rinses on oral retention and plaque-
inhibiting effect. Caries Res 9:74-82.
Jenkins S, Addy M, Wade W (1988). The mechanism of action of chlorhexi-
dine. A study of plaque growth on enamel inserts in vivo. J Clin
Periodontol 15:415-424.
Kohlbecker G (1989). Toxische verunreinigungen in chlorhexidin-digluconat.
Dtsch Zahndrztl Z 44:273-276.
Lemons JE (1990). Dental implant biomaterials. JAm DentAssoc 121:716-
719.
Nordbo H (1972). The affinity of chlorhexidine for hydroxyapatite and tooth
surfaces. Scand J Dent Res 80:465-473.
Perdok JF, van der Mei HC, Genet MJ, Rouxhet PG, Busscher HJ (1989).
Elemental surface concentration ratios and surface free energies of
human enamel after application of chlorhexidine and adsorption of
salivary constituents. Caries Res 23:297-302.
Rolla G, Lee H, Schi0tt CR (1970). The affinity of chlorhexidine for
hydroxyapatite and salivary mucins. J Periodont Res 5:90-95.
vanAttekum PMT, TroosterJM (1977). Removal ofx-ray satellites from Mg
Ka excited photoelectron spectra. J Electron Spectrosc Relat Phenom
11:363-370.
Zhou X-L, White JM (1990). Photon- and electron-induced chemistry of
chlorobenzene on Ag(111). J Chem Phys 92:5612-5621.