1. The document describes a study on the preparation and properties of silver-doped phosphate antibacterial glasses. 2. Silver-doped phosphate glasses were prepared by melting and their antibacterial effects against bacteria were investigated using agar disk-diffusion assays. 3. The tested silver-free and silver-doped glasses demonstrated different antibacterial effects. Silver-doped glasses showed increased inhibition of bacterial growth with higher silver content. The release of phosphate and silver ions from dissolving glasses contributed to their antibacterial properties.

1. Solid State Sciences 13 (2011) 981e992
Contents lists available at ScienceDirect
Solid State Sciences
journal homepage: www.elsevier.com/locate/ssscie
Study on the preparation and properties of silver-doped phosphate
antibacterial glasses (Part I)
A.A. Ahmed a, A.A. Ali a, *, Doaa A.R. Mahmoud b, A.M. El-Fiqi a
a
Glass Research Department, National Research Centre, Dokki, Cairo, Egypt
b
Natural and Microbial Products Laboratory, National Research Centre, Dokki, Cairo, Egypt
a r t i c l e i n f o a b s t r a c t
Article history: Silver-doped phosphate antibacterial glasses were prepared by the melting method. The antibacterial
Received 14 October 2010 effects of some undoped and silveredoped glasses of compositions 65P2O5e10CaOe(25ex) Na2O,
Received in revised form 70P2O5e20CaOe(10ex) Na2Oand (70ex) P2O5e30CaO, (where x ¼ 0, 0.5, 1.2 Ag2O), against Staphylococcus
26 January 2011
aureus, Pseudomonas aeruginosa and Escherichia coli micro-organisms using agar disk-diffusion assays
Accepted 11 February 2011
Available online 19 February 2011
were investigated. The structures of some glasses were studied by XRD, FT-IR, and UVeVIS spectroscopy.
The variation of pH with dissolution rate was studied. The tested silver-free and silver-doped glasses
demonstrated different antibacterial effects against the tested micro-organisms. For silver-free glasses, an
Keywords:
Antibacterial glasses
increase in inhibition zone diameter (zone of no bacterial growth) was seen with the decrease in water pH.
Silver-doped phosphate-based glasses Silver-doped glasses showed an increase in inhibition zone diameter with increasing Ag2O content. The
Glass dissolution low pH produced by glass dissolution was certainly a critical factor for glass antibacterial effect. The more
the phosphate ions released the lower is the pH and the greater the antibacterial effect.
Ó 2011 Elsevier Masson SAS. All rights reserved.
1. Introduction Silver ions are effective against a broad range of micro-organisms
including G- bacteria e.g., Pseudomonas aeruginosa, yeast e.g.,
Glasses in the P2O5eCaOeNa2O system have a chemical Candida albicans, and Gþ bacteria e.g. Staphylococcus aureus [4,5].
composition similar to that of the inorganic phase of bone. These Therefore, silver ions have been commercially used to take
glasses consist of PO4 tetrahedra, which can be attached to advantage of its antibacterial properties e.g. silver nitrate, colloidal
a maximum of three neighboring tetrahedra forming a three silver, and certain other silver compounds are among the most
dimensional network as in vitreous P2O5 [1]. Adding metal oxides to generally used bactericidal agents. A large number of healthcare
the glass leads to a depolymerization of the network, with the products now contain silver ions, principally due to its low toxicity
breaking of PeOeP linkages and the creation of non-bridging to human cells and high antibacterial effect. Such products include
oxygens. The modifying cations can provide ionic cross-linking silver-coated catheters, and wound dressings [6]. Phosphate-based
between the non-bridging oxygens of two phosphate chains, thus glasses are materials of technological importance due to their
increasing the bond strength of this ionic cross-link and improving superior physical properties compared to silicate glasses e.g., low
the mechanical strength and chemical durability of the glasses [2]. melting temperatures, low glass transition and low softening
These phosphate-based glasses are a unique class of materials in that temperatures, and high thermal expansion coefficients [7,8]. Thus
they are completely degradable; whereas silica-based glasses are PBGs can be prepared and processed easily at lower temperatures.
relatively stable to hydrolysis. Furthermore, the degradation of In addition, phosphate-based glasses enjoy a range of composi-
phosphate-based glasses can be tailored to suit the end application tional and structural possibilities (ultra, meta, pyro, and ortho) that
and the rate at which they hydrolyze can vary quite considerably [3]. facilitate tailoring chemical and physical properties of interest for
Various types of silver-doped inorganic antibacterial materials specific technological applications. Controlled-release glasses
have been developed e.g. zeolites, calcium phosphate, silica gel, and (CRGs) were first developed in the 1970s primarily for use in food
borosilicate glass and some of them are now in commercial use. production industries [9]. Drake and Allen [10] found that PBGs
with a suitable composition would dissolve in water with zero-
order rate constant, and by controlling the composition it was
* Corresponding author. possible to produce glasses which would completely degrade in
E-mail address: ali_nrc@hotmail.com (A.A. Ali). water from hours to years thus can, over a prolonged period, release
1293-2558/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved.
doi:10.1016/j.solidstatesciences.2011.02.004

2. 982 A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992
any additional constituents incorporated into them. Hence it has 2.3. FT-IR absorption measurements
been possible to use pellets of CRGs containing metal ions such as
copper and cobalt, as pesticides, fungicides and in animal feeds. The FT-IR absorption spectra of some selected undoped and
CRGs are manufactured in a similar way to conventional soda-lime silver-doped glasses (0, 0.5, 1 and 2 mol% Ag2O) were recorded at
silica glasses in that the constituents are heated to temperatures room temperature in the frequency range 400e4000 cmÀ1 using an
above 1000 C, then cast into various forms such as solid blocks, infrared spectrometer (Jasco FT-IR 6100). The measurements were
powder, granules, tubes, fiber or wool. The incorporation of well- made by the KBr disc technique in which discs were prepared by
known silver, copper or zinc antibacterial metal ions in several glass mixing and grinding a small amount of glass powder with spec-
systems has a proven negative influence on the growth of bacteria troscopic grade anhydrous KBr powder and then pressed under
and fungi [11,12]. Where in presence of an aqueous medium or vacuum and pressure of 6 ton/cm2 into clear disks (1.2 cm in
moisture, the glass will gradually dissolve and at the same time, diameter and about 0.5 mm in thickness). All measurements were
silver, copper, or zinc ions are released during its dissolution to recorded with a resolution of 4 cmÀ1.
provide an antibacterial effect. Generally, antibacterial glasses can
be manufactured either by addition of an antibacterial agent to the 2.4. UVeVIS absorption measurements
glass batch prior to their manufacture or by post-treatment
processes e.g. ion-exchange or surface coating. This work is an UVeVIS absorption spectra were measured for some undoped
attempting to prepare and study the antibacterial effect of high and silver-doped glasses (0, 0.5, 1 and 2 mol% Ag2O). Polished glass
dissolution silver-doped phosphate glasses. samples having dimensions 3 cm  1 cm and of the same thickness
(2 mm) were scanned in the range from 200 to 1000 nm using
a UVeVIS spectrometer (T80þ, PG instruments Ltd.).
2. Experimental
2.5. pH measurements
2.1. Glass preparation
The pH changes of the distilled water during the dissolution of
All batches were prepared from chemically pure grade chem-
some undoped and silver-doped glasses were measured at every
icals in the powder form. P2O5 was introduced as (NH4H2PO4)
hour and up to 6 h using IQ 140 pH-meter (IQ Inc. USA). The pH
(99.0% Merck), Calcium oxide (CaO) as Calcium carbonate
electrode was calibrated using pH calibration standards (Colourkey
(CaCO3) (99.5% SRL) Sodium oxide (Na2O) as sodium
Buffer Solutions BDH, UK).
carbonate (Na2CO3) (99.5% s.d. fine-chem), and silver oxide (Ag2O)
as silver nitrate (AgNO3) (99.9% SRL).The appropriate amounts of
2.6. Antibacterial activity test
the starting materials of each batch equivalent to 50 g glass were
accurately weighed, thoroughly mixed and then transferred to
The antibacterial activities of undoped and silver-doped
porcelain crucibles. Before melting, the batches were calcined
P2O5eCaOeNa2O glasses were tested against bacterial species of
slowly in an electric muffle furnace at a temperature in the range of
American Type Culture Collection (ATCC); S. aureus (ATCC, 25923),
350e550 C in order to get rid of the gaseous decomposition
E. coli (ATCC, 25922), and P. aeruginosa (ATCC, 27853) using the agar
products of the batch materials, e.g. H2O, NH3, NO2, and CO2 and to
disk-diffusion assays.
minimize the evaporation tendency of P2O5. Calcination was
continued until the decomposition of the batch materials and
3. Results
evolution of gaseous products came to an end. All the batches were
melted in disposable porcelain crucibles inside an electrically
3.1. Glass forming region (GFR)
heated furnace in the range 800e1200 C. The melting time was
continued for 1 h to 2 h depending upon the chemical composition.
The glass forming regions and the compositions prepared in the
During melting, the melt was stirred manually by swirling about
systems P2O5eCaOeNa2Oex Ag2O and P2O5eCaO-x Ag2O, x ¼ 0.5, 1
several times to ensure homogeneity and to get ride of gas bubbles.
and 2 mol % are illustrated in Figs. 1e3. Clear circles denote
The melt was then cast on a preheated stainless steel plate in the
homogeneous, transparent and colorless glasses as confirmed by
form of rectangular slabs which subsequently annealed in a muffle
XRD. Black circles denote samples that showed metallic silver
furnace maintained at a temperature in the range 200e450 C for
particles precipitation. Fig. 1 and Fig. 2 showed that the composi-
20 min. The muffle furnace was then switched off and the glass
tions containing ! 60 mol% of P2O5 in the quaternary system
samples were left overnight to cool slowly to room temperature.
P2O5eCaOeNa2Oex Ag2O, x ¼ 0.5 and 1 mol%, formed homoge-
The visible characteristics e.g. color; transparency, and homoge-
neous, transparent and colorless silver-doped glasses, whereas it
neity, of all samples prepared in this work were investigated using
was not possible to obtain homogeneous glasses for the composi-
the normal visual observations.
tions containing 55 mol% of P2O5 since these compositions
showed precipitations of metallic silver particles. Also it can be
2.2. XRD measurements seen from Fig. 3 that the compositions containing ! 65 mol% of
P2O5 in the quaternary system P2O5eCaOeNa2Oe2Ag2O formed
To ensure the glassy state, some selected samples were char- homogeneous, transparent and colorless silver-doped glasses.
acterized with powder X-ray diffraction technique which is Nevertheless, among compositions containing 60 mol% of P2O5,
commonly used to verify the amorphous state of glassy materials. only three compositions which contain 10, 15 and 20 mol% of Na2O
In the XRD spectra of glassy materials a halo is seen instead of formed homogeneous, transparent and colorless silver-doped
diffraction peaks. the samples were finely ground in an agate glasses. For other compositions containing 5, 25, 30 and 35 mol %
mortar and X-ray diffraction spectra were obtained using a Bruker Na2O, homogenous silver-doped glasses could not be obtained
D8 Advance X-ray diffractometer at room temperature with Ni- since these compositions showed precipitations of metallic silver
filtered Cu Ka radiation (l ¼ 0.15418 nm), generated at 40 kV and particles. Overall, A glass forming region containing ! 60 mol% of
40 mA. Scans were performed with a step size of 0.02 and a step P2O5 was observed in the quaternary system P2O5eCaOeNa2Oex
time of 0.4 s over an angular range 2q from 4 to 70 . Ag2O, x ¼ 0.5, 1 and 2 mol%.

3. A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992 983
P2O5 P2O5
0
100 A 0
5 100 A
10
95 B 5
90 C 0.5 mol% Ag2O 95 B
15 10
20
85 D 90 C 2 mol % Ag 2O
80 E 15
25
75 F
85 D
30 20
70 G 80 E
35 25
.
40
65 H
30
75 F
. ........
. 60 I
45
55 J 70 G
50 35
65
60
55
..
50 K
45 L
40 M
35 N
45
40
. ... 65 H
60 I
55 J
70 50
30
CaO 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Na O + 0.5 Ag O 50 K
2 2 CaO 0
15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 5 10 15 20 25 30 35 40 45 50 Na2O + 2 Ag2O
Mol % 2
11 10 9 8 7 6 5 4 3 1
Fig. 1. Ternary phase diagram showing the GFR and the glass forming compositions in Mol %
the quaternary system P2O5eCaOeNa2Oe0.5 Ag2O.
Fig. 3. Ternary phase diagram showing the GFR and the glass forming compositions in
the quaternary system P2O5eCaOeNa2Oe2Ag2O. B: Homogeneous glass (Transparent
X-ray diffraction measurements were performed on some and Colorless). C: Metallic silver particles precipitation.
P2O5eCaOeNa2OeAg2O glasses containing 2 mol % Ag2O. The X-ray
diffraction patterns obtained for these glasses are displayed in
Fig. 4. As shown in this figure, the X-ray diffraction patterns showed position, whereas few bands shifted to slightly higher frequencies.
no sharp peaks thus indicating the absence of formation of any It can also be seen that the intensities of some bands increased with
crystalline phases and ensuring the amorphous nature of these the increase in the Ag2O content. Six main absorption bands located
prepared samples. in the regions at about 445e472 cmÀ1, 720e750 cmÀ1,
907e918 cmÀ1, 1024e1050 cmÀ1, 1100e1109 cmÀ1 and
1298e1332 cmÀ1 along with two shoulders at about 530 cmÀ1 and
3.2. FT-IR absorption spectroscopy 1170 cmÀ1 were observed in all FT-IR spectra of P2O5eCaOeNa2O
glasses, Table 2.
FT-IR absorption spectroscopy was used to detect any change in
the structure of some P2O5eCaOeNa2O and P2O5eCaO glasses,
Table 1, that may have occurred as a result of introducing Ag2O into 3.3. UVeVIS absorption spectra
these glasses and to obtain essential information concerning the
arrangement of the phosphate structural units in the phosphate Fig. 6a, b and c shows the UVeVIS absorption spectra recorded
glass network. As shown in Fig. 5a, b and c no new absorption bands in the wavelength range 200e1000 nm at room temperature for
were detected on the addition of x Ag2O, x ¼ 0.5, 1 and 2 mol% 70P2O5e20CaOe(10ex) Na2Oex Ag2O, (70ex) P2O5e30CaO-x Ag2O
to the P2O5eCaOeNa2O base glasses. Also, it can be seen that most and 65P2O5e10CaOe(25ex) Na2O-x Ag2O glasses, x ¼ 0, 0.5, 1 and
of the bands observed in the base glasses do not show significant 2 mol%. From this figure, it can be seen that the undoped and silver-
shifts with the Ag2O addition. Most of bands appeared on the same doped glasses reveal no absorption peaks in the visible region. Two
absorption peaks were observed in the ultraviolet region, one at
about 210 nm was observed for all undoped glasses. A small red
P2O5
0
100 A
5
95 B
10
90 C 1 mol % Ag2O
15
85 D
20
80 E
25
75 F
30
70 G
35
65 H
40
60 I
45
55 J
50
50 K
CaO 0 5 10 15 20 25 30 35 40 45 50 Na2O + 1 Ag2O
11 10 9 8 7 6 5 4 3 2 1
Mol %
Fig. 2. Ternary phase diagram showing the GFR and the glass forming compositions in Fig. 4. The XRD patterns for some P2O5eCaOeNa2OeAg2O glasses containing 2 mol %
the quaternary system P2O5eCaOeNa2Oe1 Ag2O. Ag2O.

4. 984 A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992
a
x=2
Absorbance (a.u.)
x=1
450 x = 0.5
530
911 1017 1096 1298
735 1170 1644
x=0
400 600 800 1000 1200 1400 1600 1800 2000
wavenumber (cm-1)
b x=2
Absorbance (a.u.)
x=1
x = 0.5
445
528 917 1024 1332
11091170 1644
743
x=0
400 600 800 1000 1200 1400 1600 1800 2000
Wavenumber (cm-1)
c
x=2
Absorbance (a.u.)
x=1
x = 0.5
448
530
907 1024
720 1100 1300 1644 Fig. 6. UVeVIS absorption spectra (a): 70P2O5e20CaOe(10ex) Na2Oex Ag2O,
1170
(b):(70ex) P2O5e30CaO-x Ag2O and (c) :65P2O5e10CaOe(25ex) Na2O-x Ag2O glasses,
x=0 x ¼ 0, 0.5, 1 and 2 mol%. The inset shows the region 200e320 nm.
400 600 800 1000 1200 1400 1600 1800 2000 3.4. pH measurements and dissolution rates
wavenumber (cm-1)
The pH changes and dissolution rates measured during the
Fig. 5. FT-IR absorption spectra for (a): 70P2O5e20CaOe(10ex) Na2Oex Ag2O,
dissolution of some undoped and Ag2O-doped P2O5eCaOeNa2O
(b):(70ex) P2O5e30CaO-x Ag2O and (c) :65P2O5e10CaOe(25ex) Na2O-x Ag2O glasses,
x ¼ 0, 0.5, 1 and 2 mol%. and P2O5eCaO glasses in distilled water at 37 C for different time
intervals up to 6 h are listed in Tables 1 and 3 and Fig. 7. As shown in
shift (a shift to longer wavelength) in the position of this absorption Table 3 and Fig. 7 a fast drop in pH of distilled water (w5.5) was
band was observed (in all UV-spectra of Ag2O-doped glasses) with seen through the first hour of glass dissolution and then the pH
increasing Ag2O content and strong absorption peak at about decreased slowly with increasing time of dissolution. The glass
230 nm was observed for all silver-doped glasses. with the highest dissolution rate, H3, shows a higher decrease in pH

5. A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992 985
a 6.0 b 6.0
G5 G7
5.5 5.5
G5Ag0.5 G7Ag0.5
G5Ag1 G7Ag1
5.0 5.0
G5Ag2 G7Ag2
4.5 4.5
pH
pH
4.0 4.0
3.5 3.5
3.0 3.0
2.5 2.5
2.0 2.0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
Dissolution Time ( Hours) Dissolution Time ( Hours)
c 6.0
H3
5.5
H3Ag0.5
H3Ag1
5.0
H3Ag2
4.5
pH
4.0
3.5
3.0
2.5
2.0
0 1 2 3 4 5 6 7
Dissolution Time ( Hours)
Fig. 7. pH variation with time during the dissolution of (a): 70P2O5e20CaOe(10ex) Na2Oex Ag2O, (b):( 70ex) P2O5e30CaO-x Ag2O and (c):65P2O5e10CaOe(25ex) Na2O-x Ag2O
glasses, x ¼ 0, 0.5, 1 and 2 mol%.
than that showed by the glasses having lower dissolution rates. was seen with increasing Ag2O content in the glass. The biggest zone
That means the pH is dependant on the glass dissolution rate. of inhibition among silver-doped glasses was observed for the
highest silver releasing glass, namely H3Ag2 against S. aureus micro-
organism. Figs. 13e15 show the variation of the inhibition zone
3.5. Antibacterial activity
diameter with dissolution rate for silver free glasses. The biggest
zone of inhibition among silver free glasses was observed for the
The antibacterial effects of undoped and2 mol % Ag2O-doped
glass with the highest dissolution rate, namely H3 against S. aureus
P2O5eCaOeNa2O and P2O5eCaO glasses were tested in vitro against
micro-organism as shown in Figs. 13e15.
S. aureus as Gþ, P. aeruginosa and E. coli as G- micro-organisms using
agar disk-diffusion assays. The results of agar disk-diffusion assays
conducted for 24 h at 37 C are shown in Figs. 8 and 9. The anti-
bacterial activity of the glass was confirmed by the presence of an 4. Discussion
inhibitory zone (i.e. zone of no bacterial growth) around each tested
glass disk. The measured inhibition zone diameters (minus the 4.1. Glass forming region
diameter of the glass disk, 12 mm) are given in Table 1. It can be seen
from Figs. 8 and 9 that all tested glasses (even silver free glasses) A glass forming region which contains ! 60 mol% of P2O5 was
demonstrated different antibacterial effects against the tested observed in the quaternary system P2O5eCaOeNa2Oex Ag2O,
micro-organisms as indicated by the clear zone around each glass x ¼ 0.5, 1 and 2 mol%, whereas compositions containing 55 mol%
disk. Figs. 8 and 9 also show that the glass antibacterial effect of P2O5 showed metallic silver particles precipitation. The
depends on the glass composition, Ag2O content and type of the P2O5eCaOeNa2OeAg2O glasses were prepared under normal
tested micro-organism. S. aureus was found to be the most suscep- melting conditions without any special precautions.
tible micro-organism to the tested antibacterial glasses. The degree Since silver is a noble element, its oxide is easily reduced at high
of susceptibility of the tested micro-organisms to the tested anti- temperatures. The decomposition temperature of Ag2O under
bacterial glasses was in this order S. aureus P. aeruginosa E. coli. ambient conditions is calculated from its thermodynamic data to be
Figs. 10e12 show the variation of the inhibition zone diameter 421 K. It is well known that silver may exist in glass in one or more
with the Ag2O content in the all studied glasses. As displayed in than one of its common states (Ag0, Agþ or Ag2þ). The solubility of
Figs. 10e12, a gradual increase in the inhibition zone diameter (the Ag2O in glass melts is an important factor for effective production of
antibacterial activity is proportional to the size of inhibition zone) Ag2O containing glasses as silver is known to have different degrees

6. 986 A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992
Fig. 8. Photos of Petri dishes after conducting agar disk-diffusion assays at 37 C for 24 h with (a) S. aureus, (b) P. aeruginosa and (c) E. coli as test micro-organisms. (For base glasses).
of solubility in glass melts. In phosphate melts, silver has good positions. It is accepted that the structure of ultraphosphates,
solubility. metaphosphates, and polyphosphates are dominated by (Q3 Q2),
The Ag2O added to the glass batch over its solubility is reduced (Q2), and (Q2 Q1) units, respectively.
to metallic silver and it is of no use to increase Ag2O content in the The FT-IR spectra of some undoped and Ag2O-doped
glass beyond its solubility limit. It is known that the solubility of P2O5eCaOeNa2O glasses shown in Fig. 5 are typical of phosphate-
Ag2O in glass melts is limited by thermodynamic factors and is based glasses showing the characteristic absorption bands of PO4
highly dependent on the glass composition, temperature during groups.
melting, and the oxygen potential [13,14]. Therefore the glass The FT-IR spectra of these glasses appear to be dominated by
composition should be selected carefully because it is difficult to metaphosphate (Q2) and ultraphosphate structural units (Q3).
control the oxygen potential during the melting process. Three main spectral regions can be distinguished in the FT-IR
spectra of the glasses as follows: 400e800 cmÀ1, 800e1200 cmÀ1
4.2. Structure of studied glasses and 1200e1400 cmÀ1. In the spectral region ( 400e800 cmÀ1), the
band at w 450e470 cmÀ1 and the shoulder at w 530 cmÀ1 may be
It is well known that the phosphate network structure consists attributed to harmonics of bending vibrations of OePeO and O]
of a series of PO4 tetrahedral units connected by three bridging PeO linkages.
oxygens. The network connectivity can be described in terms of Another band in the frequency region (400e800 cmÀ1) is the
a Qn distribution as shown, where n is the number of bridging absorption band at about 720e750 cmÀ1. This band may be
oxygens. Q0 corresponds to isolated tetrahedra (orthophosphate attributed to the symmetric stretching vibrations of the PeOeP
groups), Q1 to end groups (pyrophosphate), Q2 to middle groups linkages, ys (PeOeP) modes [15]. This absorption band at about
(metaphosphate) and Q3 to branching groups (ultraphosphate). 750 cmÀ1 shifts towards lower frequency with increasing the
When a modifier oxide is added, disruption of the main network concentration of Ag2O. The variation of the frequency of PeOeP
occurs with the modifier oxide cations occupying interstitial bonds with increasing Ag2O content is consistent with breakage of

7. A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992 987
Fig. 9. Photos of Petri dishes after conducting agar disk-diffusion assays at 37 C for 24 h with (a) S. aureus, (b) P. aeruginosa and (c) E. coli as test micro-organisms. (For glasses
containing 2 mol % Ag2O).
cyclic PeOeP bonds in the glass when the Ag2O acts as a network concentrations of the Ag2O. The band at w 1100 cmÀ1 can be
modifier. The FT-IR absorptions in the spectral region associated with an overlap of several modes like stretching of the
800e1200 cmÀ1 was found to be sensitive for the different meta- PO3 terminal and PO2 middle groups. In addition, the y3 (F2) mode
phosphate groups in the form of chain-, ring-, terminal groups of the ortho anion can also contribute to the absorption at
[16,17]. This area in the FT-IR spectra of all the studied glasses is 1100 cmÀ1 [15,22]. The band at w 1100 cmÀ1 was also assigned to
characterized by the presence of four absorption bands at about asymmetric stretching mode of chain-terminating Q1 groups, yas
910e920, 1017e1050, w1100 and 1160e1170 cmÀ1. The absorption (PO3)2-. The shoulder at about 1160e1170 cmÀ1 may be attributed
band at w 910 cmÀ1 is assigned to the asymmetric stretching to symmetric stretching mode of the two non-bridging oxygen
vibrations of PeOeP groups linked with linear metaphosphate atoms in the Q2 tetrahedral sites, ys (PO2)À mode in metaphosphate
chain [15]. Phosphate-based glasses with a metaphosphate struc- groups [23].
ture can have a chain and/or ring structure. The occurrence of the The spectral region 1200e1400 cmÀ1 showed a strong broad
yas (PeOeP) at around 900 cmÀ1 in all the samples studied is an absorption band in the range 1298e1330 cmÀ1. The appearance of
indication of a chain structure, as the analogous vibration occurs at this band (at w1300 cmÀ1) could be due to the presence of a frac-
around 1000 cmÀ1 in ring structures [18]. The shift of PeOeP tion of Q3 tetrahedral units which are the most characteristic
asymmetric stretching vibration around 900 cmÀ1 to higher wave feature of an ultraphosphate glass. This band appeared in all the FT-
numbers [19], always indicates increase of covalency proportion of IR spectra of the studied phosphate-based glasses and it is attrib-
the PeOeP bond and strengthening of glass structure with uted to the asymmetric stretching of the doubly bonded oxygen, yas
improved chemical durability. The weak absorption band at (P]O) modes, which are only present in the glasses with a P2O5
1017e1050 cmÀ1 is attributed to a normal vibration mode (y3 content 50 mol% [15,16]. Osaka et al. [24] have shown that PO4
symmetric stretching) of the PO43- groups [20,21]. The position of units have two bridging oxygen bonds along with two non-bridging
this band is constant and does not change by the addition of varying bonds such as P]O and PeOÀ, which are in resonance with each

8. 988 A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992
Table 1
Glass compositions, Dissolution Rate, Density and the measured inhibition zone
diameters (minus the diameter of the glass disk, 12 mm).
Glass Glass Density D.R Inhibition zone
Composition code No. (g cmÀ2 hÀ1) diameter (mm)
*10À4
P2O5 CaO Na2O S. P. E. coli
aureus aeruginosa
70 20 10 G5 2.3937 3.91 27 25 22
70 20 9.5 G5Ag0.5 2.4186 3.52 31 28 25
70 20 9 G5Ag1 2.4415 3.12 34 31 28
70 20 8 G5Ag2 2.4827 2.55 38 35 32
70 30 0 G7 2.4098 3.24 25 22 19
69.5 30 0 G7Ag0.5 2.4302 2.95 29 28 23
69 30 0 G7Ag1 2.4523 2.67 32 30 26
68 30 0 G7Ag2 2.4910 2.19 36 33 29
65 10 25 H3 2.4229 6.46 30 28 25
65 10 24.5 H3Ag0.5 2.4440 5.85 34 31 29
65 10 24 H3Ag1 2.4647 5.32 37 35 33
65 10 23 H3Ag2 2.5066 4.36 41 38 36
Fig. 11. Variation of IZD with Ag2O content for (70ex) P2O5e30CaOex Ag2O glasses,
other. Therefore, the FT-IR spectra, as in the present study, are split x ¼ 0, 0.5, 1 and 2 mol%.
into two bands with a higher energy and strong double bond
character (1298e1330 cmÀ1) and a lower energy band to the yas (PeOeP) and ys (PeOeP) modes shift to higher
(w1100 cmÀ1). This means that the P]O double bond is more frequencies as Na2O is replaced by CaO showing an intensified
strongly localized in the central position of phosphate groups. This PeOeP bonds in the glass structure and showing that the covalency
assumption is confirmed by the larger relative intensity of the proportion in the bond of the metal ions with the non-bridging
1298e1330 cmÀ1 envelope compared with that of the band at w oxygen on the [PO4] tetrahedron units increased probably because
1100 cmÀ1. of the increased polarizability of the calcium ions which has
The weak absorption band at around 1615e1644 cmÀ1 observed a smaller electrovalency in the Ca..O bond than in the Na..O bond.
in all spectra of phosphate glasses prepared in this work is attrib- This explains the improvement in chemical durability with
uted to the bending vibration modes of OeH groups, d (H2O) modes increasing CaO content [19]. The band shift can be explained by the
[21]. This band demonstrates the presence of small amounts of increase in covalent character of PeOeP bonds, indicating that the
water absorbed from air moisture (due to the hygroscopic character PeOeP bonds are strengthened as the Na2O is substituted by CaO.
of the phosphate glasses) during the preparation of the KBr pellets This explains the improvement in the chemical durability of these
for infrared measurements [16]. The band positions of various glasses as the CaO content increases.
phosphate structural groups observed for the studied glass samples Overall, the FT-IR spectra of the base P2O5eCaOeNa2O showed
were found to be well within the ranges reported in the literature that the shift of yas (PeOeP) and ys (PeOeP) to higher frequencies as
for different phosphate-based glasses. The positions and assign- CaO replaces Na2O is witnessed with improved chemical durability
ments [15, 19, and 23] of the IR absorption bands observed in the of P2O5eCaOeNa2O glasses and this was explained in terms of:
FT-IR spectra of the studied glasses are summarized in Table 2.
As the P2O5 content increases the band around 1320 cmÀ1 (i) Strengthened Ca..O bond formed by the non-bridging oxygen
become broader and this is typically for ultraphosphate glasses. with the Ca ion introduced on the [PO4] tetrahedron;
Also it can be seen from the FT-IR spectra of the base glasses that (ii) Strengthened PeOeP bond formed by the bridging oxygen
the absorption bands at about w 910 cmÀ1 and 720 cmÀ1 assigned between two [PO4] tetrahedrons.
Fig. 10. Variation of IZD with Ag2O content for 70P2O5e20CaOe(10ex ) Na2Oex Ag2O Fig. 12. Variation of IZD with Ag2O content for 65P2O5e10CaOe(25ex) Na2Oex Ag2O
glasses, x ¼ 0, 0.5, 1 and 2 mol%. glasses, x ¼ 0, 0.5, 1 and 2 mol%.

9. A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992 989
Fig. 13. Variation of IZD with D.R for some P2O5eCaOeNa2O glasses in case of S. aureus Fig. 15. Variation of IZD with D.R for some P2O5eCaOeNa2O glasses in case of E. coli
micro-organism. micro-organism.
According to the dissolution mechanism of the phosphate A strong absorption peak at about 230 nm was observed for all
glasses [25], these two factors determine the dissolution rate of the silver-doped glasses as shown in Fig. 6a which is presumably due to
phosphate glasses by both limiting the velocity of hydration reac- Agþ ions. Ahmed and Abdallah [29] observed three absorption
tion and the network breakage reaction, respectively. bands at 305 nm, 350 nm and 420 nm in the UVeVIS spectra of
silver-containing soda-lime silica glass prepared by the ion-
4.3. UVeVIS spectra exchange process. They assigned these bands to silver ions Agþ,
elemental silver Ag0 and silver crystallites (Ag0)n (n is the number
It is well known [26,27] that the optical properties of silver of silver atoms forming the crystallite) respectively. Paje et al. [30]
atoms are entirely determined in the visible region by their free observed a broad absorption band centered at w260 nm for
electrons (Ag: [Kr] 4d10 5s1), while that of silver ions are deter- a silver-doped silicate glass prepared by the melt-quenching
mined in the ultraviolet region (Agþ: [Kr] 4d10 5s0). Thus absorp- technique. However, they also observed the same peak in the host
tions in the UVeVIS spectral regions can help in the elucidation of glass, so they did not attribute it to silver. Borsella et al. [31] per-
silver states within a glass network. formed UVeVIS spectroscopic investigations on a range of silver
In the present work, the UVeVIS absorption spectra of some concentrations in soda-lime glass prepared by the ion-exchange
undoped and silver-doped P2O5eCaOeNa2O glasses revealed no process. In contrast to Paje et al. [30] they observed an absorption
absorption peaks in the visible region, whereas two absorption band at about 268 nm for silver ions in the soda-lime glass. The
peaks were observed in the ultraviolet region. A peak at bout absorption was ascribed to transitions involving the promotion of
210 nm was observed for all undoped glasses as shown in Fig. 6a, an electron from the 4d10 ground state to the 4d9 5s1 state. Jimenez
b and c except for the binary glass, G7, where this peak was et al. [32] observed an absorption band centered at about 275 nm
observed at about 240 nm and it was weak and broad as shown in for silver-doped aluminophosphate glass and they ascribed it to
Fig. 6b. This band (either at about 210 nm or at about 240 nm) is silver ions. It was reported [28] that besides the pronounced silver
attributed to electronic absorption given by the phosphate glass plasmon resonances that appear between 400 and 500 nm, the
network (the threshold absorption of the host glass matrix) [28]. electronic transitions involving the Agþ ion give rise to absorption
Table 2
Assignment of IR absorption bands of studied glasses.
Band Assignment
frequency (cmÀ1)
450e470 Bending vibrations of OePeO bonds, d (PO2) modes.
530 Fundamental bending vibrations of O]PeO bonds,
d (O]PeO) modes.
660 Bending mode of PO4 units.
720e750 Symmetric stretching vibrations of PeOeP linkages,
ys (PeOeP).
900e920 Asymmetric stretching vibrations of PeOeP linkages,
yas (PeOeP) linked with linear metaphosphate chain.
1020e1050 Vibration mode (y3 symmetric stretching) of the PO43-
groups.
1100 Asymmetric stretching mode of chain-terminating
Q1 groups, yas (PO3)2-.
1170 Symmetric stretching mode of the tow non-bridging
oxygen atoms in the Q2 tetrahedral sites, ys (PO2)À
mode in metaphosphate groups Q2.
1300e1330 Asymmetric stretching of doubly bonded oxygen,
yas (P]O) modes.
1610e1640 Bending vibration modes of OeH groups in HeOeH,
Fig. 14. Variation of IZD with D.R for some P2O5eCaOeNa2O glasses in case of P. d (H2O) modes.
aeruginosa micro-organism.

10. 990 A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992
bands located between 200 and 230 nm, whereas the electronic the pH and produces acidic media (since more acidic phosphate
transitions of metallic Ago atoms appear in the 250e330 nm ions are released).
spectral range. The pH value of the medium (in which the glass dissolves) and
Accordingly, it is reasonable to ascribe the absorption band at inhibition nation zone diameter were found to be related to the
around 235 nm which appeared in the UVeVIS spectra of the Ag2O- glass dissolution rate (Table 1 and Figs. 13e15). The dissolution rate
doped P2O5eCaOeNa2O glasses to electronic transitions involving of P2O5eCaOeNa2O glasses depends on the contents of P2O5, CaO
Agþ ions, [4d10 / 4d9 5 s1]. Although these transitions are parity- and Na2O in the glass. From the results of dissolution of such
forbidden, they are partially allowed in a solid due to vibrational glasses, it was found that the glass dissolution rate is directly
coupling. The observed red shift of the absorption edge (the shift proportional to either P2O5 or Na2O contents and inversely
from 230 to 245 nm) associated with increasing Ag2O content, proportional to CaO content. The glass dissolution study showed
appears to be related to several effects. The progressive introduc- that the dissolution rate of these glasses as shown in Table 1 and
tion of silver oxide is responsible for the red shift of the absorption Figs. 13e15 are in this order H3 G5 G7 i.e. H3 has the fastest
edge. Regarding silver and its role as a network modifying oxide dissolution rate, whereas G7 has the slowest dissolution rate in this
implies that the red shift (caused by Ag2O addition) is related to the glass series. As shown in Tables 1 and 3 a decrease in the pH was
incorporation of silver ions in glass structure. Also this shift may be seen with the increases in the dissolution rate since more acidic
arise from an increase in the concentration of non-bridging phosphate ions are released.
oxygens with increasing silver content or may be due to the It is well known that the acidity or alkalinity of the medium
decrease of the smallest AgeO distance existing in the family of affects the growth of bacteria. The pH affects the rate of enzyme
silver sites. action and plays a role in determining the ability of bacteria to grow
or survive in particular environments. Most bacteria survive near
4.4. Antibacterial activity neutral conditions and grow optimally within a narrow range of pH
between 6.7 and 7.5. Since the optimal pH of all tested bacteria in
An antibacterial activity of a substance is an indication of its this study is close to neutral. Thus, the decrease in pH during the
ability to kill bacteria or inhibit their growth. In the present study, dissolution of P2O5eCaOeNa2O glasses could explain the bacterial
the tested silver free and silver-doped phosphate-based glasses growth inhibition produced by such glasses. The more the phos-
showed different antibacterial effects against S. aureus, P. aerugi- phate ions released the lower is the pH and the greater the anti-
nosa and E. coli micro-organisms as shown in Figs. 8 and 9. The bacterial effect. Thus the low pH produced by glass dissolution was
antibacterial effect is indicated by the clear zone (zone of no certainly a critical factor for glass antibacterial effect. Overall, the
bacterial growth) around each glass disk. As shown in these figures, antibacterial effect of silver free glasses was influenced by glass
the silver-doped glasses demonstrated more potent antibacterial composition, glass dissolution rate and the dissolution conditions
effects than the silver free glasses. According to the results of the in the glass surroundings. This might explain at least some of the
agar disk-diffusion assays, all glass compositions examined were differences in the antibacterial action of glass with varying chem-
able to inhibit the growth of Gþ and G- bacteria. The antibacterial ical compositions. The antibacterial activity of silver free glasses
effect as shown in Figs. 8 and 9 varies according to the glass was closely related to the dissolution rate of the glasses because
composition, Ag2O contents and the type of the micro-organism. high dissolution rates cause a decrease of pH or an increase of the
The degree of susceptibility of the tested micro-organisms to the ion concentration of the media, and this resulted in antibacterial
tested antibacterial glasses was in this order S. aureus P. activity. Therefore, the results of silver free glass dissolution, their
aeruginosa E. coli. The explanations of the antibacterial effects of pH changes during dissolution and their antibacterial effects
silver free and silver-doped glasses are given in the following correlate well with each other. The antibacterial effect of the silver
sections: free glasses may be attributed to other factors [34e36] e.g. the ionic
strength of the medium where high concentrations of calcium,
4.4.1. Antibacterial activity of silver free glasses sodium and phosphates ions likely to be released from the glass
In the present work, some P2O5eCaOeNa2O glasses were used during dissolution could cause perturbations of the membrane
as control samples in the antibacterial activity test. It was thought potential of bacteria and such as osmotic effects caused by the
that these silver free glasses would not show any antibacterial nonphysiological concentration of ions such as sodium, calcium,
effects against the tested micro-organisms. Interestingly, these and phosphate ions dissolved from the glass and lead to change the
silver free glasses namely, G5, G7, and H3 produced inhibitory zones osmotic pressure in the vicinity of the glass.
of different sizes (Fig. 8). The exact mechanism of the antibacterial action of these silver
This antibacterial effect of these silver free glasses can be free glasses is unknown. Therefore, we conclude that the
explained in terms of the glass composition, the glass dissolution
rate, and the pH changes of the medium. Valappil et al. [33] Table 3
pH values of studied glasses in different times.
observed a zone of inhibition for S. aureus, MRSA and C.difficile in
testing the antibacterial activity of gallium free Glass Code No. pH
45P2O5e16CaOe39Na2O glass. They attributed the antibacterial 1h 2h 3h 4h 5h 6h
effect of this gallium free glass to the change in pH during the glass G5 2.98 2.87 2.81 2.72 2.65 2.54
degradation. Pickup et al. [23] investigated the antibacterial activ- G5Ag0.5 3.05 2.93 2.87 2.79 2.71 2.63
ities of a Ga-doped solegel phosphate-based glasses of composition G5Ag1 3.14 3.06 2.98 2.90 2.81 2.74
(CaO)0.30 (Na2O)0.20Àx (Ga2O3)x (P2O5)0.50 where x ¼ 0 and 0.03 mol G5Ag2 3.30 3.21 3.15 3.04 2.95 2.85
G7 3.07 2.95 2.88 2.80 2.73 2.62
%. They observed a small zone of inhibition (7 mm) for the gallium
G7Ag0.5 3.12 3.05 2.96 2.89 2.80 2.69
free glass and they attributed it to either a change in pH as the glass G7Ag1 3.25 3.13 3.03 2.95 2.87 2.77
dissolves or by reduced water activity as ions leach out. G7Ag2 3.41 3.32 3.23 3.13 3.01 2.92
The glass compositions investigated contain 65 and 70 mol% of H3 2.84 2.75 2.68 2.60 2.52 2.41
P2O5 and the remainder is made up of CaO and Na2O. Accordingly, H3Ag0.5 2.91 2.82 2.75 2.67 2.58 2.50
H3Ag1 3.01 2.93 2.85 2.77 2.68 2.59
due to their high P2O5 contents, these glasses have acidic compo- H3Ag2 3.20 3.11 3.01 2.94 2.85 2.76
sition and this means that the dissolution of such glasses changes

11. A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992 991
antibacterial effect observed with these glasses can be explained by shown to be associated with the cell wall [47], cytoplasm and the
the dramatic changes in the physicochemical characteristics of the cell envelope [48]. Chappell and Greville [49] acknowledged that
culture medium (pH, ionic strength, and osmotic pressure) which low levels of Agþ ions collapsed the proton motive force on the
occur as a consequence of the glasses dissolution. Thus the anti- membrane of bacteria, and this was reinforced by Mitchell’s
bacterial action of glass is influenced by its chemical composition work [50]. Dibrov et al. [51] showed that low concentrations of
and the dissolution conditions in its surroundings. Agþ ions induced a massive proton leakage through the bacterial
membrane, resulting in complete de-energization and, ultimately,
4.4.2. Antibacterial activity of silver-doped glasses cell death. Overall, there is consensus that surface binding and
In the present work, the addition of Ag2O to P2O5eCaOeNa2O or damage to membrane function are the most important mecha-
P2O5eCaO glasses has been found to potentate its antibacterial nisms for the killing of bacteria by Agþ ions. The results of anti-
activity. Silver-doped glasses showed increased antibacterial bacterial activity showed that the silver free and silver-doped
activities (depending upon the Ag2O content) more than silver free glasses exhibited different antibacterial effects against the tested
glasses as shown in Figs. 8 and 9 and as indicated by the measured bacteria and the sensitivity of GÀ and Gþ bacteria to the anti-
inhibition zone diameters Table 1 and Figs. 10e12. This increase is bacterial glasses was different. A remarkable difference was seen
attributed to the release of the Agþ ions which are well known as between G- bacterium (E. coli) and Gþ bacterium (S. aureus),
antibacterial metal ions [37]. while a small difference was found between the two G- bacteria. It
An increase in the antibacterial activity as represented by the is known that the cell wall of G- bacteria is composed of high
increase in the inhibition zone diameter was seen with increasing proportion of phospholipids, lipopolysaccharides and proteins
the Ag2O content in the glass as displayed in Figs. 10e12. This was in (the cell wall of G- bacterium is chemically more complex than
a good agreement with the results of Agþ ions release seen in water that of Gþ bacteria), whereas peptidoglycan is the major
where an increase in concentrations of silver ions released from component of the cell wall of Gþ bacteria. This fact would possibly
glass into water was seen with the increase in Ag2O content. contribute to the difference of antibacterial effects between GÀ
Generally, in an aqueous medium or in presence of moisture, the and Gþ bacteria.
silver-doped glass gradually dissolves depending on its dissolution
rate and during its dissolution, the silver ions (the antibacterial
active agents) incorporated into its structure are released into the
medium and inhibit the growth of bacteria. Thus the mechanism 5. Conclusion
for antibacterial action of silver-doped glasses is bacterial growth
inhibition by the silver ions released from the glass. Bellantone 1- A glass forming region containing ! 60 mol% of P2O5 was
et al. [38] investigated the antibacterial effects of Ag2O-doped observed in the quaternary P2O5eCaOeNa2Oex Ag2O system,
bioactive glasses on S. aureus, E. coli, and P. aeruginosa. The anti- x ¼ 0.5, 1 and 2 mol%.
bacterial action of the silver-doped bioactive glass was attributed to 2- Generally, all the prepared silver free and silver-doped glasses
the leaching out of Agþ ions from the glass matrix. Kim et al. [39] were (as observed visually) homogeneous, free from solid
investigated the antimicrobial effects of various ceramics against inclusions, transparent and colorless. For 2 mol% silver-doped
E. coli using viable count and growth rate studies. They concluded P2O5eCaOeNa2O glasses, the XRD patterns indicated the
that Agþ ions interfered with the metabolism of the micro- absence of formation of any crystalline phases and thus XRD
organism, thus inhibiting its growth. ensured the amorphous nature of these glasses.
The exact mechanism of antibacterial action of silver ions is still 3- Measurements of pH changes during dissolution of silver-free
unknown. Antibacterial mechanisms of silver ions might differ and silver-doped glasses in water revealed a decrease of water
according to the species of bacteria. Generally, most antibacterial pH with increasing time of glass dissolution. It was found that
agents exert their antibacterial action by four principal modes of the magnitude of the pH drop increases with the increase in
action. These modes include: inhibition of bacterial cell wall glass dissolution rate.
synthesis, inhibition of protein synthesis, inhibition of synthesis of 4- Silver-doped glasses showed less pH drop than silver-free
bacterial RNA and DNA or inhibition of a metabolic pathway. In glasses. In agar disk-diffusion assays, all the tested silver-free
bacteria, silver ions are known to react with bacterial nucleophilic and silver-doped glasses demonstrated different antibacterial
amino acid residues in proteins, and attach to sulphydryl, amino, effects (depending on the glass composition and the type of the
imidazole, phosphate, and carboxyl groups of membranes or tested micro-organism) against S. aureus, P. aeruginosa and E.
enzymes, resulting in protein denaturation [40]. Silver is also coli micro-organisms as indicated by the clear zone (zone of no
known to inhibit a number of oxidative enzymes such as yeast bacterial growth) around each tested glass disk.
alcohol dehydrogenase [41] the uptake of succinate by membrane 5- For silver-free glasses an increase in bacterial growth inhibition
vesicles [42] and the respiratory chain of E. coli, as well as causing zone diameter was observed with the increase in the glass
metabolite efflux [43] and interfering with DNA replication. Holt dissolution rate and with the decrease in pH, whereas for silver-
and Bard [44] examined the interaction of silver ions with the doped glasses an increase in bacterial growth inhibition zone
respiratory chain of E. coli. They found that an addition of 10 mM diameter was observed with increasing Ag2O content.
AgNO3 to suspended or immobilized E. coli resulted in stimulated 6- S. aureus as a Gþ bacterium was found to be the most suscep-
respiration before death, signifying uncoupling of respiratory tible micro-organism to the tested antibacterial glasses. The
control from ATP synthesis. This was a symptom of the interaction degree of susceptibility of the tested micro-organisms to the
of Agþ with enzymes of the respiratory chain. Feng et al. [4] studied tested antibacterial glasses was found in this order S. aureus P.
the antibacterial effect of silver ions on E. coli and S. aureus and aeruginosa E. coli.
suggested that the antibacterial mechanism was due to DNA not 7- There is a good agreement between the antibacterial activities of
being able to replicate, and proteins becoming inactivated after silver-free glasses, their dissolution rates and their pH changes
contact with Agþ ions. during their dissolution. Also there is a good agreement
One of the primary targets of Agþ ions, specifically at low between the antibacterial activities of silver-doped glasses and
concentrations, appears to be the Naþ-translocating NADH: the concentrations of silver ions released into water during their
ubiquinone oxidoreductase system [45,46]. Silver has also been dissolution.

12. 992 A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992
References [24] A. Osaka, K. Takahashi, M. Ikeda, J. Mater. Sci. Lett. 3 (1984) 36.
[25] B. C Bunker, G.W. Arnold, J.A. Wilder, J. Non-Cryst. Solids 64 (1984) 291.
[26] P. Chakraborty, J. Mater. Sci. 33 (1998) 2235.
[1] J. Clement, J.M. Manero, J.A. Planell, G. Avila, S. Martinez, J. Mater. Sci. Mater.
[27] R.H. Doremus, J. Chem. Phys. 42 (1965) 414.
Med. 10 (1999) 729.
[28] J. Lu, J.J. Bravo, A. Takahashi, M. Haruta, S.T. Oyama, J. Catal. 232 (2005) 85.
[2] I. Ahmed, M.P. Lewis, I. Olsen, J.C. Knowles, Biomaterials 25 (3) (2004) 491.
[29] A.A. Ahmed, E.W. Abd Allah, J. Am. Ceram. Soc. 78 (10) (1995) 2777e2784.
[3] K. Franks, I. Abrahams, G. Georgiou, J.C. Knowles, Biomaterials 22 (2001) 497.
[30] S.E. Paje, J. Llopis, M.A. Villegas, J.M. Fernandez, Appl. Phys. A 63 (1996) 431.
[4] Q.L. Feng, J. Wu, G.Q. Chen, F.Z. Cui, T.N. Kim, J.O. Kim, J. Biomed. Mater. Res. 52
[31] E. Borsella, G. Battaglin, M.A. García, F. Gonella, P. Mazzoldi, R. Polloni,
(2000) 662e668.
A. Quaranta, Appl. Phys. A 71 (2000) 125.
[5] M. Kawashita, S. Tsuneyama, F. Miyaji, T. Kokubo, H. Kozuka, K. Yamamoto,
[32] J.A. Jimenez, S. Lysenko, G. Zhang, H. Liu, J. Electronic Mater. 36 (2007).
Biomaterials 21 (2000) 393e398.
[33] S.P. Valappil, D. Ready, E.A. Abou Neel, D.M. Pickup, W. Chrzanowski,
[6] A.P. Adams, E.M. Santschi, M.A. Mellencamp, Vet. Surg. 28 (1999) 219e225.
L.A. O’Dell, R.J. Newport, M.E. Smith, M. Wilson, J.C. Knowles, Adv. Funct.
[7] Y. He, D.E. Day, Glass Technol. 33 (1992) 214.
Mater. 18 (2008) 732e741.
[8] A.E. Marino, S.R. Arrasmith, L.L. Gregg, S.D. Jacobs, G. Chen, Y. Duc, J. Non-Cryst
[34] P. Stoor, E. Söderling, J.I. Salonen, Acta Odontol Scand. 56 (1998) 161e165.
Solids 289 (2001) 37.
[35] E. Munukka, O. Lepparanta, M. Korkeamaki, M. Vaahtio, T. Peltola, D. Zhang,
[9] C.F. Drake, Pest Sci. 9 (1978) 441.
L. Hupa, H. Ylanen, J.I. Salonen, M.K. Viljanen, E. Eerola, J. Mater. Sci. Mater.
[10] C.F. Drake, W.M. Allen, Biochem. Soc. Trans. 13 (1985) 520.
Med. 19 (2008) 27e32.
[11] A.M. Mulligan, M. Wilson, J.C. Knowles, J. Biomed. Mater. Res. 67A (2003)
[36] O. Lepparanta, M. Vaahtio, T. Peltola, D. Zhang, L. Hupa, M. Hupa, H. Ylänen,
401e412.
J.I. Salonen, M.I.K. Viljanen, E. Eerola, J. Mater. Sci. Mater. Med. 19 (2008)
[12] I. Ahmed, D. Ready, M. Wilson, J.C. Knowles, J. Biomed. Mater. Res. 79A (2006)
547e551.
618e626.
[37] A.B.G. Lansdown, J. Wound Care 11 (4) (2002) 125.
[13] D.A. McKeown, H. Gan, I.L. Pegg, J. Non-Cryst. Solids 351 (2005) 3826e3833.
[38] M. Bellantone, H.D. Williams, L.L. Hench, Antimicrobial. Agents. Chemother.
[14] M. Bosca, L. Pop, G. Borodi, P. Pascuta, E. Culea, J. Alloys Compounds 479
46 (2002) 1940e1945.
(2009) 579e582.
[15] R.C. Lucacel, A.O. Hulpus, V. Simon, I. Ardelean, J. Non-Cryst. Solids 355 (2009) [39] T.N. Kim, Q.L. Feng, J.O. Kim, J. Wu, H. Wang, G.C. Chen, F.Z. Cui, J. Mater. Sci.
Mater. Med. 9 (1998) 129e134.
425e429.
[40] S.L. Percival, P.G. Bowler, D. Russell, J. Hosp. Infect. 60 (2005) 1e7.
[16] Y.M. Moustafa, K. El-Egili, J. Non-Cryst. Solids 240 (1998) 144.
[41] P.J. Snodgrass, B.I. Vallee, F.L. Hoch, J. Biol. Chem. 235 (1960) 504e508.
[17] E.I. Kamitsos, A.P. Patsis, M.A. Karakassides, G.D. Chryssikos, J. Non-Cryst.
[42] M.K. Rayman, T.C.Y. Lo, B.D. Sanwal, J. Biol. Chem. 247 (1972) 6332e6339.
Solids 126 (1990) 52.
[43] W.J.A. Schreurs, H. Rosenberg, J. Bacteriol. 152 (1982) 7e13.
[18] A. Rulmont, R. Cahay, M.L. Duyckaerts, P. Tarte, Eur. J. Solid State Inorg. Chem.
[44] K.B. Holt, A.J. Bard, Biochemistry 44 (2005) 13214e13223.
28 (1991) 207.
[45] A.L. Semeykina, V.P. Skulachev, FEBS Lett. 269 (1990) 69e72.
[19] H. Gao, T. Tan, D. Wang, J. Control. Release. 96 (2004) 21e28.
[46] M. Hayashi, T. Miyoshi, M. Sato, T. Unemoto, Biochim. Biophys. Acta 1099
[20] A.H. Khafagy, M.A. Ewaida, A.A. Higazy, M.M.S. Ghoneim, I.Z. Hager,
(1992) 145e151.
R.R. El-Bahnasawy, J. Mater. Sci. 27 (1992) 1435.
[47] H.S. Rosenkranz, H.S. Carr, Antimicrob. Chemother. 2 (1972) 367e372.
[21] S.S. Das, B.P. Baranwal, P. Singh, V. Srivastava, Prog. Crystal Growrh Characr 45
[48] P.A. Goddard, A.T. Bull, Appl. Microbiol. Biotechnol. 31 (1989) 314.
(2002) 89e96.
[49] J.B. Chappell, G.D. Greville, Nature 174 (1954) 930e931.
[22] D. Ilieva, B. Jivov, G. Bogacev, C. Petkov, I. Penkov, Y. Dimitriev, J. Non-Cryst.
[50] P. Mitchell, Biol. Rev. 41 (1966) 445e502.
Solids 238 (2001) 195.
[51] P. Dibrov, J. Dzioba, K.K. Gosink, C. Hase, Antimicrob. Agents Chemother.
[23] D.M. Pickup, S.P. Valappil, R.M. Moss, H.L. Twyman, P. Guerry, M.E. Smith,
8 (2002) 2668e2670.
M. Wilson, J.C. Knowles, R.J. Newport, J. Mater. Sci. 44 (2009) 1858e1867.