Study on the preparation and properties of silver-doped phosphate antibacterial glasses (Part I)

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 b

Glass Research Department, National Research Centre, Dokki, Cairo, Egypt 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: Received 14 October 2010 Received in revised form 26 January 2011 Accepted 11 February 2011 Available online 19 February 2011

Silver-doped phosphate antibacterial glasses were prepared by the melting method. The antibacterial effects of some undoped and silveredoped glasses of compositions 65P2O5e10CaOe(25ex) Na2O, 70P2O5e20CaOe(10ex) Na2Oand (70ex) P2O5e30CaO, (where x ¼ 0, 0.5, 1.2 Ag2O), against Staphylococcus aureus, Pseudomonas aeruginosa and Escherichia coli micro-organisms using agar disk-diffusion assays 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 increase in inhibition zone diameter (zone of no bacterial growth) was seen with the decrease in water pH. Silver-doped glasses showed an increase in inhibition zone diameter with increasing Ag2O content. The 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.

Keywords: Antibacterial glasses Silver-doped phosphate-based glasses Glass dissolution

1. Introduction Glasses in the P2O5eCaOeNa2O system have a chemical composition similar to that of the inorganic phase of bone. These glasses consist of PO4 tetrahedra, which can be attached to a maximum of three neighboring tetrahedra forming a three dimensional network as in vitreous P2O5 [1]. Adding metal oxides to the glass leads to a depolymerization of the network, with the breaking of PeOeP linkages and the creation of non-bridging oxygens. The modifying cations can provide ionic cross-linking between the non-bridging oxygens of two phosphate chains, thus increasing the bond strength of this ionic cross-link and improving the mechanical strength and chemical durability of the glasses [2]. These phosphate-based glasses are a unique class of materials in that they are completely degradable; whereas silica-based glasses are relatively stable to hydrolysis. Furthermore, the degradation of phosphate-based glasses can be tailored to suit the end application and the rate at which they hydrolyze can vary quite considerably [3]. Various types of silver-doped inorganic antibacterial materials have been developed e.g. zeolites, calcium phosphate, silica gel, and borosilicate glass and some of them are now in commercial use.

* Corresponding author. E-mail address: [email protected] (A.A. Ali). 1293-2558/$ e see front matter Ó 2011 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.solidstatesciences.2011.02.004

Silver ions are effective against a broad range of micro-organisms including G- bacteria e.g., Pseudomonas aeruginosa, yeast e.g., Candida albicans, and Gþ bacteria e.g. Staphylococcus aureus [4,5]. Therefore, silver ions have been commercially used to take advantage of its antibacterial properties e.g. silver nitrate, colloidal silver, and certain other silver compounds are among the most generally used bactericidal agents. A large number of healthcare products now contain silver ions, principally due to its low toxicity to human cells and high antibacterial effect. Such products include silver-coated catheters, and wound dressings [6]. Phosphate-based glasses are materials of technological importance due to their superior physical properties compared to silicate glasses e.g., low melting temperatures, low glass transition and low softening temperatures, and high thermal expansion coefficients [7,8]. Thus PBGs can be prepared and processed easily at lower temperatures. In addition, phosphate-based glasses enjoy a range of compositional and structural possibilities (ultra, meta, pyro, and ortho) that facilitate tailoring chemical and physical properties of interest for specific technological applications. Controlled-release glasses (CRGs) were first developed in the 1970s primarily for use in food production industries [9]. Drake and Allen [10] found that PBGs with a suitable composition would dissolve in water with zeroorder rate constant, and by controlling the composition it was possible to produce glasses which would completely degrade in water from hours to years thus can, over a prolonged period, release

982

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

any additional constituents incorporated into them. Hence it has been possible to use pellets of CRGs containing metal ions such as copper and cobalt, as pesticides, fungicides and in animal feeds. CRGs are manufactured in a similar way to conventional soda-lime silica glasses in that the constituents are heated to temperatures above 1000  C, then cast into various forms such as solid blocks, powder, granules, tubes, fiber or wool. The incorporation of wellknown silver, copper or zinc antibacterial metal ions in several glass systems has a proven negative influence on the growth of bacteria and fungi [11,12]. Where in presence of an aqueous medium or moisture, the glass will gradually dissolve and at the same time, silver, copper, or zinc ions are released during its dissolution to provide an antibacterial effect. Generally, antibacterial glasses can be manufactured either by addition of an antibacterial agent to the glass batch prior to their manufacture or by post-treatment processes e.g. ion-exchange or surface coating. This work is an attempting to prepare and study the antibacterial effect of high dissolution silver-doped phosphate glasses. 2. Experimental 2.1. Glass preparation All batches were prepared from chemically pure grade chemicals in the powder form. P2O5 was introduced as (NH4H2PO4) (99.0% Merck), Calcium oxide (CaO) as Calcium carbonate (CaCO3) (99.5% SRL) Sodium oxide (Na2O) as sodium carbonate (Na2CO3) (99.5% s.d. fine-chem), and silver oxide (Ag2O) as silver nitrate (AgNO3) (99.9% SRL).The appropriate amounts of the starting materials of each batch equivalent to 50 g glass were accurately weighed, thoroughly mixed and then transferred to porcelain crucibles. Before melting, the batches were calcined slowly in an electric muffle furnace at a temperature in the range of 350e550  C in order to get rid of the gaseous decomposition products of the batch materials, e.g. H2O, NH3, NO2, and CO2 and to minimize the evaporation tendency of P2O5. Calcination was continued until the decomposition of the batch materials and evolution of gaseous products came to an end. All the batches were melted in disposable porcelain crucibles inside an electrically heated furnace in the range 800e1200  C. The melting time was continued for 1 h to 2 h depending upon the chemical composition. During melting, the melt was stirred manually by swirling about several times to ensure homogeneity and to get ride of gas bubbles. The melt was then cast on a preheated stainless steel plate in the form of rectangular slabs which subsequently annealed in a muffle furnace maintained at a temperature in the range 200e450  C for 20 min. The muffle furnace was then switched off and the glass samples were left overnight to cool slowly to room temperature. The visible characteristics e.g. color; transparency, and homogeneity, of all samples prepared in this work were investigated using the normal visual observations. 2.2. XRD measurements To ensure the glassy state, some selected samples were characterized with powder X-ray diffraction technique which is commonly used to verify the amorphous state of glassy materials. In the XRD spectra of glassy materials a halo is seen instead of diffraction peaks. the samples were finely ground in an agate mortar and X-ray diffraction spectra were obtained using a Bruker D8 Advance X-ray diffractometer at room temperature with Nifiltered Cu Ka radiation (l ¼ 0.15418 nm), generated at 40 kV and 40 mA. Scans were performed with a step size of 0.02 and a step time of 0.4 s over an angular range 2q from 4 to 70 .

2.3. FT-IR absorption measurements The FT-IR absorption spectra of some selected undoped and silver-doped glasses (0, 0.5, 1 and 2 mol% Ag2O) were recorded at room temperature in the frequency range 400e4000 cm1 using an infrared spectrometer (Jasco FT-IR 6100). The measurements were made by the KBr disc technique in which discs were prepared by mixing and grinding a small amount of glass powder with spectroscopic grade anhydrous KBr powder and then pressed under vacuum and pressure of 6 ton/cm2 into clear disks (1.2 cm in diameter and about 0.5 mm in thickness). All measurements were recorded with a resolution of 4 cm1. 2.4. UVeVIS absorption measurements UVeVIS absorption spectra were measured for some undoped and silver-doped glasses (0, 0.5, 1 and 2 mol% Ag2O). Polished glass 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.). 2.5. pH measurements The pH changes of the distilled water during the dissolution of some undoped and silver-doped glasses were measured at every hour and up to 6 h using IQ 140 pH-meter (IQ Inc. USA). The pH electrode was calibrated using pH calibration standards (Colourkey Buffer Solutions BDH, UK). 2.6. Antibacterial activity test The antibacterial activities of undoped and silver-doped P2O5eCaOeNa2O glasses were tested against bacterial species of American Type Culture Collection (ATCC); S. aureus (ATCC, 25923), E. coli (ATCC, 25922), and P. aeruginosa (ATCC, 27853) using the agar disk-diffusion assays. 3. Results 3.1. Glass forming region (GFR) The glass forming regions and the compositions prepared in the systems P2O5eCaOeNa2Oex Ag2O and P2O5eCaO-x Ag2O, x ¼ 0.5, 1 and 2 mol % are illustrated in Figs. 1e3. Clear circles denote homogeneous, transparent and colorless glasses as confirmed by XRD. Black circles denote samples that showed metallic silver particles precipitation. Fig. 1 and Fig. 2 showed that the compositions containing  60 mol% of P2O5 in the quaternary system P2O5eCaOeNa2Oex Ag2O, x ¼ 0.5 and 1 mol%, formed homogeneous, transparent and colorless silver-doped glasses, whereas it was not possible to obtain homogeneous glasses for the compositions containing  55 mol% of P2O5 since these compositions showed precipitations of metallic silver particles. Also it can be seen from Fig. 3 that the compositions containing  65 mol% of P2O5 in the quaternary system P2O5eCaOeNa2Oe2Ag2O formed homogeneous, transparent and colorless silver-doped glasses. Nevertheless, among compositions containing 60 mol% of P2O5, only three compositions which contain 10, 15 and 20 mol% of Na2O formed homogeneous, transparent and colorless silver-doped glasses. For other compositions containing 5, 25, 30 and 35 mol % Na2O, homogenous silver-doped glasses could not be obtained since these compositions showed precipitations of metallic silver particles. Overall, A glass forming region containing  60 mol% of P2O5 was observed in the quaternary system P2O5eCaOeNa2Oex Ag2O, x ¼ 0.5, 1 and 2 mol%.

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

983

P2O5

P2O5 0 5

0

100 A 0.5 mol% Ag2O

90 C

15 20

75 F

45

.

. ......... ..

50 55 60 65

35

50 K

40

45 L 40 M 35 N

Fig. 1. Ternary phase diagram showing the GFR and the glass forming compositions in the quaternary system P2O5eCaOeNa2Oe0.5 Ag2O.

X-ray diffraction measurements were performed on some 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 no sharp peaks thus indicating the absence of formation of any crystalline phases and ensuring the amorphous nature of these prepared samples.

3.2. FT-IR absorption spectroscopy 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 these glasses and to obtain essential information concerning the arrangement of the phosphate structural units in the phosphate glass network. As shown in Fig. 5a, b and c no new absorption bands were detected on the addition of x Ag2O, x ¼ 0.5, 1 and 2 mol% to the P2O5eCaOeNa2O base glasses. Also, it can be seen that most of the bands observed in the base glasses do not show significant shifts with the Ag2O addition. Most of bands appeared on the same P2O5

0

75 F

30

60 I 55 J

70 30 CaO 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 Na O + 0.5 Ag O 2 2 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 Mol %

5

80 E

25

65 H

40

45

2 mol % Ag 2O

85 D

20

70 G

35

90 C

15

80 E

30

95 B

10

85 D

25

100 A

5

95 B

10

...

.

70 G 65 H 60 I 55 J

50 CaO 0

5 11 10

50 K

10 15 20 25 30 35 40 45 50 Na2O + 2 Ag2O 9

8

7

6

5

4

3

2

1

Mol % Fig. 3. Ternary phase diagram showing the GFR and the glass forming compositions in the quaternary system P2O5eCaOeNa2Oe2Ag2O. B: Homogeneous glass (Transparent and Colorless). C: Metallic silver particles precipitation.

position, whereas few bands shifted to slightly higher frequencies. It can also be seen that the intensities of some bands increased with the increase in the Ag2O content. Six main absorption bands located in the regions at about 445e472 cm1, 720e750 cm1, 907e918 cm1, 1024e1050 cm1, 1100e1109 cm1 and 1298e1332 cm1 along with two shoulders at about 530 cm1 and 1170 cm1 were observed in all FT-IR spectra of P2O5eCaOeNa2O glasses, Table 2.

3.3. UVeVIS absorption spectra Fig. 6a, b and c shows the UVeVIS absorption spectra recorded in the wavelength range 200e1000 nm at room temperature for 70P2O5e20CaOe(10ex) Na2Oex Ag2O, (70ex) P2O5e30CaO-x Ag2O and 65P2O5e10CaOe(25ex) Na2O-x Ag2O glasses, x ¼ 0, 0.5, 1 and 2 mol%. From this figure, it can be seen that the undoped and silverdoped 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

100 A

10

95 B 90 C

15

1 mol % Ag2O

85 D

20

80 E

25

75 F

30

70 G

35

65 H

40

60 I 55 J

45 50

50 K

CaO 0

5 10 15 20 25 30 35 40 45 50 11 10 9 8 7 6 5 4 3 2 1

Na2O + 1 Ag2O

Mol % Fig. 2. Ternary phase diagram showing the GFR and the glass forming compositions in the quaternary system P2O5eCaOeNa2Oe1 Ag2O.

Fig. 4. The XRD patterns for some P2O5eCaOeNa2OeAg2O glasses containing 2 mol % Ag2O.

984

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

a Absorbance (a.u.)

x=2

x=1

450

x = 0.5

530

911 1017 1096 1298 1170

735

1644

x=0

400

600

800

1000

1200

1400

1600

1800

2000

wavenumber (cm-1)

Absorbance (a.u.)

b

x=2

x=1

445 528

x = 0.5 743

917 1024 110911701332

1644

x=0

400

600

800

1000

1200

1400

1600

1800

2000

Wavenumber (cm-1)

c

Absorbance (a.u.)

x=2

x=1

x = 0.5 448

530

907 1024 1300 1100 1170

720

1644

x=0 400

600

800

1000

1200

1400

1600

1800

2000

Fig. 6. UVeVIS absorption spectra (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%. The inset shows the region 200e320 nm.

3.4. pH measurements and dissolution rates

wavenumber (cm-1) Fig. 5. FT-IR absorption spectra for (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%.

shift (a shift to longer wavelength) in the position of this absorption band was observed (in all UV-spectra of Ag2O-doped glasses) with increasing Ag2O content and strong absorption peak at about 230 nm was observed for all silver-doped glasses.

The pH changes and dissolution rates measured during the dissolution of some undoped and Ag2O-doped P2O5eCaOeNa2O 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 Table 3 and Fig. 7 a fast drop in pH of distilled water (w5.5) was seen through the first hour of glass dissolution and then the pH decreased slowly with increasing time of dissolution. The glass with the highest dissolution rate, H3, shows a higher decrease in pH

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

a

b 6.0

6.0 5.5

G5

5.0

G5Ag1

G5Ag0.5 G5Ag2

5.5

G7

5.0

G7Ag1

G7Ag0.5 G7Ag2

4.5

pH

4.5

pH

985

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

c

2

3

4

5

6

7

Dissolution Time ( Hours)

Dissolution Time ( Hours) 6.0 5.5

H3

5.0

H3Ag1

H3Ag0.5 H3Ag2

pH

4.5 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. That means the pH is dependant on the glass dissolution rate. 3.5. Antibacterial activity The antibacterial effects of undoped and2 mol % Ag2O-doped P2O5eCaOeNa2O and P2O5eCaO glasses were tested in vitro against 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 antibacterial activity of the glass was confirmed by the presence of an inhibitory zone (i.e. zone of no bacterial growth) around each tested glass disk. The measured inhibition zone diameters (minus the 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) demonstrated different antibacterial effects against the tested micro-organisms as indicated by the clear zone around each glass disk. Figs. 8 and 9 also show that the glass antibacterial effect depends on the glass composition, Ag2O content and type of the tested micro-organism. S. aureus was found to be the most susceptible micro-organism to the tested antibacterial glasses. The degree of susceptibility of the tested micro-organisms to the tested antibacterial glasses was in this order S. aureus > P. aeruginosa > E. coli. Figs. 10e12 show the variation of the inhibition zone diameter with the Ag2O content in the all studied glasses. As displayed in Figs. 10e12, a gradual increase in the inhibition zone diameter (the antibacterial activity is proportional to the size of inhibition zone)

was seen with increasing Ag2O content in the glass. The biggest zone of inhibition among silver-doped glasses was observed for the highest silver releasing glass, namely H3Ag2 against S. aureus microorganism. Figs. 13e15 show the variation of the inhibition zone diameter with dissolution rate for silver free glasses. The biggest zone of inhibition among silver free glasses was observed for the glass with the highest dissolution rate, namely H3 against S. aureus micro-organism as shown in Figs. 13e15.

4. Discussion 4.1. Glass forming region A glass forming region which contains  60 mol% of P2O5 was observed in the quaternary system P2O5eCaOeNa2Oex Ag2O, x ¼ 0.5, 1 and 2 mol%, whereas compositions containing  55 mol% of P2O5 showed metallic silver particles precipitation. The P2O5eCaOeNa2OeAg2O glasses were prepared under normal melting conditions without any special precautions. Since silver is a noble element, its oxide is easily reduced at high temperatures. The decomposition temperature of Ag2O under ambient conditions is calculated from its thermodynamic data to be 421 K. It is well known that silver may exist in glass in one or more than one of its common states (Ag0, Agþ or Ag2þ). The solubility of Ag2O in glass melts is an important factor for effective production of Ag2O containing glasses as silver is known to have different degrees

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 solubility. The Ag2O added to the glass batch over its solubility is reduced to metallic silver and it is of no use to increase Ag2O content in the glass beyond its solubility limit. It is known that the solubility of Ag2O in glass melts is limited by thermodynamic factors and is highly dependent on the glass composition, temperature during melting, and the oxygen potential [13,14]. Therefore the glass composition should be selected carefully because it is difficult to control the oxygen potential during the melting process. 4.2. Structure of studied glasses It is well known that the phosphate network structure consists of a series of PO4 tetrahedral units connected by three bridging oxygens. The network connectivity can be described in terms of a Qn distribution as shown, where n is the number of bridging oxygens. Q0 corresponds to isolated tetrahedra (orthophosphate groups), Q1 to end groups (pyrophosphate), Q2 to middle groups (metaphosphate) and Q3 to branching groups (ultraphosphate). When a modifier oxide is added, disruption of the main network occurs with the modifier oxide cations occupying interstitial

positions. It is accepted that the structure of ultraphosphates, metaphosphates, and polyphosphates are dominated by (Q3 & Q2), (Q2), and (Q2 & Q1) units, respectively. The FT-IR spectra of some undoped and Ag2O-doped P2O5eCaOeNa2O glasses shown in Fig. 5 are typical of phosphatebased glasses showing the characteristic absorption bands of PO4 groups. The FT-IR spectra of these glasses appear to be dominated by metaphosphate (Q2) and ultraphosphate structural units (Q3). Three main spectral regions can be distinguished in the FT-IR spectra of the glasses as follows: 400e800 cm1, 800e1200 cm1 and 1200e1400 cm1. In the spectral region ( 400e800 cm1), the band at w 450e470 cm1 and the shoulder at w 530 cm1 may be attributed to harmonics of bending vibrations of OePeO and O] PeO linkages. Another band in the frequency region (400e800 cm1) is the absorption band at about 720e750 cm1. This band may be attributed to the symmetric stretching vibrations of the PeOeP linkages, ys (PeOeP) modes [15]. This absorption band at about 750 cm1 shifts towards lower frequency with increasing the concentration of Ag2O. The variation of the frequency of PeOeP bonds with increasing Ag2O content is consistent with breakage of

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 modifier. The FT-IR absorptions in the spectral region 800e1200 cm1 was found to be sensitive for the different metaphosphate groups in the form of chain-, ring-, terminal groups [16,17]. This area in the FT-IR spectra of all the studied glasses is characterized by the presence of four absorption bands at about 910e920, 1017e1050, w1100 and 1160e1170 cm1. The absorption band at w 910 cm1 is assigned to the asymmetric stretching vibrations of PeOeP groups linked with linear metaphosphate chain [15]. Phosphate-based glasses with a metaphosphate structure can have a chain and/or ring structure. The occurrence of the yas (PeOeP) at around 900 cm1 in all the samples studied is an indication of a chain structure, as the analogous vibration occurs at around 1000 cm1 in ring structures [18]. The shift of PeOeP asymmetric stretching vibration around 900 cm1 to higher wave numbers [19], always indicates increase of covalency proportion of the PeOeP bond and strengthening of glass structure with improved chemical durability. The weak absorption band at 1017e1050 cm1 is attributed to a normal vibration mode (y3 symmetric stretching) of the PO43- groups [20,21]. The position of this band is constant and does not change by the addition of varying

concentrations of the Ag2O. The band at w 1100 cm1 can be associated with an overlap of several modes like stretching of the PO3 terminal and PO2 middle groups. In addition, the y3 (F2) mode of the ortho anion can also contribute to the absorption at 1100 cm1 [15,22]. The band at w 1100 cm1 was also assigned to asymmetric stretching mode of chain-terminating Q1 groups, yas (PO3)2-. The shoulder at about 1160e1170 cm1 may be attributed to symmetric stretching mode of the two non-bridging oxygen atoms in the Q2 tetrahedral sites, ys (PO2) mode in metaphosphate groups [23]. The spectral region 1200e1400 cm1 showed a strong broad absorption band in the range 1298e1330 cm1. The appearance of this band (at w1300 cm1) could be due to the presence of a fraction of Q3 tetrahedral units which are the most characteristic feature of an ultraphosphate glass. This band appeared in all the FTIR spectra of the studied phosphate-based glasses and it is attributed to the asymmetric stretching of the doubly bonded oxygen, yas (P]O) modes, which are only present in the glasses with a P2O5 content > 50 mol% [15,16]. Osaka et al. [24] have shown that PO4 units have two bridging oxygen bonds along with two non-bridging bonds such as P]O and PeO, which are in resonance with each

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). Inhibition zone Glass Density D.R code No. (g cm2 h1) diameter (mm) *104 CaO Na2O S. P. E. coli aureus aeruginosa

Glass Composition P2O5 70 70 70 70 70 69.5 69 68 65 65 65 65

20 20 20 20 30 30 30 30 10 10 10 10

10 9.5 9 8 0 0 0 0 25 24.5 24 23

G5 G5Ag0.5 G5Ag1 G5Ag2 G7 G7Ag0.5 G7Ag1 G7Ag2 H3 H3Ag0.5 H3Ag1 H3Ag2

2.3937 2.4186 2.4415 2.4827 2.4098 2.4302 2.4523 2.4910 2.4229 2.4440 2.4647 2.5066

3.91 3.52 3.12 2.55 3.24 2.95 2.67 2.19 6.46 5.85 5.32 4.36

27 31 34 38 25 29 32 36 30 34 37 41

25 28 31 35 22 28 30 33 28 31 35 38

22 25 28 32 19 23 26 29 25 29 33 36

other. Therefore, the FT-IR spectra, as in the present study, are split into two bands with a higher energy and strong double bond character (1298e1330 cm1) and a lower energy band (w1100 cm1). This means that the P]O double bond is more strongly localized in the central position of phosphate groups. This assumption is confirmed by the larger relative intensity of the 1298e1330 cm1 envelope compared with that of the band at w 1100 cm1. The weak absorption band at around 1615e1644 cm1 observed in all spectra of phosphate glasses prepared in this work is attributed to the bending vibration modes of OeH groups, d (H2O) modes [21]. This band demonstrates the presence of small amounts of water absorbed from air moisture (due to the hygroscopic character of the phosphate glasses) during the preparation of the KBr pellets for infrared measurements [16]. The band positions of various phosphate structural groups observed for the studied glass samples were found to be well within the ranges reported in the literature for different phosphate-based glasses. The positions and assignments [15, 19, and 23] of the IR absorption bands observed in the FT-IR spectra of the studied glasses are summarized in Table 2. As the P2O5 content increases the band around 1320 cm1 become broader and this is typically for ultraphosphate glasses. Also it can be seen from the FT-IR spectra of the base glasses that the absorption bands at about w 910 cm1 and 720 cm1 assigned

Fig. 10. Variation of IZD with Ag2O content for 70P2O5e20CaOe(10ex ) Na2Oex Ag2O glasses, x ¼ 0, 0.5, 1 and 2 mol%.

Fig. 11. Variation of IZD with Ag2O content for (70ex) P2O5e30CaOex Ag2O glasses, x ¼ 0, 0.5, 1 and 2 mol%.

to the yas (PeOeP) and ys (PeOeP) modes shift to higher frequencies as Na2O is replaced by CaO showing an intensified PeOeP bonds in the glass structure and showing that the covalency proportion in the bond of the metal ions with the non-bridging oxygen on the [PO4] tetrahedron units increased probably because of the increased polarizability of the calcium ions which has a smaller electrovalency in the Ca..O bond than in the Na..O bond. This explains the improvement in chemical durability with increasing CaO content [19]. The band shift can be explained by the increase in covalent character of PeOeP bonds, indicating that the PeOeP bonds are strengthened as the Na2O is substituted by CaO. This explains the improvement in the chemical durability of these glasses as the CaO content increases. Overall, the FT-IR spectra of the base P2O5eCaOeNa2O showed that the shift of yas (PeOeP) and ys (PeOeP) to higher frequencies as CaO replaces Na2O is witnessed with improved chemical durability of P2O5eCaOeNa2O glasses and this was explained in terms of: (i) Strengthened Ca..O bond formed by the non-bridging oxygen with the Ca ion introduced on the [PO4] tetrahedron; (ii) Strengthened PeOeP bond formed by the bridging oxygen between two [PO4] tetrahedrons.

Fig. 12. Variation of IZD with Ag2O content for 65P2O5e10CaOe(25ex) Na2Oex Ag2O glasses, x ¼ 0, 0.5, 1 and 2 mol%.

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 micro-organism.

Fig. 15. Variation of IZD with D.R for some P2O5eCaOeNa2O glasses in case of E. coli micro-organism.

According to the dissolution mechanism of the phosphate glasses [25], these two factors determine the dissolution rate of the phosphate glasses by both limiting the velocity of hydration reaction and the network breakage reaction, respectively.

A strong absorption peak at about 230 nm was observed for all silver-doped glasses as shown in Fig. 6a which is presumably due to Agþ ions. Ahmed and Abdallah [29] observed three absorption bands at 305 nm, 350 nm and 420 nm in the UVeVIS spectra of silver-containing soda-lime silica glass prepared by the ionexchange process. They assigned these bands to silver ions Agþ, elemental silver Ag0 and silver crystallites (Ag0)n (n is the number of silver atoms forming the crystallite) respectively. Paje et al. [30] observed a broad absorption band centered at w260 nm for a silver-doped silicate glass prepared by the melt-quenching technique. However, they also observed the same peak in the host glass, so they did not attribute it to silver. Borsella et al. [31] performed UVeVIS spectroscopic investigations on a range of silver concentrations in soda-lime glass prepared by the ion-exchange process. In contrast to Paje et al. [30] they observed an absorption band at about 268 nm for silver ions in the soda-lime glass. The absorption was ascribed to transitions involving the promotion of an electron from the 4d10 ground state to the 4d9 5s1 state. Jimenez et al. [32] observed an absorption band centered at about 275 nm for silver-doped aluminophosphate glass and they ascribed it to silver ions. It was reported [28] that besides the pronounced silver plasmon resonances that appear between 400 and 500 nm, the electronic transitions involving the Agþ ion give rise to absorption

4.3. UVeVIS spectra It is well known [26,27] that the optical properties of silver atoms are entirely determined in the visible region by their free electrons (Ag: [Kr] 4d10 5s1), while that of silver ions are determined in the ultraviolet region (Agþ: [Kr] 4d10 5s0). Thus absorptions in the UVeVIS spectral regions can help in the elucidation of silver states within a glass network. In the present work, the UVeVIS absorption spectra of some undoped and silver-doped P2O5eCaOeNa2O glasses revealed no absorption peaks in the visible region, whereas two absorption peaks were observed in the ultraviolet region. A peak at bout 210 nm was observed for all undoped glasses as shown in Fig. 6a, b and c except for the binary glass, G7, where this peak was observed at about 240 nm and it was weak and broad as shown in Fig. 6b. This band (either at about 210 nm or at about 240 nm) is attributed to electronic absorption given by the phosphate glass network (the threshold absorption of the host glass matrix) [28].

Table 2 Assignment of IR absorption bands of studied glasses. Band frequency (cm1)

Assignment

450e470 530

Bending vibrations of OePeO bonds, d (PO2) modes. Fundamental bending vibrations of O]PeO bonds, d (O]PeO) modes. Bending mode of PO4 units. Symmetric stretching vibrations of PeOeP linkages, ys (PeOeP). Asymmetric stretching vibrations of PeOeP linkages, yas (PeOeP) linked with linear metaphosphate chain. Vibration mode (y3 symmetric stretching) of the PO43groups. Asymmetric stretching mode of chain-terminating Q1 groups, yas (PO3)2-. Symmetric stretching mode of the tow non-bridging oxygen atoms in the Q2 tetrahedral sites, ys (PO2) mode in metaphosphate groups Q2. Asymmetric stretching of doubly bonded oxygen, yas (P]O) modes. Bending vibration modes of OeH groups in HeOeH, d (H2O) modes.

660 720e750 900e920 1020e1050 1100 1170

1300e1330

Fig. 14. Variation of IZD with D.R for some P2O5eCaOeNa2O glasses in case of P. aeruginosa micro-organism.

1610e1640

990

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

bands located between 200 and 230 nm, whereas the electronic transitions of metallic Ago atoms appear in the 250e330 nm spectral range. Accordingly, it is reasonable to ascribe the absorption band at around 235 nm which appeared in the UVeVIS spectra of the Ag2Odoped P2O5eCaOeNa2O glasses to electronic transitions involving Agþ ions, [4d10 / 4d9 5 s1]. Although these transitions are parityforbidden, they are partially allowed in a solid due to vibrational coupling. The observed red shift of the absorption edge (the shift from 230 to 245 nm) associated with increasing Ag2O content, appears to be related to several effects. The progressive introduction of silver oxide is responsible for the red shift of the absorption edge. Regarding silver and its role as a network modifying oxide implies that the red shift (caused by Ag2O addition) is related to the incorporation of silver ions in glass structure. Also this shift may be arise from an increase in the concentration of non-bridging oxygens with increasing silver content or may be due to the decrease of the smallest AgeO distance existing in the family of silver sites. 4.4. Antibacterial activity An antibacterial activity of a substance is an indication of its ability to kill bacteria or inhibit their growth. In the present study, the tested silver free and silver-doped phosphate-based glasses showed different antibacterial effects against S. aureus, P. aeruginosa and E. coli micro-organisms as shown in Figs. 8 and 9. The antibacterial effect is indicated by the clear zone (zone of no bacterial growth) around each glass disk. As shown in these figures, the silver-doped glasses demonstrated more potent antibacterial effects than the silver free glasses. According to the results of the agar disk-diffusion assays, all glass compositions examined were able to inhibit the growth of Gþ and G- bacteria. The antibacterial effect as shown in Figs. 8 and 9 varies according to the glass composition, Ag2O contents and the type of the micro-organism. The degree of susceptibility of the tested micro-organisms to the tested antibacterial glasses was in this order S. aureus > P. aeruginosa > E. coli. The explanations of the antibacterial effects of silver free and silver-doped glasses are given in the following sections: 4.4.1. Antibacterial activity of silver free glasses In the present work, some P2O5eCaOeNa2O glasses were used as control samples in the antibacterial activity test. It was thought that these silver free glasses would not show any antibacterial effects against the tested micro-organisms. Interestingly, these silver free glasses namely, G5, G7, and H3 produced inhibitory zones of different sizes (Fig. 8). This antibacterial effect of these silver free glasses can be explained in terms of the glass composition, the glass dissolution rate, and the pH changes of the medium. Valappil et al. [33] observed a zone of inhibition for S. aureus, MRSA and C.difficile in testing the antibacterial activity of gallium free 45P2O5e16CaOe39Na2O glass. They attributed the antibacterial effect of this gallium free glass to the change in pH during the glass degradation. Pickup et al. [23] investigated the antibacterial activities of a Ga-doped solegel phosphate-based glasses of composition (CaO)0.30 (Na2O)0.20x (Ga2O3)x (P2O5)0.50 where x ¼ 0 and 0.03 mol %. They observed a small zone of inhibition (7 mm) for the gallium free glass and they attributed it to either a change in pH as the glass dissolves or by reduced water activity as ions leach out. The glass compositions investigated contain 65 and 70 mol% of P2O5 and the remainder is made up of CaO and Na2O. Accordingly, due to their high P2O5 contents, these glasses have acidic composition and this means that the dissolution of such glasses changes

the pH and produces acidic media (since more acidic phosphate ions are released). The pH value of the medium (in which the glass dissolves) and inhibition nation zone diameter were found to be related to the glass dissolution rate (Table 1 and Figs. 13e15). The dissolution rate of P2O5eCaOeNa2O glasses depends on the contents of P2O5, CaO and Na2O in the glass. From the results of dissolution of such glasses, it was found that the glass dissolution rate is directly proportional to either P2O5 or Na2O contents and inversely proportional to CaO content. The glass dissolution study showed that the dissolution rate of these glasses as shown in Table 1 and Figs. 13e15 are in this order H3 > G5 > G7 i.e. H3 has the fastest dissolution rate, whereas G7 has the slowest dissolution rate in this glass series. As shown in Tables 1 and 3 a decrease in the pH was seen with the increases in the dissolution rate since more acidic phosphate ions are released. It is well known that the acidity or alkalinity of the medium affects the growth of bacteria. The pH affects the rate of enzyme action and plays a role in determining the ability of bacteria to grow or survive in particular environments. Most bacteria survive near 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 this study is close to neutral. Thus, the decrease in pH during the dissolution of P2O5eCaOeNa2O glasses could explain the bacterial growth inhibition produced by such glasses. The more the phosphate ions released the lower is the pH and the greater the antibacterial effect. Thus the low pH produced by glass dissolution was certainly a critical factor for glass antibacterial effect. Overall, the antibacterial effect of silver free glasses was influenced by glass composition, glass dissolution rate and the dissolution conditions in the glass surroundings. This might explain at least some of the differences in the antibacterial action of glass with varying chemical compositions. The antibacterial activity of silver free glasses was closely related to the dissolution rate of the glasses because high dissolution rates cause a decrease of pH or an increase of the ion concentration of the media, and this resulted in antibacterial activity. Therefore, the results of silver free glass dissolution, their pH changes during dissolution and their antibacterial effects correlate well with each other. The antibacterial effect of the silver free glasses may be attributed to other factors [34e36] e.g. the ionic strength of the medium where high concentrations of calcium, sodium and phosphates ions likely to be released from the glass during dissolution could cause perturbations of the membrane potential of bacteria and such as osmotic effects caused by the nonphysiological concentration of ions such as sodium, calcium, and phosphate ions dissolved from the glass and lead to change the osmotic pressure in the vicinity of the glass. The exact mechanism of the antibacterial action of these silver free glasses is unknown. Therefore, we conclude that the Table 3 pH values of studied glasses in different times. Glass Code No.

pH 1h

2h

3h

4h

5h

6h

G5 G5Ag0.5 G5Ag1 G5Ag2 G7 G7Ag0.5 G7Ag1 G7Ag2 H3 H3Ag0.5 H3Ag1 H3Ag2

2.98 3.05 3.14 3.30 3.07 3.12 3.25 3.41 2.84 2.91 3.01 3.20

2.87 2.93 3.06 3.21 2.95 3.05 3.13 3.32 2.75 2.82 2.93 3.11

2.81 2.87 2.98 3.15 2.88 2.96 3.03 3.23 2.68 2.75 2.85 3.01

2.72 2.79 2.90 3.04 2.80 2.89 2.95 3.13 2.60 2.67 2.77 2.94

2.65 2.71 2.81 2.95 2.73 2.80 2.87 3.01 2.52 2.58 2.68 2.85

2.54 2.63 2.74 2.85 2.62 2.69 2.77 2.92 2.41 2.50 2.59 2.76

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

antibacterial effect observed with these glasses can be explained by the dramatic changes in the physicochemical characteristics of the culture medium (pH, ionic strength, and osmotic pressure) which occur as a consequence of the glasses dissolution. Thus the antibacterial action of glass is influenced by its chemical composition and the dissolution conditions in its surroundings. 4.4.2. Antibacterial activity of silver-doped glasses In the present work, the addition of Ag2O to P2O5eCaOeNa2O or P2O5eCaO glasses has been found to potentate its antibacterial activity. Silver-doped glasses showed increased antibacterial activities (depending upon the Ag2O content) more than silver free glasses as shown in Figs. 8 and 9 and as indicated by the measured inhibition zone diameters Table 1 and Figs. 10e12. This increase is attributed to the release of the Agþ ions which are well known as antibacterial metal ions [37]. An increase in the antibacterial activity as represented by the increase in the inhibition zone diameter was seen with increasing the Ag2O content in the glass as displayed in Figs. 10e12. This was in a good agreement with the results of Agþ ions release seen in water where an increase in concentrations of silver ions released from glass into water was seen with the increase in Ag2O content. Generally, in an aqueous medium or in presence of moisture, the 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 for antibacterial action of silver-doped glasses is bacterial growth inhibition by the silver ions released from the glass. Bellantone et al. [38] investigated the antibacterial effects of Ag2O-doped bioactive glasses on S. aureus, E. coli, and P. aeruginosa. The antibacterial action of the silver-doped bioactive glass was attributed to the leaching out of Agþ ions from the glass matrix. Kim et al. [39] investigated the antimicrobial effects of various ceramics against E. coli using viable count and growth rate studies. They concluded that Agþ ions interfered with the metabolism of the microorganism, thus inhibiting its growth. The exact mechanism of antibacterial action of silver ions is still unknown. Antibacterial mechanisms of silver ions might differ according to the species of bacteria. Generally, most antibacterial agents exert their antibacterial action by four principal modes of action. These modes include: inhibition of bacterial cell wall synthesis, inhibition of protein synthesis, inhibition of synthesis of bacterial RNA and DNA or inhibition of a metabolic pathway. In bacteria, silver ions are known to react with bacterial nucleophilic amino acid residues in proteins, and attach to sulphydryl, amino, imidazole, phosphate, and carboxyl groups of membranes or enzymes, resulting in protein denaturation [40]. Silver is also known to inhibit a number of oxidative enzymes such as yeast alcohol dehydrogenase [41] the uptake of succinate by membrane vesicles [42] and the respiratory chain of E. coli, as well as causing metabolite efflux [43] and interfering with DNA replication. Holt and Bard [44] examined the interaction of silver ions with the respiratory chain of E. coli. They found that an addition of 10 mM AgNO3 to suspended or immobilized E. coli resulted in stimulated respiration before death, signifying uncoupling of respiratory control from ATP synthesis. This was a symptom of the interaction of Agþ with enzymes of the respiratory chain. Feng et al. [4] studied the antibacterial effect of silver ions on E. coli and S. aureus and suggested that the antibacterial mechanism was due to DNA not being able to replicate, and proteins becoming inactivated after contact with Agþ ions. One of the primary targets of Agþ ions, specifically at low concentrations, appears to be the Naþ-translocating NADH: ubiquinone oxidoreductase system [45,46]. Silver has also been

991

shown to be associated with the cell wall [47], cytoplasm and the cell envelope [48]. Chappell and Greville [49] acknowledged that low levels of Agþ ions collapsed the proton motive force on the membrane of bacteria, and this was reinforced by Mitchell’s work [50]. Dibrov et al. [51] showed that low concentrations of Agþ ions induced a massive proton leakage through the bacterial membrane, resulting in complete de-energization and, ultimately, cell death. Overall, there is consensus that surface binding and damage to membrane function are the most important mechanisms for the killing of bacteria by Agþ ions. The results of antibacterial activity showed that the silver free and silver-doped glasses exhibited different antibacterial effects against the tested bacteria and the sensitivity of G and Gþ bacteria to the antibacterial glasses was different. A remarkable difference was seen between G- bacterium (E. coli) and Gþ bacterium (S. aureus), while a small difference was found between the two G- bacteria. It is known that the cell wall of G- bacteria is composed of high proportion of phospholipids, lipopolysaccharides and proteins (the cell wall of G- bacterium is chemically more complex than that of Gþ bacteria), whereas peptidoglycan is the major component of the cell wall of Gþ bacteria. This fact would possibly contribute to the difference of antibacterial effects between G and Gþ bacteria.

5. Conclusion 1- A glass forming region containing  60 mol% of P2O5 was observed in the quaternary P2O5eCaOeNa2Oex Ag2O system, x ¼ 0.5, 1 and 2 mol%. 2- Generally, all the prepared silver free and silver-doped glasses were (as observed visually) homogeneous, free from solid inclusions, transparent and colorless. For 2 mol% silver-doped P2O5eCaOeNa2O glasses, the XRD patterns indicated the absence of formation of any crystalline phases and thus XRD ensured the amorphous nature of these glasses. 3- Measurements of pH changes during dissolution of silver-free and silver-doped glasses in water revealed a decrease of water pH with increasing time of glass dissolution. It was found that the magnitude of the pH drop increases with the increase in glass dissolution rate. 4- Silver-doped glasses showed less pH drop than silver-free glasses. In agar disk-diffusion assays, all the tested silver-free and silver-doped glasses demonstrated different antibacterial effects (depending on the glass composition and the type of the tested micro-organism) against S. aureus, P. aeruginosa and E. coli micro-organisms as indicated by the clear zone (zone of no bacterial growth) around each tested glass disk. 5- For silver-free glasses an increase in bacterial growth inhibition zone diameter was observed with the increase in the glass dissolution rate and with the decrease in pH, whereas for silverdoped glasses an increase in bacterial growth inhibition zone diameter was observed with increasing Ag2O content. 6- S. aureus as a Gþ bacterium was found to be the most susceptible micro-organism to the tested antibacterial glasses. The degree of susceptibility of the tested micro-organisms to the tested antibacterial glasses was found in this order S. aureus > P. aeruginosa > E. coli. 7- There is a good agreement between the antibacterial activities of silver-free glasses, their dissolution rates and their pH changes during their dissolution. Also there is a good agreement between the antibacterial activities of silver-doped glasses and the concentrations of silver ions released into water during their dissolution.

992

A.A. Ahmed et al. / Solid State Sciences 13 (2011) 981e992

References [1] J. Clement, J.M. Manero, J.A. Planell, G. Avila, S. Martinez, J. Mater. Sci. Mater. Med. 10 (1999) 729. [2] I. Ahmed, M.P. Lewis, I. Olsen, J.C. Knowles, Biomaterials 25 (3) (2004) 491. [3] K. Franks, I. Abrahams, G. Georgiou, J.C. Knowles, Biomaterials 22 (2001) 497. [4] Q.L. Feng, J. Wu, G.Q. Chen, F.Z. Cui, T.N. Kim, J.O. Kim, J. Biomed. Mater. Res. 52 (2000) 662e668. [5] M. Kawashita, S. Tsuneyama, F. Miyaji, T. Kokubo, H. Kozuka, K. Yamamoto, Biomaterials 21 (2000) 393e398. [6] A.P. Adams, E.M. Santschi, M.A. Mellencamp, Vet. Surg. 28 (1999) 219e225. [7] Y. He, D.E. Day, Glass Technol. 33 (1992) 214. [8] A.E. Marino, S.R. Arrasmith, L.L. Gregg, S.D. Jacobs, G. Chen, Y. Duc, J. Non-Cryst Solids 289 (2001) 37. [9] C.F. Drake, Pest Sci. 9 (1978) 441. [10] C.F. Drake, W.M. Allen, Biochem. Soc. Trans. 13 (1985) 520. [11] A.M. Mulligan, M. Wilson, J.C. Knowles, J. Biomed. Mater. Res. 67A (2003) 401e412. [12] I. Ahmed, D. Ready, M. Wilson, J.C. Knowles, J. Biomed. Mater. Res. 79A (2006) 618e626. [13] D.A. McKeown, H. Gan, I.L. Pegg, J. Non-Cryst. Solids 351 (2005) 3826e3833. [14] M. Bosca, L. Pop, G. Borodi, P. Pascuta, E. Culea, J. Alloys Compounds 479 (2009) 579e582. [15] R.C. Lucacel, A.O. Hulpus, V. Simon, I. Ardelean, J. Non-Cryst. Solids 355 (2009) 425e429. [16] Y.M. Moustafa, K. El-Egili, J. Non-Cryst. Solids 240 (1998) 144. [17] E.I. Kamitsos, A.P. Patsis, M.A. Karakassides, G.D. Chryssikos, J. Non-Cryst. Solids 126 (1990) 52. [18] A. Rulmont, R. Cahay, M.L. Duyckaerts, P. Tarte, Eur. J. Solid State Inorg. Chem. 28 (1991) 207. [19] H. Gao, T. Tan, D. Wang, J. Control. Release. 96 (2004) 21e28. [20] A.H. Khafagy, M.A. Ewaida, A.A. Higazy, M.M.S. Ghoneim, I.Z. Hager, R.R. El-Bahnasawy, J. Mater. Sci. 27 (1992) 1435. [21] S.S. Das, B.P. Baranwal, P. Singh, V. Srivastava, Prog. Crystal Growrh Characr 45 (2002) 89e96. [22] D. Ilieva, B. Jivov, G. Bogacev, C. Petkov, I. Penkov, Y. Dimitriev, J. Non-Cryst. Solids 238 (2001) 195. [23] D.M. Pickup, S.P. Valappil, R.M. Moss, H.L. Twyman, P. Guerry, M.E. Smith, M. Wilson, J.C. Knowles, R.J. Newport, J. Mater. Sci. 44 (2009) 1858e1867.

[24] [25] [26] [27] [28] [29] [30] [31] [32] [33]

[34] [35]

[36]

[37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51]

A. Osaka, K. Takahashi, M. Ikeda, J. Mater. Sci. Lett. 3 (1984) 36. B. C Bunker, G.W. Arnold, J.A. Wilder, J. Non-Cryst. Solids 64 (1984) 291. P. Chakraborty, J. Mater. Sci. 33 (1998) 2235. R.H. Doremus, J. Chem. Phys. 42 (1965) 414. J. Lu, J.J. Bravo, A. Takahashi, M. Haruta, S.T. Oyama, J. Catal. 232 (2005) 85. A.A. Ahmed, E.W. Abd Allah, J. Am. Ceram. Soc. 78 (10) (1995) 2777e2784. S.E. Paje, J. Llopis, M.A. Villegas, J.M. Fernandez, Appl. Phys. A 63 (1996) 431. E. Borsella, G. Battaglin, M.A. García, F. Gonella, P. Mazzoldi, R. Polloni, A. Quaranta, Appl. Phys. A 71 (2000) 125. J.A. Jimenez, S. Lysenko, G. Zhang, H. Liu, J. Electronic Mater. 36 (2007). S.P. Valappil, D. Ready, E.A. Abou Neel, D.M. Pickup, W. Chrzanowski, L.A. O’Dell, R.J. Newport, M.E. Smith, M. Wilson, J.C. Knowles, Adv. Funct. Mater. 18 (2008) 732e741. P. Stoor, E. Söderling, J.I. Salonen, Acta Odontol Scand. 56 (1998) 161e165. E. Munukka, O. Lepparanta, M. Korkeamaki, M. Vaahtio, T. Peltola, D. Zhang, L. Hupa, H. Ylanen, J.I. Salonen, M.K. Viljanen, E. Eerola, J. Mater. Sci. Mater. Med. 19 (2008) 27e32. O. Lepparanta, M. Vaahtio, T. Peltola, D. Zhang, L. Hupa, M. Hupa, H. Ylänen, J.I. Salonen, M.I.K. Viljanen, E. Eerola, J. Mater. Sci. Mater. Med. 19 (2008) 547e551. A.B.G. Lansdown, J. Wound Care 11 (4) (2002) 125. M. Bellantone, H.D. Williams, L.L. Hench, Antimicrobial. Agents. Chemother. 46 (2002) 1940e1945. 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. S.L. Percival, P.G. Bowler, D. Russell, J. Hosp. Infect. 60 (2005) 1e7. P.J. Snodgrass, B.I. Vallee, F.L. Hoch, J. Biol. Chem. 235 (1960) 504e508. M.K. Rayman, T.C.Y. Lo, B.D. Sanwal, J. Biol. Chem. 247 (1972) 6332e6339. W.J.A. Schreurs, H. Rosenberg, J. Bacteriol. 152 (1982) 7e13. K.B. Holt, A.J. Bard, Biochemistry 44 (2005) 13214e13223. A.L. Semeykina, V.P. Skulachev, FEBS Lett. 269 (1990) 69e72. M. Hayashi, T. Miyoshi, M. Sato, T. Unemoto, Biochim. Biophys. Acta 1099 (1992) 145e151. H.S. Rosenkranz, H.S. Carr, Antimicrob. Chemother. 2 (1972) 367e372. P.A. Goddard, A.T. Bull, Appl. Microbiol. Biotechnol. 31 (1989) 314. J.B. Chappell, G.D. Greville, Nature 174 (1954) 930e931. P. Mitchell, Biol. Rev. 41 (1966) 445e502. P. Dibrov, J. Dzioba, K.K. Gosink, C. Hase, Antimicrob. Agents Chemother. 8 (2002) 2668e2670.