Optical processes in SiO 2 Sol-Gel Glass Colored with organic dyes

Inorganic Materials, Vol 36, No. 10, 2000, pp. 1060-1069. From Neorganicheskie Material); Vol. 36, No. I0, 2000, pp. 1258-1266. Original English Text Copyright 9 2000 by P~rez-Bueno, Dfaz-FIorez, P~rez-Robles, Espinoza-Beltrdn, Manzano-Ramirez, Ramlrez-Bon, Gonzdlez-Herndndez, Vorobiev.

Optical Processes in SiO2 Sol-Gel Glass Colored with Organic Dyes I J. J. P6rez-Bueno*, L. L. Diaz-Florez**, J. F. P6rez-Robles***, F. J. Espinoza-Beltrfin***, A. Manzano-Ramlrez***, R. Ramffez-Bon***, J. Gonzfilez-Hernfindez***, and Y. V. Vorobiev*** * Divisidn de Estudios de Postgrado, Facultad de lngenierfa, UniversidadAutdnoma de Quer~taro, Apdo. Postal 1-1010, 76010, Querdtaro, Qro., M~xico ** lnstituto Tecnolrgico de Saltillo, V. Carranza 2400 Saltillo, Coahuila, M(xico *** Centro de lnvestigacirn y de Estudios Avanzados del IPN, Unidad Quer~taro, Apdo. Postal 1-1010, 76010, Quer(taro, Qro., M(xico Received October 26, 1999

Abstract---Colored SiO 2 coatings were prepared using the sol-gel process. The color was obtained by adding organic dyes to the starting solutions. The dyes incorporated were the Brilliant Blue, Brilliant Black, Fast Green, Yellow 5, Tartrazine, and Erythrozine. It is observed that mechanical treatment of the starting solution using ball milling reduces the size of aggregated dye particles in the coatings. The material obtained reveals an efficient photoluminescence in the visible and infrared regions of the spectrum. Investigations of the optical absorption, luminescence excitation, and emission spectra show that each of these systems is characterized by a well-defined set of discrete electronic energy levels. A relation is found between the level separation and the structure of the colorant's molecule and also with the doping level. Besides, the size of the molecular aggregates greatly influences the efficiency of light absorption and emission. It is shown that a simple quantum-mechanical description of the system, treating the organic molecule as a two-dimensional potential well, accounts for the observed optical transitions. The results from this simple approach are compared with those obtained using the modified FEMO and LCAO approaches. A reasonable agreement of theory with experiment was obtained. From this work it is concluded that, by using the sol-gel technique, it is possible to produce systems in which nanometer-scale potential wells are embedded in a SiO 2 matrix. The discrete energy levels of the wells correspond to the molecular electronic transitions active in the visible region. INTRODUCTION Over the last two decades, growing interest has been centered on the investigation of sol-gel glasses colored with organic and inorganic dyes [1-15]. One of the main applications of these systems may be in new tunable dye lasers. It is now well established that sol-gel glasses enhance the life and stability of the organic dyes entrapped in them. In particular, the SiO 2 sol-gel glass provides a stable and easily made matrix for this purpose. This matrix has been widely used in previous investigations. There are many different organic dyes; among them, a great number has been produced for use in the food industry. Food dyes are cheap, easy to handle, and nontoxic and are produced in large quantities. To our knowledge, not much work has been done to incorporate this family of dyes into SiO2 sol-gel glasses, although they appear very appropriate for that. In the present study, we have investigated the preparation conditions necessary to introduce organic food dyes into a SiO2 sol-gel matrix and the optical properl This article was submitted by the authors in English.

ties of the trapped dyes. It is shown that large-area glasses of different uniform colors can be produced using the low-temperature sol-gel technology at low cost. The dispersion of the dye in glass was improved by milling the starting solution in a commercial ballmilling apparatus. It is found that the dye dispersion has an important effect on the color glass characteristics. The system was modeled using a simple quantummechanical treatment based on the FEMO (free-electron molecular orbital) and modified LCAO (linear combination of atomic orbitals) approximations. The molecular agglomeration and the possibility of electron interchange between molecules was taken into account. The results of the calculations are in good agreement with experiment. EXPERIMENTAL Uniformly colored glass coatings were obtained by incorporating dye powders into the starting or precursor solution composed of tetraethyl orthosilicate (TEOS), water, and ethanol. The ethanol-to-TEOS and water-to-TEOS molar ratios in the solution were 4 : 1

0020-1685/00/3610-1060525.00 9 2000 MAIK "Nauka/lnterperiodica"

OPTICAL PROCESSES IN SiO2 SOL-GEL GLASS COLORED WITH ORGANIC DYES (a) R I

(b) Bk I

OH

NaO. . . .

NaO3S-i '

U

1061

-I

o

'--N ----N-(

COONa

o O

--N = N - -

NHCOCH3

no SO3Na

803Na (c) B

(d) G

SO3Na /

HO-NaO3So N(C2H,)_( COH

/

SO3Na

-/O)__

)- N(C2Hs)CH2--((

O

)

~-C SO

O

)-- N(C2Hs)CH2-I

\ S03

SO3Na

(f) Y5 (e) YM 0 HCI"NH:=

N a O 3 S - - I O ) - - N----N--~- - CII- - COONa HO-C N \/ N

N(CH3)2 + H20

I

O I

SO3Na

Fig. 1. Molecularstructureof the organicdyes used: (a) red, (b) black, (c) blue, (d) green, (e) yellowM, and (f) yellow5. For details, see text. and 11 : 1, respectively. This composition has proven to produce high-quality SiO2 coatings [8, 16, 17]. TEOS was dissolved in ethanol using magnetic stirring for 15 min. The dye was dissolved in water and added to the TEOS-ethanol mixture to form the starting solution. To catalyze the gelation/condensation reaction, 3.4 • 10-4 mol of phosphoric acid per mole of TEOS was added. Coatings with the estimated amount of 3.6 and 7.2 wt % of the dye were studied. Colored coatings on microscope slide glass substrates were obtained using a dip-coating apparatus. The glass substrate was immersed in the starting solution, when its viscosity was about 3 cP, and then removed at a constant speed of 8 mm/s. After that the films were dried at 90~ for 1 h in an oven at atmospheric conditions. The films thus obtained were about 850 nm in thickness. In our experiments, six different types of dyes were used; their molecular structure is shown in Fig. 1. The common names of these dyes are: (1) FD&C red no. 3 (erythrosine, R), (2) brilliant black BN (Bk), (3) FD&C INORGANIC MATERIALS

Vol. 36

No. 10

2000

blue no. 1 (B), (4) FD&C green no. 3 (G), (5) 41000 C.I. Basic Yellow no. 2 (bright yellow or macrillon yellow, YM), and (6) FD&C Yellow no. 5 (Tartrazine, Y5). The letter or letters in parentheses will be used below to denote the dyes. Note that the shapes of the molecules of the blue and green colorants are very similar. Before the preparation of the coatings, in some cases the starting solution was subjected to ball milling in a plastic container 80 mm in diameter operating at 55 rpm (which corresponds to 0.39 of the critical velocity V c [18]), using ceramic balls 1 and 2 mm in diameter. The balls to solution volume ratio was about 2 : 1. The milling time was 2 or 4 h. The milling process was found to improve the dye dispersion in the glass coating. The room-temperature emission and excitation spectra were taken with a Fluorometer Fluoromax SPEX. The optical absorption spectra were obtained using a UV-Vis Perkin-Elmer Lambda 2 spectrometer. The surface of the coating was examined by atomic

1062

PI~REZ-BUENO et al.

9~ 4 8.5% 4"J1% I, \ ~

9~ 2 2

B273 nm

c-

E

~0

o

525

R200 nm

G318 nn

0

300

""

400

600 675 750 . Wavelength, nm

B320 nm

'

500

'

'

600

~

700

800 Wavelength, nm

Fig. 2. Emission spectra for the films colored with four different dyes (indicated by letters; the numbers show the excitation wavelength). The inset shows the long-wavelength band of the emission spectrum of the blue dye for different dye concentrations.

force microscopy (AFM) using a IIIa Digital Instruments nanoscope. Microfluorescence measurements were made with a home-made apparatus. In that experiment, the sample was placed at the bench of an optical microscope and illuminated with blue light for excitation. The room-temperature fluorescence was collected by a video camera and registered in a TV monitor. RESULTS Figure 2 shows the emission spectra for coatings containing 7.2 wt % of the B, R, G, and Bk dyes. The numbers used as subscripts show the excitation wavelength. The inset shows the effect of the dye concentration on the band at about 670 nm for the B dye; this result will be discussed below. Figure 3 shows the excitation and absorption spectra for the same dyes as in Fig. 2. The number used as a subscript in the letters identifying the dye indicates the position of the maximum in the emission band produced by the excitation spectra. The spectra identified with letters without subscripts correspond to the absorption spectra. Figure 4 shows the absorption spectra for the two yellow dyes; curve 1 for the Y5, and the rest for the YM dye. The inset in Fig. 4 will be discussed below. All the spectra contain well-defined bands, which correspond to different electronic energy levels of the molecules. The energies of the electronic transitions corresponding to each

4.5

3.0 --t

o

1.5

,=

0

300

400

500

600

700

Wavelength, nm Fig. 3. Absorption (letters without subscripts) and excitation (letters with subscripts) spectra of colored coatings with different dyes. The letters indicate the color; subscripts show the emission wavelength corresponding to a particular excitation spectrum. INORGANIC MATERIALS

Vol. 36

No. 10

2000

OPTICAL PROCESSES IN SiO2 SOL-GEL GLASS COLORED WITH ORGANIC DYES

1063

2.0 YM; Y5(1)

% 7 wt .2 %

0.75

. o.5o

1.5 .,.q

0.25 .~ = 1.0~ "~ |

4 h .***'~'~ / I ~ 300 *** ~ , 1 %

L

"[

,,

/

9

,,

i 400

~ ~""'~ . . . . r . . . . 500 Wavelength, nm

"

9

0.5

I- ~ . u ~ . , , 7 0

~ "--,r-. - - . . . . . . . . . . . . . . I

I

I

300

% K.__

--~mm[mmmmmmmmmmn~lmm ummmammp

400

500

600 Wavelength, nm

Fig. 4. Absorption spectra for the coating with 7.2 wt % of the yellow dye. Curve I correspondsto the Y5 dye without ball milling,

and the other curves to the YM dye after ball milling for the indicated times. Inset shows the effects of concentration on the absorption spectra for coatings containing the YM without milling. band are listed in Tables 1-5. To determine the transition energy, we took the values of the wavelength corresponding to half the maximum value on the longwavelength side of the absorption and excitation bands, and on the short-wavelength side of the emission bands. Note that the transition values for the blue and green colorants in sol-gel glass (Table 1) are practically identical. The effect of the ball-milling process is illustrated in Fig. 5, which shows 50 x 50 ~tm2 AFM images of the surface of the film containing the YM dye without milling and after 4 h of milling. The vertical scale is indicated by the shaded bars on the right side of the micrographs; the white spots in the surface micrographs corTable 1. Well 1.33 x 2.0 nm2 (blue and green dyes) nx

ny

E, e V

1 1

1 2

1.23

2 1 2 2 1

1 3 2 3 4

2.36 3.79 4.25 4.92 6.81 6.89

12 11 11 ~ 11 11 ~ 11

9- 13:1.89 ,. 21:2.56 13:3.02 - 22:3.69 23:5.58 - 14:5.66

INORGANIC MATERIALS

Vol. 36

1.88

1.88

2.92 3.65 5.52

2.53 2.92 3.70 5.52

No. 10

The influence of the dye concentration on the transition energy levels is illustrated by the inset in Fig. 2, which shows the emission band around 670 nm of the film containing the B dye. It can be seen that an Table 2. Well 1.22 x 2.55 nm 2 (black dye)

Eexp Eexp (blue), eV (green), eV

AE, eV

respond to elevated regions. Dye aggregates about 1 [tm in size determine the surface features before milling; after milling, these features disappear. Figure 6 shows microluminescence for a coating containing the YM dye; the white spots are fluorescent dye aggregates. After milling, uniform fluorescence is observed throughout the illuminated area; this indicates that the fluorescent aggregates are of much smaller size and cannot be resolved with the optical microscope used.

2000

nx

ny

E, eV

AE, eV

1

1

1.25

1

2

2.20

11 ~

1

3

3.10

2

1

1

Eexp, eV

13:1.85

1.85

12

- 21:2.09

2.07

4.29

12

,. 14:2.53

2.48

4

4.73

11

,- 21:3.04

3.1

2

2

4.99

11

- 22:3.74

3.65

1

5

6.81

I1

,- 15:5.57

5.52

1064

PI~REZ-BUENO et al.

Table 3. Well 1.40 x 1.13 nm2 (red dye)

nx

n:

E, eV

1

1

1.95

2

1

4.26

11

---21:2.31

2.4

1

2

5.50

lI

- 12:3.55

3.55

AE, eV

E~xp, eV 2.2

before [10, 14]. Thus, we conclude that in all cases we have dye molecular aggregates trapped in cases, which are probably pores in the glass, and that the interactions among the molecules within the same aggregate are much more essential than the interaction of an aggregate with its environment. DISCUSSION

Table 4. Well 1.27 x 1.0 nm 2 (yellow M)

nx

ny

E, eV

AE, eV

1

1

2.45

2

1

5.25

11

1

2

6.97

11 ----- 12:4.52

,'21:2.8

2

2

9.8

21 ~

22:4.55

3

1

9.93

21 ~

31:4.68

E~xp, eV

2.76 >4.4

9

linear size given by aNo

and a cross section given by

2 a r2/3

Table 5. Well 1.32 x 1.05 nm 2 (yellow 5)

AE, eV

Effects of aggregation. According to our results, a better dispersion of dye particles produces a higher optical density of the coatings. A simple model to account for this effect could be of the following type: the molecule is assumed to have a cubic shape of size a; the aggregates, also cubic, with No molecules, have a

nx

nv

E, eV

1

1

2.24

2

1

4.84

11

,- 21:2.6

2.6

1

2

6.34

11

--- 12:4.1

4.14

Eexp, eV

a ~v0 . The density of the aggregates, assumed to be uniformly distributed in the glass, is denoted by n. If N is the density of the colorant molecules, then n = N/No. There are two cases, depending on the amount of the aggregates and the thickness of the coating L. I. The degree of aggregation is very high (No >> 1), only the fraction Z of the total layer area shaded by the aggregates will be absorbing, having the optical density O' = d N 01/3, d being 9 the optical density of one molecule. If the total illuminated area is S, the corresponding coating volume is LS and the number of aggregates in this volume is nLS, then, Z is given by Z = nLSa2N2•/S =

increase in the dye concentration (indicated near the curves in wt %) increases the width of the emission band and shifts the position of its maximum to shorter wavelengths. Notice that the short-wavelength side of this band remains unchanged. The absorption spectra of the films containing the YM dye with concentrations of 10.8, 7.2 and 3.6 wt % and prepared from solutions without milling are shown in the inset to Fig. 4. As can be seen, the two spectra obtained from the coatings with the higher dye concentrations have two bands, whereas the spectrum of the sample with the lowest dye concentration has only the short-wavelength band. The general conclusion is that larger dye aggregation increases the interaction among the dye molecules and therefore reduces the electron transition energy. An additional effect of the improved dye dispersion is the increase observed in the absorption intensity due to longer milling treatment (see Fig. 4). The insets of Figs. 2 and 4 show that there is a noticeable influence of the degree of dye aggregation on the energy spectrum of the systems studied. Besides that, we have not noticed any effect on the spectrum due to interactions of the dye molecules with their environment. All the main bands in the optical spectra are practically the same for dyes in powder, in solution, and in the glass; similar observations have been reported

nLa2 N~o 3 "

Then we can write the light intensity passing through the layer as I = IoZe -~

= foe

-D*

,

where D* denotes the effective absorption coefficient of the whole layer. Therefore, e

-D*

= l-Z(1-e-~

(1)

The second term in the right-hand part is much less than 1. Taking the logarithm, we obtain D* = Z(1 - e -D') or D* = a2~v0"2/3nt, tl'" - exp [-dN~/3] .

(2)

F o r N 0 large enough (below, we will see that in our case 9 rl/3

No ~ 106, which gives an aggregate size, alv0 , of about 100 nm, which is large enough to justify the following approximation), the exponential in Eq. (2) is small compared to 1, so that D*

2/3 = a 2 .~v 0 nL. = a2NLNoo u3.

(3)

According to Eq. (3), the effective absorption coefINORGANIC MATERIALS

Vol. 36

No. 10

2000

OPTICAL PROCESSES IN SiO2 SOL-GEL GLASS COLORED WITH ORGANIC DYES

1065

Fig. 5. AFM images of the coatings containingthe YM dye: (a) without milling, (b) after millingfor 4 h. ficient decreases with an increase in the degree of aggregation (i.e., with No) and does not depend on the individual optical characteristic of the molecule, d. For the case considered (106 molecules in an aggregate) Eq. (3) gives an optical density of 0.4, which is in good agreement with experiment (Fig. 4). INORGANIC MATERIALS

Wol. 36

No. 10

2000

II. The degree of aggregation is not too high, but the percentage of the colorant is rather large, so that the parameter Z defined above is larger than 1. In this case, the coating can be viewed as made of several layers (Z) of colorant aggregates, without a free space for light passing among them.

1066

PI~REZ-BUENO et al.

Fig. 6. Fluorescence microphotograph of a coating containing the YM dye and prepared from solutions without milling.

(3)

(4)

¢)

100

101

102

103

104

105

106

107

108

109

1010

N O(Number of molecules for aggregate) Fig. 7. Effective optical density as a function of the size of dye aggregates.

For example, if we take No = 1, a = 10-7 cm, L = 4 x 10-5 cm, and n = 102o cm 3 (about 1% of the atomic density in glass), then

thus, the total optical density D of the doped coating will be Z times larger:

Z = 4 0 > > 1.

It is found then that for the case of small and welldispersed aggregates, the value of D does not depend on the degree of aggregation (No does not enter into expression (4), only the total amount of molecules N). From the experimental results, we consider the case of

If d is the optical loss due to the absorption by one molecule, expressed by e -d, then, the corresponding absorption probability for an aggregate will be d N o ;

D = ZdNlo/3 = a 2,,2/3 ~vo n ~. .a. t. v oi/3 = d a 2 L N .

INORGANIC MATERIALS

Vol. 36

No. 10

(4)

2000

OPTICAL PROCESSES IN SiO2 SOL-GEL GLASS COLORED WITH ORGANIC DYES coatings prepared from solutions after 4 h of milling, that is, a well-dispersed dye. For this case, D = 1; using the rest of the values as given above, we obtain d = 1 / ( a 2 L N ) = 0.025.

This value is used in subsequent calculations. The boundary between the two cases considered corresponds to Z = 1, or a2NLNoU3 = 1.

The corresponding degree of aggregation is given by o NIl3

= aZNL.

For L = 0.4 Bm, with the other values given above, we get N~r3 = 40. Figure 7 presents the calculated dependence of the optical density on the aggregation parameter No and also shows how the expression given by Eq. (2) describes the general behavior, while Eqs. (3) and (4) correspond to the limiting cases I and II. Energy spectra. The simplest and widely used model for the description of the energy spectra of organic molecules is the Free Electron Molecular Orbital (FEMO) model (see, for example, [20-22]). This model considers electrons delocalized inside a molecule, so that the molecular shape and dimensions determine the energy spectrum. In all cases, it is the quantum-mechanical "particle in a box" problem; depending on the shape of the molecule, it can correspond to a particle in a one-, two- or three-dimensional potential well or in a circular well. The depth of the well is usually considered infinite, which determines the zero boundary condition for the W-function and excludes the possibility of interchange between the molecules. In the case of a two-dimensional rectangular potential well with dimensions L x and Lv, this approximation gives the following expression for the electron energy levels [22]:

22 2)

h'fnx nv E = 8m(L2x + ..~ Ly ,

(5)

where h is the Planck constant, m is the electron mass, and nx and ny are integers starting from I.

It can be seen from Fig. 1 that all the dye molecules studied can be considered, to a first approximation, as rectangles or two-dimensional structures. The dimensions of the rectangles can be determined knowing the chemical bond lengths; these were taken from [23]. The calculated molecular dimensions are given for each dye at the upper part of Tables 1-5. As it is usually done [21, 24], one bond length on each molecule side was added. INORGANIC MATERIALS Vol. 36

No. 10

2000

1067

We estimate that the accuracy of our estimation is no worse than 10%. For these dimensions, the energy values calculated from the expression for E given above (for relatively small n values) are much lower than those observed in experiment. This inconsistency is certainly due to the fact that the expression used does not take into account the dye aggregation and the molecular interaction within aggregates, which is definitely present in our case. In order to introduce the interaction between the molecules in the same aggregate, we consider potential wells with a finite depth, so that the probability of an electron to be found outside a molecule (i.e., the possibility of an electron interchange) is nonzero. The most appropriate and general boundary conditions for this situation are the periodic (Born-Carman) boundary conditions. In this case, the values of the W-function repeat in a two-dimensional sequence of the molecules with periods equal to L x and Ly in x and y directions, respectively. This immediately leads to an expression for the system's energy levels different from that given above, namely:

h'( x n,,

E = 2 m . L ~ + -:3" Ly

(6)

Tables 1-5 list the energy levels and the electron transition energies determined by this expression; it is evident that the calculated values are very close to the experimental ones. Another approach frequently used to calculate the energy levels of molecular structures is the linear combination of atomic orbitals (LCAO) [20-22]. The application of this model, in its direct form, to our systems is rather complicated if one considers that the number of atoms in the molecules studied is almost one hundred; this determines the order of the corresponding secular equation and the number of solutions which give the energy levels of the system studied. Thus, in our case its use is not encouraging. We propose a modified version of the LCAO, which we call the Linear Combination of Complexes Orbitals (LCCO). This approximation can give some insight into the character of the molecular energy spectrum on a much simpler basis. Here we take into account the benzene ring, or other similar complexes, as the basic unit for calculations (instead of individual atoms as in the classic LCAO method) and consider the interaction between these complexes in the usual way. Thus, the order of the secular equation is drastically reduced, as well as the number of the energy levels; their sequence is obtained from the solution of the equation. Figure 8 shows the schemes for the calculations for the type of dyes studied: (a) for black, (b) for blue and

PI~REZ-BUENO et al.

1068

~

-6

.

~,,~=p i,j= 1-5

i

67

-6

i i ,

i

i

i

e=

~,,~ ~

;

>

4-

-4 "2

1112 = 1113 = 1135 ---- 1156 = 1146 ---- 1]

111,= 0.71]; 1134: 0.31]

(i,j 1...5)

~

=

2-

>4"

i

o

ii-

al 2 -

-2

i i

61

J

ibla

(h)

0

0

t~

i

t

(a)

-4 "~

i

0

i

i

i

i

"6 i i i ,

////,

i

-4 "~

i

i

"2

i i i i

t

I

i

i J

I I

R iY5"YM"

(c)

0"

t~

0

Fig. 8. Schematicillustrationof the LCCOapproachappliedto differentdye molecules. green, and (c) for red and yellow. Thus, for the black dye we obtain the following determinant: x

l

0

0

0

1

x

1

0

1

0

1

x

1

{3=0,

0

0

1

x

C

0

1

0

0

which gives a 5 th order equation for x, which is a linear function of the energy. The values of the so-called resonance integrals (denoted by ~ij) determine the interaction between the elements of the molecules and can be varied according to the type of the interaction between different parts of the molecule. The energy diagrams in Fig. 8 compare the experimental energy levels for the different dyes with those calculated using the LCCO approximation. It can be stated that an essential agreement exists between the calculated and experimental spectra and, furthermore, that this agreement can be better when the interactions among the elements of the molecule are properly selected. As an example, we show (Fig. 8b) two sets of calculated energy levels for the blue/green dyes. The set of levels identified with the letter b, were obtained considering an additional interaction in the left part of the molecule (Fig. 8b); this results in much better agreement with experiment. This shows that our sim-

plified LCCO approach can provide a useful analysis of complicated organic molecules. CONCLUSION It is shown that the sol-gel technology is an efficient, simple, and cheap method to produce systems in which organic molecules are embedded in inorganic matrices and that the molecules can be considered as two-dimensional quantum wells for the calculation of the energy spectra. The energy levels of six types of the food dyes were investigated experimentally with the help of the optical absorption, luminescence excitation, and emission techniques. The effect of the organic molecules aggregation on the energy spectra was studied with the use of atomic force and fluorescence microscopy. A new version of the FEMO model to calculate the energy spectra of complex molecules was developed; it takes into account molecule aggregation effects. A simplified version of the LCAO method is also proposed for calculating the energy spectra of molecules. Both models give reasonably good description of the experimental data; in the case of the FEMO, no adjustable parameters are needed. REFERENCES 1. Tentorio, A., Matijevic, E., and Kratohvil, J.P., Preparation and Optical Properties of Spherical Colloidal Aluminium Hydroxide Particles Containing a Dye, J. Colloid Interface Sci., 1980, vol. 77 (2), pp. 418--426. INORGANIC MATERIALS Vol. 36

No. 10

2000

OPTICAL PROCESSES IN SiO2 SOL-GEL GLASS COLORED WITH ORGANIC DYES 2. Knobbe, K.T., Dunn, B., Fugua, P.D., and Nishida, F., Laser Behavior and Photostability Characteristics of Organic Dye Doped Silicate Gel Materials, Appl. Opt., 1990, vol. 29 (18), pp. 2729-2733. 3. S~inchez, C. and In, M., Molecular Design of Alkoxide Precursor for the Synthesis of Hybrid Organic-Inorganic Gels, J. Non-Cryst. Solids, 1992, vol. 147/148, pp. 1-12. 4. Hinsch, A., Organic Fluorescent Dyes in Organically Modified A1203-SiO2 or TiO2-SiO 2 Coatings with Varying Polymethacrylate Content, J. Non-Cryst. Solids, 1992, vol. 147/148, pp. 478-482. 5. Ocafia, M., Levy, D., and Serna, C.J., Preparation and Optical Properties of Spherical Metal Oxide Particles Containing Fluorescent Dyes, J. Non-Cryst. Solids, 1992, vol. 147/148, pp. 622-626. 6. Hou, L. and Schmidt, H., Effect of Additives on the Photostability of Sol-Gel Derived Organic-Inorganic Photochromic Coatings, J. Mater. Sci. Lett., 1997, vol. 16, pp. 415-416. 7. Avnir, D., Levy, D., and Reisfeld, R., The Nature of the Silica Cage as Reflected by Spectral Changes and Enhanced Photostability of Trapped Rhodamine 6G, J. Phys. Chem., 1984, vol. 88, pp. 5956-5959. 8. Avnir, D., Kaufman, V.R., and Reisfeld, R., Organic Fluorescent Dyes Trapped in Silica and Silica-Titania Thin Films by the Sol-Gel Method, J. Non-Cryst. Solids, 1985, vol. 74, pp. 395-406. 9. Fournier, T., Tran-Thi, T.-H., Herlet, N., and S~inchez,C., Charge Transfer Dynamics of Porphyrin-Phthalocyanine Heterodimers in Hybrid Sol-Gel Films, Chem. Phys. Lett., 1993, vol. 208, pp. 101-105. 10. De Matteis, E, Prosposito, P., Sarcinelli, E, et al., SilicaBased Sol--Gel Films Optically Functionalized through Doping with Organic Molecules, J. Non-Cryst. Solids, 1999, vol. 245, pp. 15-19. 11. Chen, O., Gu, D., Shu, J., et aL, Optical and Recording Properties of Copper Phthalocyanine Films, Mater Sci. Eng, B., 1994, vol. 25, pp. 171-174. 12. Salin, E, Le Saux, G., Georges, P., et al. Efficient Tunable Solid-State Laser near 630 nm Using Sulfor-

INORGANIC MATERIALS

Vol. 36

No. 10

2000

13.

14. 15.

16.

17.

18.

19.

20. 21. 22. 23. 24.

1069

hodamine 640-Doped Silica Gel, Opt. Lett., 1989, vol. 14, pp. 785-787. Canva, M., Georges, P., Brun, A., et al., Impregnated SiO2 Gels Used as Dye Laser Matrix Hosts, J. NonCryst. Solids, 1992, vol. 147/148, pp. 636-640. Jager, S., Dye-Metal Oxide Composite Thin Films for Decorative Applications, Thin Solid Films, 1996, vol. 286, pp. 154-158. Atkinson, A., Doorbar, J., Hudd, A., et al. Continuous Ink-Jet Printing Using Sol-Gel "Ceramic" Inks, J. SolGel Sci. Technol., 1997, vol. 8, pp. 1093-1097. Brinker, C.J. and Sherer, G.W., Sol - Gel ,- Glass: Gelation and Gel Structure, J. Non-Cryst. Solids, 1985, vol. 70, pp. 301-322. Martfnez, J.R., Ruiz, F., Vorobiev, Y.V., et al., Infrared Spectroscopy Analysis of the Local Atomic Structure in Silica Prepared by Sol-Gel, J. Chem. Phys., 1998, vol. 109, pp. 7511-7514. Wills, B.A., Mineral Processing Technology: An Introduction to the Practical Aspects of Ore Treatment and Mineral Recovery, London: Pergamon, 1987, 2nd ed., p. 176. Dfaz-Flores, L.L., Pfrez-Bueno, J.J., Vorobiev, Y.V., et al. The effect of Ball Milling Dispersion on the Optical Properties of Organic Dyes Trapped in Silica Films by Sol-Gel Method, Mater Lett., 1999. Griffits, J., Color and Constitution of Organic Molecules, London: Academic, 1976, pp. 24-29. Murrell, J.N., The Theory of the Electronic Spectra of Organic Molecules, New York: Wiley, 1963, pp. 30--46. Birks, J.B., Photophysics of Aromatic Molecules, London: Interscience, 1970, pp. 1-26. CRC Handbook of Chemistry and Physics, Weast, R.C. and Astle, M.J., Eds., Boca Raton: CRC, 1979, pp. F216-219. Platt, J.S., The Box Model and Electron Densities in Conjugated Systems, J. Chem. Phys., 1954, vol. 22, pp. 1448-1455.