Dual-functionality glasses: Optimizing optical transmission and gamma-ray attenuation in transition metal oxide-doped bismuth silicates for next-generation windows

2.1. Materials

Pure reagent chemicals, including Bismuth Oxide (Bi₂O₃) was supplied by Panreac Co (Spain), Silicon Dioxide (SiO₂) Lanxess Co. (India) as the major glass network. Analytical grade chemicals of Copper Oxide (CuO), Nickel oxide (NiO), Cobalt Oxide (CoO), Iron Oxide (Fe₂O₃), Manganese oxide (MnO), Chromium Oxide (Cr₂O₃), Vanadium Oxide (V₂O₅), and Titanium Dioxide (TiO₂) was supplied by Sigma Aldrich Co. and used as received.

2.2. Sample preparation

Glass samples from the system were 60Bi₂O₃-40SiO₂ mol%, and other samples contained 1% from one of the first raw TMO (Ti→Cu) at the expense of SiO₂. The precisely weighed batches were ground and mixed using an agate mortar and then placed in porcelain crucibles. They were gradually heated to a temperature ranging between 1150 and 1200°C, based on the sample composition and type of TMO. The obtained melt was swirled several times at fixed time intervals to ensure homogenous and bubble-free samples. The melt was then poured into a preheated stainless-steel mold with the required dimensions.

2.3. Characterization techniques

Fourier transform- infrared (FT-IR) spectra measurements were carried out using a Bruker FT-IR spectrometer (Invenio S, Germany). The spectral resolution of 4 cm^-1 and scan number of 64 within the wavenumber range of 4000–400 cm^-1 were applied to the collection of ATR spectra. The UV/Visible spectrophotometer (V-570 UV/VIS/NIR, Japan) measured optical behavior within the range of wavelength λ=200–1000 nm. A user-friendly online Photon Shielding and Dosimetry (PSD) software available at https://phy-x.net/PSD was used to calculate the parameters relevant to shielding and dosimetry including linear and mass attenuation coefficients (LAC, MAC), half and tenth value layers (HVL, TVL), mean free path (MFP), effective atomic number and electron density (Z _eff, N _eff), effective conductivity (C _eff), and energy absorption and exposure buildup factors (EABF, EBF). Data were generated in the continuous energy region (1 keV-100 GeV) [17].

3.1. X-ray analysis

Figure 1 shows the X-ray diffraction (XRD) patterns of parent bismuth silicate glass and other samples that were doped with 1 mol% of different TMOs, including V₂O₅, TiO, Cr₂O₃, Fe₂O₃, CuO, MnO, CoO, and NiO, along with sample images. The XRD pattern of pure bismuth silicate glass exhibits a characteristic broad, intense peak at approximately 28°, indicative of an amorphous structure with a short-range order typical of glasses. This peak likely corresponds to Bi-O and Si-O bond distances in the glass network [14]. The broad nature of the peak suggests a lack of long-range crystalline order and variability in interatomic distances and bond angles, which is expected in a glass structure. The addition of 1% of first-row TMOs (from TiO₂ to CuO) to the bismuth silicate glass results in XRD patterns that maintain the broadband at around 28°. This persistence indicates that the glass structure remains predominantly amorphous, with the low concentration of additives not significantly altering the overall glass network. However, restrained changes may be present, such as slight shifts in peak position, minor variations in peak width or intensity, or the appearance of very small additional features. These restrained modifications could reflect changes in average interatomic distances, slight alterations in local ordering, or variations in electron density distribution due to the incorporation of transition metals into the glass structure [14,18,19].

Close

Figure 1.

(a) Glass sample images. (b) XRD of base bismuth silicate glass and samples doped with different TMOs.

Export to PPT

The consistent presence of the broad peak across all samples, including those with TMO additions, confirms that the basic bismuth silicate glass network remains intact. The absence of sharp crystalline peaks indicates that the 1% addition of TMOs does not lead to crystallization, suggesting that these metals are likely dissolved in the glass, occupying interstitial sites or substituting for Bi or Si in small amounts.

3.2. FTIR interpretations

Figure 2 illustrates the Fourier transform infrared (FTIR) absorption spectra of binary bismuth silicate glasses doped with 1 mol% of different TMOs in the range of 4000 - 400 cm^-1. The infrared absorption spectra of all the glasses exhibit general characteristics like those typically observed in silicate glasses, rather than in borate or phosphate glasses. In the FTIR spectra of the investigated glasses, there are no major changes after the addition of 1 mol% of TM oxides because the coordination shells around the IR luminescence centers stay stable. The first shell consists of oxygen atoms, and the second shell is made up of positive ions from the doped oxide, maintaining consistency despite changes in the dopant oxide type [3,20]. The main high-intensity absorption bands are concentrated in the region of 400–2000 cm^-1, with two particularly prominent bands at 471–456 cm^-1 and 700–1250 cm^-1. These specific absorption bands are mainly attributed to vibrations of characteristic structural chains, including silicate groups and transition metal ions, within the glass network. There is a minor change because of transition metal dopants. A broad band in the range of 700-1250 cm^-1 is assigned to the vibrations of Si–O–NBO bonds due to the addition of bismuth oxide as an intermediate. This band is also attributed to the stretching vibrations of Bi–O bonds and can be deconvoluted into separate peaks as shown in Figure 2. The peak at about 860 cm^-1 is due to [BiO₆] octahedra, which act as glass modifiers, and the peak at about 847 cm^-1 is due to [BiO₃] polyhedra, which act as network formers [21]. The band appearing at low wavenumbers is related to vibrations originating from Bi–O bonds in the [BiO₆] octahedra and the cation vibrations resulting from the incorporation of transition metals as dopants. The FTIR absorption bands located at 3440 cm^-1 and 1640 cm^-1 are generally attributed to vibrations of H₂O and hydroxyl (OH) groups, respectively.

Close

Figure 2.

(a) FTIR absorption spectral data (FTIR of base bismuth silicate glass and samples doped with different TMOs) (b) exemplified analysis of deconvoluted spectra for glass containing NiO.

Export to PPT

3.3. Physical properties

Table 1 gives the density, molar volume, and other physical parameters of the studied glasses. The variation in density values depends on the type of TMO doped and the relative molecular weight of the TMO relative to SiO₂. The density increased with the substitution of SiO₂ by TM oxides, and it was found that glass doped with nickel oxide (NiO) has the highest density compared to those doped with other TMO, as shown in Figure 3(a). This increase is because NiO has a high intrinsic density (around 6.67 g/cm³), and incorporating NiO into the glass network increases the overall density of the glass, leading to a more compact glass network structure with a lower molar volume. Ni cations may occupy interstitial vacancies when incorporated into the glass matrix. Consequently, increasing the NiO content (5% wt) should reduce the number of structural defects (free space) within the glass network, thereby decreasing the volumetric density of defects. This reduction in defects should lead to a decrease in the molar volume and an increase in the rigidity of the glass. In contrast, among all the TM oxides doped in the studied glasses, V₂O₅ exhibited the opposite behavior, with both density and molar volume increasing compared to the base sample, Figure 3(b). These increases are related to the higher molar mass of V₂O₅, which, when substituted for SiO₂, raises the density, and the former role of V₂O₅ in forming new groups (VO₄), which leads to an increase in molar volume.

Close

Figure 3.

(a) Density vs. 1 mol% of TMO for all investigated glasses, (b) molar volume vs. 1 mol% of TMO for all investigated glasses.

Export to PPT

Figure 4(a), displays that free volume is reduced by incorporating TM oxides into the glass network, and the packing density takes variation values depending on the structural role of each one, Figure 4(b). For instance, the incorporation of 1 mol% of NiO content into the binary bismuth silicate glass network increases the overall density while reducing the molar volume. This result indicates that the glass network becomes more tightly packed. As a result, the packing density increases because more mass is contained within a smaller volume, leading to a more compact structure. The incorporation of V₂O₅ results in increases in both density and molar volume. This suggests that while the glass becomes denser, the structure does not become more tightly packed in the same way as with NiO. Instead, the formation of new structural groups (VO₄) likely creates additional space within the glass matrix, resulting in more voids or empty space compared to the NiO-doped glass. The substitution of SiO₂ by TMO has a significant impact on the average molecular weight, inter-nuclear distance, ion concentration, field strength, oxygen packing density, and oxygen molar volume, as shown in Table 1. The variation in these parameters is affected by the rearrangement of the lattice within the glass network, the compactness of the glass network, the number of bonds per unit volume, the stretching force constant of the bonds inside the glass, and the density of the glass. As a result, the addition of the TMO makes the glass network more compact and tightly packed [22,23].

Close

Figure 4.

(a) Packing density vs. 1 mol% of TMO for all investigated glasses, (b) free volume vs. 1 mol% of TMO for all investigated glasses.

Export to PPT

3.4. Optical parameters

3.4.1. UV-visible absorption spectra

All investigated glass samples exhibit a strong absorption band at about λ = 220 nm, which is attributed to iron impurities from the raw materials used for synthesizing the glass samples [24,25]. The optical absorption spectra of all studied glass samples display a broad band extending from 400 to 600 nm, which is associated with the absorption contribution of the Bi³⁺ ions (related to the transition of the molecules in the lower energy state to the higher state when absorbing a significant amount of energy from the incoming photons) together with that corresponding to TM ions present in the glass samples. For instance, the bismuth silicate glass samples containing Cr₂O₃, CoO, MnO, CuO, and NiO exhibited a broad band in the visible region, with different positions of these bands depending on the type of transition metal ions. Figure 5 displays the UV-Vis absorption spectra of bismuth silicate glasses incorporated into the glass matrix. The absorption spectrum of bismuth silicate glass doped with CoO shows absorption bands at about 520 nm due to the ⁴T_1g(F) $\to$ ²T_1g (H) octahedral transition, 644 nm due to the ⁵T_2g $\to$ ⁵E_g octahedral transition of Co³⁺ ions [26], and 720 nm. In contrast, the glass doped with Cr₂O₃ displays absorption bands at around 440 nm due to the ⁴A₂ $\to$ ⁴T₂ Cr³⁺ ions transition, and 610-700 nm due to the ⁴A_2g $\to$ ⁴T_2g, ⁴A_2g $\to$ ⁴T_1g, and ⁴A_2g $\to$ ⁴E_g Cr³⁺ ions transition [27].

Close

Figure 5.

UV-Vis absorption spectra of bismuth silicate glasses.

Export to PPT

The absorption spectrum of CuO-doped bismuth silicate glass reveals a broad band in the wavelength range of 600 – 900 nm due to the absorption by Cu²⁺ ions in distorted octahedral coordination sites within the glass matrix [28]. As observed in the absorption spectrum of the Bi₂O₃-SiO₂ glass sample doped with MnO, a weak broad absorption band appears in the wavelength range of 700 – 900 nm, while the glass doped with NiO shows a broad absorption band within the range of 400 – 900 nm. In the glass containing NiO, a broad absorption band in the visible region indicates that the transmittance decreases as a result of the decrease in the number of defects upon incorporating NiO into the glass matrix.

3.4.2. Optical band gap

Optical energy band gap values for all the prepared samples were determined utilizing the Mott and Davis formula [29]:

$\left(\alpha hv\right)=B{(hv-E_g)}^r$

For the allowed indirect and direct electronic transitions, the index number r = 2 and 0.5, the values of the optical energy gap can be determined from the linear portion of the plotting ${\left(\alpha hv\right)}^{0.5}$ vs. $hv$ (eV), and ${\left(\alpha hv\right)}^2$ vs. $hv$ (eV), respectively, as shown in Figures 6 and 7. Refractive index, molar refractivity, reflection loss, and some other optical parameters were determined depending on the obtained energy band gaps, and the calculated values have been listed in Table 2. It is observed that the optical energy gap shows varying values (nonlinear behavior) depending on the variation of the TMO dopant. Among the investigated glasses, the glass containing NiO exhibits a higher energy gap due to the combination of structural reorganization and modifications in the electronic structure that result from the incorporation of NiO. These changes can lead to a more disordered glass network with modified electronic states, resulting in an increased band gap.

Close

Figure 6.

Plotting ${\left(\alpha hv\right)}^{0.5}$ vs. $\text{hv}$ (eV).

Export to PPT

Close

Figure 7.

Plotting ${\left(\alpha hv\right)}^2$ vs. $\text{hv}$ (eV).

Export to PPT

Table 2. Calculated optical parameters of all investigated glasses.

Optical parameters	TM Glass Code
	Base	V2O5	TiO2	Cr2O3	Fe2O3	MnO	CuO	CoO	NiO
Egap	2.55	2.71	2.67	2.41	2.48	2.76	2.75	2.57	2.93
Refractive Index (n)	2.53	2.48	2.49	2.58	2.55	2.46	2.47	2.52	2.42
Molar refractivity RMolar (cm3/mol)	30.50	30.08	30.07	30.95	30.71	29.70	29.70	30.28	29.14
Molar polarizability${\alpha }^{\text{Molar}}$ 10−24 cm3	12.10	11.94	11.93	12.28	12.19	11.79	11.79	12.01	11.56
Optical transmission TOptical	0.68	0.69	0.69	0.67	0.68	0.70	0.70	0.69	0.71
Reflection loss RLoss	0.188	0.181	0.183	0.195	0.191	0.179	0.179	0.187	0.172
Metallization criterion M	0.357	0.368	0.365	0.347	0.352	0.371	0.371	0.359	0.383
Static dielectric constant ${\epsilon }^{\prime }$	6.40	6.16	6.22	6.65	6.52	6.08	6.09	6.37	5.84
Optical dielectric constant ${\epsilon }^{″}$	5.40	5.16	5.22	5.65	5.52	5.08	5.09	5.37	4.84
The electronic polarizability 1024	0.255	0.251	0.252	0.259	0.257	0.249	0.250	0.255	0.245
Optical electronegativity	0.686	0.727	0.716	0.647	0.666	0.742	0.739	0.691	0.787

3.5.1. LAC

In the studied glasses, different TMOs, including V₂O₅, TiO, Cr₂O₃, Fe₂O₃, CuO, CoO, and NiO, have varying effects on LAC. The ability of glass to attenuate the gamma radiation depends on the type of TM oxide incorporated into the glass matrix, as shown in Figure 8(a). Bismuth silicate glass doped with NiO has a higher atomic number and density compared to other TMOs, contributing to a higher LAC when doped into the glass. Different interaction processes affect LAC behavior, the photoelectric absorption mechanism is dominant at lower energy regions, while Compton scattering and pair production processes become more prominent at medium and high energy regions, respectively.

Close

Figure 8.

Polts of (a) LAC vs. E, (b) MAC vs. E, MAC values of commercial glasses and (c) NiO as selected study glasses, (d) HVL vs. E, (e) TVL vs. E, (f) MFP vs. E.

Export to PPT

3.5.2. MAC

For describing the total interaction of gamma photons with a glass matrix, the MAC is a widely used parameter. A higher MAC signifies a higher probability of a glass interacting with, absorbing, or attenuating incident photons. The MAC of a medium depends on the incident photon energy and the chemical composition of the interacting glass. Additionally, MAC is a linear combination of the MACs resulting from the major interaction processes. Generally, photons interact with glass in five different ways: coherent scattering, photoelectric ionization, Compton scattering, electron-positron pair production, and photodisintegration [30]. Figure 8(b) shows the variation of MAC for all prepared glasses as a function of incoming photon energy in the range of 0.015 MeV to 15MeV, respectively. It is observed that MAC has a higher value at lower incident energy, and then it decreases with rising photon energy. The reason behind this behavior is the contribution of photoelectric absorption at lower energy regions due to the interaction occurring between the incident photon energy and the glass. In the medium energy region, the Compton scattering process is dominant, while in the higher energy region, the dominant process is the pair production mechanism [31]. The MAC also increased with the addition of TMOs in the present glass system due to the increase in the atomic number and its density. A comparison was performed between the MAC calculated for the present glasses and commercial Schott glasses (RS 253 G18, RS 253 G19, RS 360, and RS 520) containing the PbO compound, as shown in Figure 8(c). It is evident that the MAC of the studied glasses, NiO as a selected sample, are comparable to those measured for commercial glasses containing lead reported by several researchers [31-33]. It is higher than the MAC measured for commercial glasses containing PbO, indicating that the studied glasses are more efficient than commercial glasses in terms of γ photon absorption. This means that the NiO sample, with the highest MAC in the respective power spectrum, is the best photon shield.

3.5.3. HVL – TVL - MFP

Material thickness is an important factor for designing applicable shielding materials, depending on the desired application. HVL and TVL represent the thickness of the materials required to reduce the intensity of the radiation by half and a tenth, respectively. In medical and industrial applications, they help in determining safe exposure levels and designing protective barriers to minimize radiation doses. They allow for easy comparison between different shielding materials. Materials with lower HVL and TVLs are generally more effective. Figure 8(d,e) show the HVL and TVL plots against the incoming photon energy. The glass sample named NiO achieved lower values among all the studied glasses. The HVL of the studied glasses varies from 10 micrometers at 0.015 MeV to about 2.675 cm for the base sample, 2.674 cm for V₂O₅, 2.669 cm for TiO₂, 2.666 cm for Cr₂O₃, 2.666 cm for Fe₂O₃, 2.664 cm for MnO, 2.660 cm for CuO, 2.659 cm for CoO and 2.658 cm for NiO glasses. MFP represents the average distance a photon travels through a material before interacting with it. Figure 8(f) displays the plot of MFP versus energy for all studied glasses. The values of MFP for the studied glasses are placed between 3.08 cm to 3.06 cm at 15 MeV, and the glass coded as NiO achieved the lowest value (3.06 cm), indicating a high frequency of interactions between radiation and the material. This means the material is highly effective at attenuating radiation because the radiation particles are likely to interact and lose energy over shorter distances within the material. Therefore, a low MFP signifies that the material is a good radiation shield. Combining HVL, TVL, and MFP offers a comprehensive approach to shielding design. HVL and TVL provide practical measurements for determining the necessary thickness of shielding materials, while MFP offers a deeper understanding of how radiation interacts within the materials.

Table 3 presents the investigated HVL values of glasses at Eγ= 662 keV compared with experimental values of some standard concrete and glasses, such as Commercial window glass [34], Serpentine [35], and Ordinary concrete [36]. As shown in Table 3, the current glasses have HVL values lower than all the standard materials compared. Therefore, the studied glasses have a higher capacity for radiation shielding than some commercial materials.

3.5.4. Effective electronic density (N_eff)-Effective conductivity (C_eff)

Effective electronic density and effective conductivity are fundamental parameters that significantly impact the shielding properties of the glass materials. They determine the material’s ability to interact with and attenuate radiation, playing a crucial role in designing and applying radiation shielding solutions across various fields. Figure 9(a,b) shows the variation of N_eff and C _eff versus photon energy. The results revealed higher values at lower energy due to the dominance of the photoelectric process in this region. The effect of Compton scattering appears in the intermediate region of photon energy, where values sharply decrease, and then start to increase as pair production takes over the interaction process at higher energy levels, as shown in Figure 9. The increase in both N_eff and C_eff is due to the addition of TM oxides into the glass matrix as a dopant. Furthermore, the N_eff and C_eff values in the NiO glass sample were the highest among all investigated glasses. This makes it more efficient at absorbing incident photon energy and a promising option for gamma radiation shielding materials.

Close

Figure 9.

(a) N _eff vs. E, (b) C _eff vs. E for all investigated glasses.

Export to PPT

3.5.5. Effective atomic number (Z_eff)-Equivalent atomic number (Z_eq)

The calculation of the effective atomic and equivalent atomic numbers for all studied glasses have been shown in Figure 10(a,b). The number of electrons in the atoms of shielding glass materials significantly influences their interaction with photon energy, determining the material’s effectiveness in attenuating different types of radiation. In this case, the values showed the same behavior with the effective electronic density parameter, Figure 9(a).

Close

Figure 10.

(a) Z _eff vs. E, (b) Z _eq vs. E for all synthesized glasses.

Export to PPT

The results indicated that Z_eff values increased with the addition of TMOs, which showed good energy absorption. From Figure 10(a), in the lower energy region, the probability of the photoelectric effect occurring increases with increasing atomic number because heavier atoms have more tightly bound electrons that are more likely to interact with the incoming photons. In the medium and higher regions of incident energy, the Compton scattering and pair production processes are dominant, respectively.

The equivalent atomic number, Z_eq​, is a value associated with the multiple scattering energy from the Compton scattering effect [37]. At very low photon energies, where the photoelectric effect predominates, Z_eq​ ​values are significantly lower. As the photon energy increases to a middle range, the Z_eq​ ​ values gradually rise due to the dominance of Compton scattering. At higher photon energies, Z_eq​values decrease again, as pair production becomes the primary interaction mechanism, see Figure 10(b).

3.5.6. Atomic cross section (σ_a)-Electronic cross section (σ_e)

It is important to know the atomic cross-section (σ_a​) and electronic cross-section (σ_e) as shielding parameters. The calculation of atomic cross-section and electronic cross-section for all studied glasses is shown in Figure 11(a,b). As can be seen, with increasing incident photon energy, both σ_a and σ_e​ values decrease. The behavior of the atomic cross section for the samples is similar, with values ranging from 1.16×10⁻²⁰ 1.16 at lower incident energy to 6.10×10⁻²⁴ at higher incident energy. Meanwhile, the electronic cross-section values range from 1.46×10⁻²² at lower energy to 1.14×10⁻²⁵ at higher incoming energy. Moreover, in every sample, σ_a​values are greater than σ_e values because the likelihood of complete atomic interaction in a material is much higher than the probability of full electrical contact with incoming photons in that material.

Close

Figure 11.

(a) σ_a vs. E, (b) σ_e vs. E for Bi₂O₃-SiO₂-1 TMOs glass system.

Export to PPT

Experimentally, studying the effects of first-row TMOs (CuO, NiO, CoO, Fe₂O₃, MnO, Cr₂O₃, and TiO₂) on the Bi₂O₃-SiO₂ glass system proves that NiO exhibits the best combination of optical and shielding properties. NiO enhances optical transparency by introducing controlled absorption and a greenish tint without excessive discoloration, while also increasing the refractive index for optical applications. It significantly improves radiation shielding by increasing the glass density and providing efficient absorption of high-energy photons. Additionally, NiO stabilizes the glass structure without causing phase separation or instability, unlike other oxides such as CuO or Fe₂O₃. Its synergistic interaction with Bi₂O₃ further enhances both optical and shielding performance, making NiO the optimal dopant for applications requiring a balance of transparency, refractive index, and radiation protection.

Finally, the structure-property relationships between the role of each oxide and optimizing glass composition are key to tailoring the glass’s optical, shielding, and structural properties. Bi₂O₃ acts as a network former/modifier, increasing density, refractive index, and radiation shielding due to its high atomic weight and polarizability, while maintaining optical transparency. SiO₂ serves as the primary network former, providing a stable tetrahedral [SiO₄] structure that ensures mechanical, thermal, and chemical stability, along with optical clarity. TMOs like NiO, CuO, Fe₂O₃, and others act as network modifiers, introducing metal ions that break the SiO₂ network and create non-bridging oxygen atoms, influencing properties such as density, coloration, and radiation absorption. Among these, NiO stands out for its balanced contribution, enhancing density and shielding without excessive coloration, while fine-tuning the refractive index. By combining Bi₂O₃ for high density and shielding, SiO₂ for structural stability, and NiO for optimized optical and shielding performance, the glass composition achieves an ideal balance for applications requiring both transparency and radiation protection. This synergy allows for the systematic design of high-performance glass tailored to specific needs.