Egyptian blue, optimum excitation of IR fluorescence
The absorption coefficient of the powdered pigment is shown as the light blue line. The three maxima at 530nm, 630nm and 790nm identify absorption from the ground state of the Cu^2+ ions (^2B_1g) to the levels ^2A_1g, ^2E_g and ^2B_2g.
A filtered high-pressure 'energy-saving' Hg bulb provides an excitation source that accesses in particular the ^2E_g level. The filters used here are a Schott OG570 (low-pass) and a BG38 (high-pass). The fluorescence spectrum (EB_Hg, orange line) is observed through a 610nm low-pass filter and shows the resulting infrared fluorescence peaking at 908nm (ambient temperature 26°C).
The pigment has also been excited by three lasers: 404nm, 532nm and 633nm. The fluorescence spectra are shown by the purple, green+turquoise and red lines respectively (stepped by 0.25 units up the vertical scale). Shining the focussed 532nm laser (300mW) directly on the powdered pigment (green line) produced a shifted and broadened fluorescence peak - probably due to local heating of the grains. When diffused with ground glass, the same laser resulted in the turquoise spectrum which is congruent with the other exciters. The three laser spectra have been scaled to have the same peak intensity but the efficiency of excitation is in the order of 404nm (lowest; actually very low), 532nm and 633nm, the latter wavelength being optimum to excite the Cu^2+ ion to the upper level (^2E_g) of the 910nm fluorescence.
The wavelengths of the three lasers are shown schematically in the figure.
The 'rough' structure near the peak of the 633nm laser-excited fluorescence peak (890-950nm) is real and may be due to diffraction effects from the powdered pigment surface. This effect can also be seen in the (green curve) spectrum from the direct 532nm laser but not in the (turquoise curve) diffused laser spectrum which is very smooth.
The temperature dependence (below 300K) of the peak fluorescence wavelength has beed addressed by Yixi Zhuang and Setsuhisa Tanabe, "Forward and back energy transfer between Cu21 and Yb31 in Ca12xCuSi4O10:Ybx crystals", J. Appl. Phys. 112, 093521, 2012. This indicates that, around room temperature, there is a redwards shift of approximately 2nm per 10°C increase. This implies that my green laser was heating the powdered sample to about 100°C.
Egyptian blue, optimum excitation of IR fluorescence
The absorption coefficient of the powdered pigment is shown as the light blue line. The three maxima at 530nm, 630nm and 790nm identify absorption from the ground state of the Cu^2+ ions (^2B_1g) to the levels ^2A_1g, ^2E_g and ^2B_2g.
A filtered high-pressure 'energy-saving' Hg bulb provides an excitation source that accesses in particular the ^2E_g level. The filters used here are a Schott OG570 (low-pass) and a BG38 (high-pass). The fluorescence spectrum (EB_Hg, orange line) is observed through a 610nm low-pass filter and shows the resulting infrared fluorescence peaking at 908nm (ambient temperature 26°C).
The pigment has also been excited by three lasers: 404nm, 532nm and 633nm. The fluorescence spectra are shown by the purple, green+turquoise and red lines respectively (stepped by 0.25 units up the vertical scale). Shining the focussed 532nm laser (300mW) directly on the powdered pigment (green line) produced a shifted and broadened fluorescence peak - probably due to local heating of the grains. When diffused with ground glass, the same laser resulted in the turquoise spectrum which is congruent with the other exciters. The three laser spectra have been scaled to have the same peak intensity but the efficiency of excitation is in the order of 404nm (lowest; actually very low), 532nm and 633nm, the latter wavelength being optimum to excite the Cu^2+ ion to the upper level (^2E_g) of the 910nm fluorescence.
The wavelengths of the three lasers are shown schematically in the figure.
The 'rough' structure near the peak of the 633nm laser-excited fluorescence peak (890-950nm) is real and may be due to diffraction effects from the powdered pigment surface. This effect can also be seen in the (green curve) spectrum from the direct 532nm laser but not in the (turquoise curve) diffused laser spectrum which is very smooth.
The temperature dependence (below 300K) of the peak fluorescence wavelength has beed addressed by Yixi Zhuang and Setsuhisa Tanabe, "Forward and back energy transfer between Cu21 and Yb31 in Ca12xCuSi4O10:Ybx crystals", J. Appl. Phys. 112, 093521, 2012. This indicates that, around room temperature, there is a redwards shift of approximately 2nm per 10°C increase. This implies that my green laser was heating the powdered sample to about 100°C.