Letter
Vol. 40, No. 14 / July 15 2015 / Optics Letters
3225
Broadband spectral transmittance measurements of complex thin-film filters with optical densities of up to 12 SIMONA LIUKAITYTE, MICHEL LEQUIME,* MYRIAM ZERRAD, THOMAS BEGOU,
AND
CLAUDE AMRA
Aix Marseille Université, CNRS, Centrale Marseille, Institut Fresnel, UMR 7249, 13013 Marseille, France *Corresponding author: [email protected] Received 30 April 2015; revised 11 June 2015; accepted 16 June 2015; posted 16 June 2015 (Doc. ID 240106); published 2 July 2015
A new transmittance measurement setup, based on the use of a tunable laser source and a low-noise scientific-grade CCD camera operating in perfect integration mode, is proposed to achieve the spectrally resolved characterization of thin-film filters with optical densities from 0 to 12 in a wavelength range between 400 and 1000 nm. The first experimental results obtained on dedicated components demonstrate the efficiency of this new measurement scheme. © 2015 Optical Society of America OCIS codes: (120.3930) Metrological instrumentation; (120.7000) Transmission; (310.6188) Spectral properties; (310.6860) Thin films, optical properties. http://dx.doi.org/10.1364/OL.40.003225
Optical interference filters are essential components in modern optical instrumentation, and their specifications are becoming increasingly demanding, especially in enabling applications such as fluorescence microscopy [1], Raman spectroscopy [2], astronomy [3], and observations of the earth from space [4,5]. Because of improvements in thin-film design software, the global automation of energetic deposition processes and the efficiency of in situ optical monitoring, it has now become possible to manufacture interference filtering functions such as bandpass, edge, and notch filters that are characterized by high levels of transmittance in their passbands (greater than 95%), high optical densities (ODs) in their blocking regions (OD > 8), and sharp spectral edges (>40 dB∕nm). Although the measurement of the spectral transmittances of such filters between 100% and 0.1% is rather easy to achieve using a standard spectrophotometer (for instance, the PerkinElmer Lambda 1050 or the Agilent Cary 7000), the determination of the blocking efficiencies of the rejection bands remains a considerable challenge [6–9] for OD values that are, in general, greater than 6 and even greater than 4 near steep edges. Dedicated methods have been proposed and implemented to overcome this limitation and extend the range of measurement up to an OD of 12; most of the time, however, these 0146-9592/15/143225-04$15/0$15.00 © 2015 Optical Society of America
methods provide data only at one or two laser lines [3,10– 12]. Among these various methods, the one based on heterodyne detection [12] is certainly the most attractive in terms of range because of the proportionality between the amplitude of the photocurrent beat signal and the square root of the filter transmittance. As emphasized by the authors, a heterodyne measurement of a change in optical transmittance of 12 orders of magnitude requires a measurement of an ac electrical signal change of only six orders of magnitude. Another advantage of this heterodyne technique over the conventional methods is its insensitivity to stray radiation, which allows measurements to be performed under full room light. To date, the experimental demonstration of this coherent scheme has been achieved only at a single wavelength [12] and, although the use of a tunable laser can, in principle, offer access to high optical density values over quite a large spectral range, the complexity of the experimental setup, which requires acousto-optic modulators to shift the laser frequency and highspeed detection devices to recover the beat signal, has prevented the widespread adoption of this method by the thin-film filters community. For this reason, we designed a simple and efficient setup to achieve the accurate (relative uncertainty of better than 1%) and spectrally resolved (typical resolution of approximately 2 nm) measurement of the transmittance of a filter over a wide range of optical densities (from 0 to 12) and wavelengths (between 400 and 1000 nm). A schematic representation of this experimental setup is given in Fig. 1. The broadband emission of a high-power supercontinuum laser source from NKT Photonics (Reference EXB-6) is launched into a volume hologram filtering device from Photon etc. (Reference LLTF VIS-2), which delivers a 1 mW free space single-mode beam with a central wavelength that is tunable between 400 and 1000 nm and a full width at half-maximum (FWHM) bandwidth of approximately 2 nm. This narrow-band and low-divergence beam passes through the thin-film filter (TFF) that is to be characterized, and is then focused by a silver coated off-axis aspheric mirror (MRC, measurement reflective collimator) into the 2a 50 μm diameter core of a step index all-silica fiber (MFOL, measure fiber-optic
3226
Vol. 40, No. 14 / July 15 2015 / Optics Letters
FOC
PIXIS 1024B
TFF MRC
LLTF
λ
MFOL
ODFW
BS NKT EXB-6
Letter
OSD
SH
ODF1 ODF2 ROD RRC
FEMTO OE-200-SI
RFOL
Fig. 1. Schematic representation of the high-OD spectral transmittance measurement setup (NKT EXB-6, supercontinuum laser source; LLTF, laser line tunable filter; OSD, order sorting device; SH, shutter; BS, beam splitter; ODF1, optical density flip 1; ODF2, optical density flip 2; ODFW, optical density filter wheel, TFF, thin-film filter; MRC, measurement reflective collimator; MFOL, measurement fiber-optic link; FOC, fiber-optic coupler; ROD, reference optical density; RRC, reference reflective collimator; RFOL, reference fiber-optic link).
link). The output end of this measurement fiber is imaged by a two-lens fiber-optic coupler (FOC, magnification γ F 2.3) on the 13 μm × 13 μm photodiode array of a Princeton Instruments PIXIS:1024B camera, a low-noise, back-illuminated, thermoelectrically cooled, 16 bit scientific-grade CCD imaging system. A non-polarizing beam splitter (BS) reflects a small portion (typically 10%) of the filtered incoming beam toward a reference channel, whose design is identical to that of the measurement channel: reference reflective collimator (RRC) + reference fiber-optic link (RFOL). The output end of this reference fiber is connected to the FC receptacle of a FEMTO OE-200-SI variable gain photoreceiver equipped with a silicon PIN detector of 1.2 mm in diameter. The conversion gain is remotely switchable from 1 × 103 to 1 × 1011 V∕W, and the output voltage is digitized by a 16 bit National Instruments USB6211 acquisition module. Three stages of reflective optical densities can be inserted between the beam splitter (BS) and the thin-film filter (TFF). The first two (ODF1 and ODF2, each with an optical density of approximately 3) are installed in two-position, highspeed 0°–90° flip mounts, whereas the last optical density filter wheel (ODFW) is mounted on a six-slot motorized fast change filter wheel (slot 1, no optical density; slot 2, OD 1; slot 3, OD 2; slot 4, OD 3; slots 5 and 6 not used). All these mechanical devices are operated remotely, which permits us to modify, at any time, the value of the optical density present in the measurement arm between zero and nine in steps of one. Moreover, an order sorting device (OSD) is placed in front of the beam splitter to remove the small parasitic peaks corresponding to the harmonics of the central optical wavelength selected by the tunable volume hologram. It consisted of two edge filters (short pass FESH0700 and long pass FELH0600, both from Thorlabs) mounted on a bistable device and whose position is switched following the value of this
central wavelength (short pass for wavelengths lower than 650 nm and long pass for wavelengths greater than 650 nm). Finally, the setup is completed by an electromechanical shutter (SH) whose purpose will be defined hereafter. To achieve the measurement of the transmittance T F λ of a thin-film filter over a very large range of optical densities, we chose to replace the direct comparison between the powers of the incident and transmitted light, which still encounters difficulties in terms of signal-to-noise ratio (SNR) and linearity, with the locking of the signal provided by the CCD camera to a high, constant level equal to 70% 10% of its full well capacity (FWC). This is achieved through the combined adjustment of two parameters, i.e., • the opening duration τλ of the electromechanical shutter (in the closed position by default) and • the effective value ODλ of the optical density present in the measurement arm. The result of a transmission measurement is thus determined by three parameters, i.e., the effective signal recorded by the camera, the opening duration, and the OD value. The smallest possible value for τλ is defined by the timing specifications of the shutter (40 ms), whereas the largest one is determined by the dark current and the influence of residual stray light at the CCD level. The latter is reduced spontaneously by the narrow angular field of view of the fiber-collimator assembly and by a careful canceling of all possible sources of parasitic light, whereas the former is intrinsically so low (92.5%) in its passband (central wavelength of 782.5 nm, bandwidth of 15 nm) and a high blocking efficiency on either side of this design wavelength (OD 10.6 at 707 nm, OD 9 at 864 nm). This filter was manufactured using the Leybold Optics HELIOS deposition machine of the Institut Fresnel with Nb2 O5 as the high index material and SiO2 as the low index material (number of layers: approximately 100). The theoretical spectral transmittance of this filter is shown in Fig. 3 (light blue line). The experimentally measured results obtained using our setup are shown on the same graph (red circles), as are the results of modeling this measurement (dark blue line) following the approach detailed above. The
Fig. 2. Normalized spectral density of the light power delivered by the tunable filter at λ0 748 nm (red circles, experimental data; blue line, modeling function; logarithmic units).
agreement between the experimental results and the modeled data is quite impressive, except at approximately 700 nm (where the measured OD is larger than the predicted value) and above 920 nm (where the long pass edge is shifted slightly toward shorter wavelengths). These small discrepancies can be explained reasonably as being the results of minor deposition errors occurring during the manufacturing of the filter. Moreover, if we replace in our model the experimental line profile shown in Fig. 2 with a theoretical profile corresponding to two LLTFs used in a series, or a Gaussian profile with a 2 nm FWHM bandwidth, then the modeled curve becomes identical to the theoretical one. All these results show that our setup functions as expected and that the observed limitation on the lowest recordable optical density is fundamentally because of a spectral crosstalk phenomenon created by the profile of the filtered line. To confirm this conclusion, we decided to measure the spectral transmittance of a “new” filter prepared by combining the previous thin-film filter with a neutral density of approximately 4 OD. Figure 4 shows the result of this additional measurement, again performed using a 1 nm pitch (dark blue circles), Design
Modeling
Experimental measurements
0 -1
Thin-Film Filter Opcal Density
number of detected photoelectrons is reduced by a factor of 350. Therefore, our measurement range effectively exceeds 12 OD. To cancel the effect of the fluctuations and drifts of the source power, we digitize, at a high rate (100 kilo-samples per second) and in 16 bits, the signal delivered by the silicon photodiode used in the reference arm, by beginning the recording 20 ms before the opening of the shutter and ending it 20 ms after its closure. This allows us to estimate the contribution of the dark current, to automatically subtract it and to perform a sort of time integration (similar to that spontaneously performed by the CCD) by summing all the processed data. Again, the measurement noise is dominated entirely by the quantum noise, and the relative uncertainty is reduced by summing the high number of acquired data (at least 4000). This transmittance measurement scheme can be modeled easily in software using a MATLAB program that accounts for the spectral dependence of the optical properties of all components of the setup. As an example, when the volume hologram filter is tuned to the wavelength λ0 , the measurement signal is given by
3227
-2 -3 -4 -5 -6 -7 -8 -9 -10 -11 400
500
600
700
800
900
1000
Wavelength (nm)
Fig. 3. OD spectral dependence of the thin-film filter used to verify the setup (light blue line, design data; dark blue line, modeled data; red circles, experimental data).
3228
Vol. 40, No. 14 / July 15 2015 / Optics Letters
OD4
TFF
OD4 + TFF
Letter
OD4 + TFF Measurements
be improved to permit an accurate determination of the optical density profile of a filter close to a band edge. To overcome these limitations, an evolution of our setup is now under study and should include a second filtering stage to increase the spectral purity of the filtered line: as underlined above, two volume hologram filters used in series would be sufficient to fulfill most of our metrological objectives, but this solution raises some problems of cost and implementation. Therefore, we choose to test in the near future two alternative solutions, based on the use of either a tunable Fabry–Perot or a dedicated monochromator.
0 -1 -2
Opcal Density
-3 -4 -5 -6 -7 -8 -9 -10 -11 -12 400
500
600
700
800
900
1000
Wavelength (nm)
Fig. 4. Spectral transmittance measurements of a filter formed by combining the original thin-film filter used to verify the setup and an OD 4 component (black line, OD 4; red line, verification TFF; light blue line, computed curve; dark blue circles, measured data).
as well as the result of the summing of the optical densities that were measured separately for each component (light blue line). The perfect agreement between these two curves, which were obtained in two different ways (through direct measurement and through the combination of two independent results) confirms our hypothesis regarding the effect of the spectral profile of the filtered line and demonstrates that our setup is, in fact, capable of measuring the spectral dependence of optical densities in the range of 0–12. In this Letter, we proposed a new method of transmission measurement using a perfect integration scheme and a method of passive synchronization between the measurement and reference channels, and we demonstrated the ability of a first setup based on this operating principle to measure accurately optical densities in a range from 0 to 12. Currently, a spectral crosstalk phenomenon limits the lowest measurable OD value to 8 in the case of a rapid spectral change in the transmittance of a complex thin-film filter. Moreover, the moderate spectral resolution of our setup (about 2 nm) has to
Funding.
Direction Générale de l'Armement (DGA).
REFERENCES 1. J. W. Lichtman and J. A. Conchello, Nat. Methods 2, 910 (2005). 2. T. Erdogan and V. Mizrahi, Spectroscopy 19, 113 (2004). 3. C. Araujo-Hauck, S. Fischer, H. Bartko, S. Gillessen, C. Straubmeier, M. Wiest, S. Yazici, F. Eisenhauer, G. S. Perrin, W. Brandner, K. Perraut, A. Amorim, and A. Eckart, Proc. SPIE 7734, 77342Z (2010). 4. A. F. H. Goetz, G. Vane, J. E. Solomon, and B. N. Rock, Science 228, 1147 (1985). 5. R. Le Goff, H. Krol, C. Grèzes-Besset, M. Lequime, T. Begou, C. Hecquet, M. Zerrad, B. Badoil, G. Montay, and K. Gasc, “Multispectral filters assemblies for Earth remote sensing imagers,” in International Conference on Space Optics, Tenerife, Canary Islands, Spain, October 7–10, 2014. 6. M. Ziter, G. Carver, S. Locknar, T. Upton, and B. Johnson, Surf. Coat. Tech. 241, 54 (2014). 7. Z. M. Zhang, L. M. Hanssen, and R. U. Datla, Opt. Lett. 20, 1077 (1995). 8. Agilent Technologies, “Measuring optical densities over 10 Abs on the Agilent Cary 7000 Universal Measurement Spectrophotometer,” Publication 5991-2528EN (2013). 9. P. Prabhat and T. Erdogan, Semrock Technical Note Series, http:// www.semrock.com/measurement‑of‑optical‑filter‑spectra.aspx. 10. Z. M. Zhang, T. R. Gentile, A. L. Migdall, and R. U. Datla, Appl. Opt. 36, 8889 (1997). 11. S. G. Kaplan, L. M. Hanssen, A. L. Migdall, and G. Lefever-Button, Proc. SPIE 3425, 56 (1998). 12. A. L. Migdall, B. Roop, Y. C. Zheng, J. E. Hardis, and G. J. Xia, Appl. Opt. 29, 5136 (1990).
Vol. 40, No. 14 / July 15 2015 / Optics Letters
3225
Broadband spectral transmittance measurements of complex thin-film filters with optical densities of up to 12 SIMONA LIUKAITYTE, MICHEL LEQUIME,* MYRIAM ZERRAD, THOMAS BEGOU,
AND
CLAUDE AMRA
Aix Marseille Université, CNRS, Centrale Marseille, Institut Fresnel, UMR 7249, 13013 Marseille, France *Corresponding author: [email protected] Received 30 April 2015; revised 11 June 2015; accepted 16 June 2015; posted 16 June 2015 (Doc. ID 240106); published 2 July 2015
A new transmittance measurement setup, based on the use of a tunable laser source and a low-noise scientific-grade CCD camera operating in perfect integration mode, is proposed to achieve the spectrally resolved characterization of thin-film filters with optical densities from 0 to 12 in a wavelength range between 400 and 1000 nm. The first experimental results obtained on dedicated components demonstrate the efficiency of this new measurement scheme. © 2015 Optical Society of America OCIS codes: (120.3930) Metrological instrumentation; (120.7000) Transmission; (310.6188) Spectral properties; (310.6860) Thin films, optical properties. http://dx.doi.org/10.1364/OL.40.003225
Optical interference filters are essential components in modern optical instrumentation, and their specifications are becoming increasingly demanding, especially in enabling applications such as fluorescence microscopy [1], Raman spectroscopy [2], astronomy [3], and observations of the earth from space [4,5]. Because of improvements in thin-film design software, the global automation of energetic deposition processes and the efficiency of in situ optical monitoring, it has now become possible to manufacture interference filtering functions such as bandpass, edge, and notch filters that are characterized by high levels of transmittance in their passbands (greater than 95%), high optical densities (ODs) in their blocking regions (OD > 8), and sharp spectral edges (>40 dB∕nm). Although the measurement of the spectral transmittances of such filters between 100% and 0.1% is rather easy to achieve using a standard spectrophotometer (for instance, the PerkinElmer Lambda 1050 or the Agilent Cary 7000), the determination of the blocking efficiencies of the rejection bands remains a considerable challenge [6–9] for OD values that are, in general, greater than 6 and even greater than 4 near steep edges. Dedicated methods have been proposed and implemented to overcome this limitation and extend the range of measurement up to an OD of 12; most of the time, however, these 0146-9592/15/143225-04$15/0$15.00 © 2015 Optical Society of America
methods provide data only at one or two laser lines [3,10– 12]. Among these various methods, the one based on heterodyne detection [12] is certainly the most attractive in terms of range because of the proportionality between the amplitude of the photocurrent beat signal and the square root of the filter transmittance. As emphasized by the authors, a heterodyne measurement of a change in optical transmittance of 12 orders of magnitude requires a measurement of an ac electrical signal change of only six orders of magnitude. Another advantage of this heterodyne technique over the conventional methods is its insensitivity to stray radiation, which allows measurements to be performed under full room light. To date, the experimental demonstration of this coherent scheme has been achieved only at a single wavelength [12] and, although the use of a tunable laser can, in principle, offer access to high optical density values over quite a large spectral range, the complexity of the experimental setup, which requires acousto-optic modulators to shift the laser frequency and highspeed detection devices to recover the beat signal, has prevented the widespread adoption of this method by the thin-film filters community. For this reason, we designed a simple and efficient setup to achieve the accurate (relative uncertainty of better than 1%) and spectrally resolved (typical resolution of approximately 2 nm) measurement of the transmittance of a filter over a wide range of optical densities (from 0 to 12) and wavelengths (between 400 and 1000 nm). A schematic representation of this experimental setup is given in Fig. 1. The broadband emission of a high-power supercontinuum laser source from NKT Photonics (Reference EXB-6) is launched into a volume hologram filtering device from Photon etc. (Reference LLTF VIS-2), which delivers a 1 mW free space single-mode beam with a central wavelength that is tunable between 400 and 1000 nm and a full width at half-maximum (FWHM) bandwidth of approximately 2 nm. This narrow-band and low-divergence beam passes through the thin-film filter (TFF) that is to be characterized, and is then focused by a silver coated off-axis aspheric mirror (MRC, measurement reflective collimator) into the 2a 50 μm diameter core of a step index all-silica fiber (MFOL, measure fiber-optic
3226
Vol. 40, No. 14 / July 15 2015 / Optics Letters
FOC
PIXIS 1024B
TFF MRC
LLTF
λ
MFOL
ODFW
BS NKT EXB-6
Letter
OSD
SH
ODF1 ODF2 ROD RRC
FEMTO OE-200-SI
RFOL
Fig. 1. Schematic representation of the high-OD spectral transmittance measurement setup (NKT EXB-6, supercontinuum laser source; LLTF, laser line tunable filter; OSD, order sorting device; SH, shutter; BS, beam splitter; ODF1, optical density flip 1; ODF2, optical density flip 2; ODFW, optical density filter wheel, TFF, thin-film filter; MRC, measurement reflective collimator; MFOL, measurement fiber-optic link; FOC, fiber-optic coupler; ROD, reference optical density; RRC, reference reflective collimator; RFOL, reference fiber-optic link).
link). The output end of this measurement fiber is imaged by a two-lens fiber-optic coupler (FOC, magnification γ F 2.3) on the 13 μm × 13 μm photodiode array of a Princeton Instruments PIXIS:1024B camera, a low-noise, back-illuminated, thermoelectrically cooled, 16 bit scientific-grade CCD imaging system. A non-polarizing beam splitter (BS) reflects a small portion (typically 10%) of the filtered incoming beam toward a reference channel, whose design is identical to that of the measurement channel: reference reflective collimator (RRC) + reference fiber-optic link (RFOL). The output end of this reference fiber is connected to the FC receptacle of a FEMTO OE-200-SI variable gain photoreceiver equipped with a silicon PIN detector of 1.2 mm in diameter. The conversion gain is remotely switchable from 1 × 103 to 1 × 1011 V∕W, and the output voltage is digitized by a 16 bit National Instruments USB6211 acquisition module. Three stages of reflective optical densities can be inserted between the beam splitter (BS) and the thin-film filter (TFF). The first two (ODF1 and ODF2, each with an optical density of approximately 3) are installed in two-position, highspeed 0°–90° flip mounts, whereas the last optical density filter wheel (ODFW) is mounted on a six-slot motorized fast change filter wheel (slot 1, no optical density; slot 2, OD 1; slot 3, OD 2; slot 4, OD 3; slots 5 and 6 not used). All these mechanical devices are operated remotely, which permits us to modify, at any time, the value of the optical density present in the measurement arm between zero and nine in steps of one. Moreover, an order sorting device (OSD) is placed in front of the beam splitter to remove the small parasitic peaks corresponding to the harmonics of the central optical wavelength selected by the tunable volume hologram. It consisted of two edge filters (short pass FESH0700 and long pass FELH0600, both from Thorlabs) mounted on a bistable device and whose position is switched following the value of this
central wavelength (short pass for wavelengths lower than 650 nm and long pass for wavelengths greater than 650 nm). Finally, the setup is completed by an electromechanical shutter (SH) whose purpose will be defined hereafter. To achieve the measurement of the transmittance T F λ of a thin-film filter over a very large range of optical densities, we chose to replace the direct comparison between the powers of the incident and transmitted light, which still encounters difficulties in terms of signal-to-noise ratio (SNR) and linearity, with the locking of the signal provided by the CCD camera to a high, constant level equal to 70% 10% of its full well capacity (FWC). This is achieved through the combined adjustment of two parameters, i.e., • the opening duration τλ of the electromechanical shutter (in the closed position by default) and • the effective value ODλ of the optical density present in the measurement arm. The result of a transmission measurement is thus determined by three parameters, i.e., the effective signal recorded by the camera, the opening duration, and the OD value. The smallest possible value for τλ is defined by the timing specifications of the shutter (40 ms), whereas the largest one is determined by the dark current and the influence of residual stray light at the CCD level. The latter is reduced spontaneously by the narrow angular field of view of the fiber-collimator assembly and by a careful canceling of all possible sources of parasitic light, whereas the former is intrinsically so low (92.5%) in its passband (central wavelength of 782.5 nm, bandwidth of 15 nm) and a high blocking efficiency on either side of this design wavelength (OD 10.6 at 707 nm, OD 9 at 864 nm). This filter was manufactured using the Leybold Optics HELIOS deposition machine of the Institut Fresnel with Nb2 O5 as the high index material and SiO2 as the low index material (number of layers: approximately 100). The theoretical spectral transmittance of this filter is shown in Fig. 3 (light blue line). The experimentally measured results obtained using our setup are shown on the same graph (red circles), as are the results of modeling this measurement (dark blue line) following the approach detailed above. The
Fig. 2. Normalized spectral density of the light power delivered by the tunable filter at λ0 748 nm (red circles, experimental data; blue line, modeling function; logarithmic units).
agreement between the experimental results and the modeled data is quite impressive, except at approximately 700 nm (where the measured OD is larger than the predicted value) and above 920 nm (where the long pass edge is shifted slightly toward shorter wavelengths). These small discrepancies can be explained reasonably as being the results of minor deposition errors occurring during the manufacturing of the filter. Moreover, if we replace in our model the experimental line profile shown in Fig. 2 with a theoretical profile corresponding to two LLTFs used in a series, or a Gaussian profile with a 2 nm FWHM bandwidth, then the modeled curve becomes identical to the theoretical one. All these results show that our setup functions as expected and that the observed limitation on the lowest recordable optical density is fundamentally because of a spectral crosstalk phenomenon created by the profile of the filtered line. To confirm this conclusion, we decided to measure the spectral transmittance of a “new” filter prepared by combining the previous thin-film filter with a neutral density of approximately 4 OD. Figure 4 shows the result of this additional measurement, again performed using a 1 nm pitch (dark blue circles), Design
Modeling
Experimental measurements
0 -1
Thin-Film Filter Opcal Density
number of detected photoelectrons is reduced by a factor of 350. Therefore, our measurement range effectively exceeds 12 OD. To cancel the effect of the fluctuations and drifts of the source power, we digitize, at a high rate (100 kilo-samples per second) and in 16 bits, the signal delivered by the silicon photodiode used in the reference arm, by beginning the recording 20 ms before the opening of the shutter and ending it 20 ms after its closure. This allows us to estimate the contribution of the dark current, to automatically subtract it and to perform a sort of time integration (similar to that spontaneously performed by the CCD) by summing all the processed data. Again, the measurement noise is dominated entirely by the quantum noise, and the relative uncertainty is reduced by summing the high number of acquired data (at least 4000). This transmittance measurement scheme can be modeled easily in software using a MATLAB program that accounts for the spectral dependence of the optical properties of all components of the setup. As an example, when the volume hologram filter is tuned to the wavelength λ0 , the measurement signal is given by
3227
-2 -3 -4 -5 -6 -7 -8 -9 -10 -11 400
500
600
700
800
900
1000
Wavelength (nm)
Fig. 3. OD spectral dependence of the thin-film filter used to verify the setup (light blue line, design data; dark blue line, modeled data; red circles, experimental data).
3228
Vol. 40, No. 14 / July 15 2015 / Optics Letters
OD4
TFF
OD4 + TFF
Letter
OD4 + TFF Measurements
be improved to permit an accurate determination of the optical density profile of a filter close to a band edge. To overcome these limitations, an evolution of our setup is now under study and should include a second filtering stage to increase the spectral purity of the filtered line: as underlined above, two volume hologram filters used in series would be sufficient to fulfill most of our metrological objectives, but this solution raises some problems of cost and implementation. Therefore, we choose to test in the near future two alternative solutions, based on the use of either a tunable Fabry–Perot or a dedicated monochromator.
0 -1 -2
Opcal Density
-3 -4 -5 -6 -7 -8 -9 -10 -11 -12 400
500
600
700
800
900
1000
Wavelength (nm)
Fig. 4. Spectral transmittance measurements of a filter formed by combining the original thin-film filter used to verify the setup and an OD 4 component (black line, OD 4; red line, verification TFF; light blue line, computed curve; dark blue circles, measured data).
as well as the result of the summing of the optical densities that were measured separately for each component (light blue line). The perfect agreement between these two curves, which were obtained in two different ways (through direct measurement and through the combination of two independent results) confirms our hypothesis regarding the effect of the spectral profile of the filtered line and demonstrates that our setup is, in fact, capable of measuring the spectral dependence of optical densities in the range of 0–12. In this Letter, we proposed a new method of transmission measurement using a perfect integration scheme and a method of passive synchronization between the measurement and reference channels, and we demonstrated the ability of a first setup based on this operating principle to measure accurately optical densities in a range from 0 to 12. Currently, a spectral crosstalk phenomenon limits the lowest measurable OD value to 8 in the case of a rapid spectral change in the transmittance of a complex thin-film filter. Moreover, the moderate spectral resolution of our setup (about 2 nm) has to
Funding.
Direction Générale de l'Armement (DGA).
REFERENCES 1. J. W. Lichtman and J. A. Conchello, Nat. Methods 2, 910 (2005). 2. T. Erdogan and V. Mizrahi, Spectroscopy 19, 113 (2004). 3. C. Araujo-Hauck, S. Fischer, H. Bartko, S. Gillessen, C. Straubmeier, M. Wiest, S. Yazici, F. Eisenhauer, G. S. Perrin, W. Brandner, K. Perraut, A. Amorim, and A. Eckart, Proc. SPIE 7734, 77342Z (2010). 4. A. F. H. Goetz, G. Vane, J. E. Solomon, and B. N. Rock, Science 228, 1147 (1985). 5. R. Le Goff, H. Krol, C. Grèzes-Besset, M. Lequime, T. Begou, C. Hecquet, M. Zerrad, B. Badoil, G. Montay, and K. Gasc, “Multispectral filters assemblies for Earth remote sensing imagers,” in International Conference on Space Optics, Tenerife, Canary Islands, Spain, October 7–10, 2014. 6. M. Ziter, G. Carver, S. Locknar, T. Upton, and B. Johnson, Surf. Coat. Tech. 241, 54 (2014). 7. Z. M. Zhang, L. M. Hanssen, and R. U. Datla, Opt. Lett. 20, 1077 (1995). 8. Agilent Technologies, “Measuring optical densities over 10 Abs on the Agilent Cary 7000 Universal Measurement Spectrophotometer,” Publication 5991-2528EN (2013). 9. P. Prabhat and T. Erdogan, Semrock Technical Note Series, http:// www.semrock.com/measurement‑of‑optical‑filter‑spectra.aspx. 10. Z. M. Zhang, T. R. Gentile, A. L. Migdall, and R. U. Datla, Appl. Opt. 36, 8889 (1997). 11. S. G. Kaplan, L. M. Hanssen, A. L. Migdall, and G. Lefever-Button, Proc. SPIE 3425, 56 (1998). 12. A. L. Migdall, B. Roop, Y. C. Zheng, J. E. Hardis, and G. J. Xia, Appl. Opt. 29, 5136 (1990).
