Abstract

A detector system incorporating the Reticon RL1024S photodiode array has been constructed at the National Oceanic and Atmospheric Administration Aeronomy Laboratory as part of a double spectrograph to be used to study the Earth’s atmosphere from ground-based and aircraft-based platforms. To determine accurately the abundances of atmospheric trace gases, this new system must be able to measure spectral absorptions as small as 0.02%. The detector, manufactured by EG&G Reticon, exhibits superior signal-to-noise characteristics at the light levels characteristic of scattered skylights, but interference in the passivating layer (a thin layer of SiO2 that is deposited during the manufacture to protect the silicon active area from water vapor) causes major problems in achieving the required precision. The mechanism of the problems and the solution we have implemented are described in detail.

© 1992 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
    [CrossRef]
  2. A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).
  3. Manufactured by MMR Technologies, Mountain View, Calif.
  4. EG&G Princeton Applied Research detector system Model 1412.
  5. See, for example, M. Born, E. Wolf, Principles of Optics (Macmillan, New York, 1964), Chap. 7.2.

1989 (1)

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

1987 (1)

G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
[CrossRef]

Born, M.

See, for example, M. Born, E. Wolf, Principles of Optics (Macmillan, New York, 1964), Chap. 7.2.

Jakoubek, R.

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

Mount, G.

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
[CrossRef]

Ravishankara, A.

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

Sanders, R.

G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
[CrossRef]

Schmeltekopf, A.

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
[CrossRef]

Solomon, S.

G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
[CrossRef]

Wahner, A.

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

Wolf, E.

See, for example, M. Born, E. Wolf, Principles of Optics (Macmillan, New York, 1964), Chap. 7.2.

J. Geophys. Res. (2)

G. Mount, R. Sanders, A. Schmeltekopf, S. Solomon, “Visible spectroscopy at McMurdo Station, Antarctica. 1, Overview and daily variations of NO2 and O3, austral spring, 1986,” J. Geophys. Res. 92, 8320–8326 (1987).
[CrossRef]

A. Wahner, R. Jakoubek, G. Mount, A. Ravishankara, A. Schmeltekopf, “Remote sensing observations of daytime column NO2 during the airborne Antarctic ozone experiment, August 22–October 2, 1987,” J. Geophys. Res. 94, 16,619–16,629 (1989).

Other (3)

Manufactured by MMR Technologies, Mountain View, Calif.

EG&G Princeton Applied Research detector system Model 1412.

See, for example, M. Born, E. Wolf, Principles of Optics (Macmillan, New York, 1964), Chap. 7.2.

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (15)

Fig. 1
Fig. 1

Spectrum of a tungsten lamp taken in the blue spectral region showing the large (~20% p.p.) interference from currently available Reticons (array 3): (a) unprocessed but arbitrarily normalized to 1.0 at pixel 512; (b) after fitting a quadratic to remove the lamp spectral shape and subtracting 1.0 to show the variation about a zero average.

Fig. 2
Fig. 2

Spectrum of a tungsten lamp observed with an array chip manufactured prior to 1981 (array 12) after fitting a quadratic and subtracting 1.0 from the normalized spectrum to remove the lamp spectral shape and show the variation about a zero average. The etaloning is much less apparent and nonsinusoidal. The p.p. amplitude modulation is 8%.

Fig. 3
Fig. 3

Spectrum of a tungsten filament lamp taken in the near UV with a commercial Princeton Applied Research Model 1412 EG&G detector system (a Reticon array with Princeton Applied Research readout electronics). The interference amplitude is ~20% p.p.

Fig. 4
Fig. 4

(a) Ratio of the lamp spectra taken 0.5 h apart with array 3. The p.p. amplitude modulation is ~0.03%. (b) The ratio of the lamp spectra taken 0.5 h apart with array 12. Despite the lack of clear sinusoidal character in the spectrum itself, there is a clear sinusoidal pattern in the ratio. The p.p. amplitude modulation is ~0.25%.

Fig. 5
Fig. 5

Normalized spectra taken 12 h apart with array 12. The phase of the interference has shifted by ~180° and indicates clearly that the interference pattern changes with time.

Fig. 6
Fig. 6

Ratio of a pair of lamp spectra taken 0.5 h apart and a second ratio taken 12 h later with array 12. The phase of the ratio changes in wavelength with time.

Fig. 7
Fig. 7

(a) Spectral ratios collected at intervals of 6.5 h with array 12 over a total time span of 100 h. Each spectrum is the average of several hundred readouts of the detector over a 0.5-h period with each ratioed to the first spectrum that is collected. The first ratio is shown by the heavy curve. A slow increase in amplitude is readily apparent. The amplitude reaches a maximum of 16% p.p. (when the modulations in the spectra that form the ratio are shifted in phase by 180°). (b) Same as (a) but after a thorough cleaning of the Dewar contents with methanol and eight days of heating at 65°C in a vacuum with a pressure of <10 × 10−6 Torr. The outgassing rate in the Dewar is diminished. Hence the rate of deposition of material onto the detector is decreased substantially

Fig. 8
Fig. 8

Time dependence of the amplitude of the successive ratios for the run shown in Fig. 7(b). The detector temperature is −84°C. The deposition rate slows considerably with time but levels off after ~70 h and remains constant with a p.p. amplitude of ~0.02%.

Fig. 9
Fig. 9

Time-dependent amplitude of the spectral ratios for array 12 at various operating temperatures.

Fig. 10
Fig. 10

Geometry of the etalon shift model. The transmitted and first internal rays interfere on the silicon side of the detector’s passivating layer (shaded area). The effective thickness of the layer increases in time as a result of the deposition of outgassed material (arrows).

Fig. 11
Fig. 11

Comparison of the model to data taken with the detector in the normal position: (a) the model of interference in a single spectrum; (b) the modeled time series of spectral ratios; (c) the time series of spectral ratios from real data. In both (b) and (c) the length of the time steps is 75 min.

Fig. 12
Fig. 12

Variation of wavelength λ, initial layer thickness x0, and intensity factor b to the diode number (a) for the normal detector position and (b) for the detector turned end for end.

Fig. 13
Fig. 13

Comparison of the model to data taken in the reversed detection position: (a) the modeled time series of spectral ratios, time step = 75 min; (b) the time series of spectral ratios from real data, time step = 15 min.

Fig. 14
Fig. 14

Schematic drawing of the desired window configuration to confine the outgassing material at atmospheric pressure in front of the detector active area.

Fig. 15
Fig. 15

Time-dependent amplitude of successive spectral ratios for array 14 with the window configuration of Fig. 14 but without the spacer and with the frit replaced by epoxy. The interference shift is much smaller than in a windowless device. The system eventually settles out at ~0.0005%, which is extremely difficult to measure.

Equations (10)

Equations on this page are rendered with MathJax. Learn more.

I S = I 1 + I 2 + 2 ( I 1 I 2 ) 1 / 2 cos δ,
δ = 2 π [ 2 n x ( t ) ] / λ ,
I = I S / I 1 = 1 + b + 2 b cos [ 4 π n x ( t ) / λ ] .
I ( Δ t ) I ( 0 ) + ( I / x ) x = x 0 ( d x / d t ) t = 0 Δ t ,
I ( Δ t ) / I ( 0 ) = 1 + ( I / x ) x = x 0 r Δ t / I ( 0 ) 1 + M ,
M = ( 8 π n r Δ t b / λ ) sin ( 4 π n x 0 / λ ) .
d A / d t = α A ,
A ( t ) = A 0 exp ( α t ) ,
D ( t ) = A ( 0 ) A ( t ) = A 0 [ 1 exp ( α t ) ] .
x ( t ) = x 0 + D ( t ) = x 0 + ( r / α ) [ 1 exp ( α t ) ] ,

Metrics