PROCEEDINGS OF SPIE

SPIEDigitalLibrary.org/conference-proceedings-of-spie

Rapid, automated, quality control of

diffraction grating efficiency

Mark Fisher, Travis Burt

Mark R. Fisher, Travis C. Burt, "Rapid, automated, quality control of diffraction

grating efficiency," Proc. SPIE 10373, Applied Optical Metrology II, 103730O
(23 August 2017); doi: 10.1117/12.2275526

Event: SPIE Optical Engineering + Applications, 2017, San Diego, California,

United States

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 26 Sep 2022 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use

 Rapid, automated, quality control of diffraction grating efficiency
Mark R. Fishera, Travis C. Burtb,
a
Agilent Technologies, 2500 Regency Parkway, Cary, North Carolina, 27518, USA;
b
Agilent Technologies, 679 Springvale Rd, Mulgrave, Victoria, 3170, Australia.

ABSTRACT

A diffraction grating is found at the heart of every modern spectrophotometer and yet, despite being used for over 60
years, a practical and efficient characterization tool has proven to be elusive. Part of the challenge can be attributed to
the unique angular dependent geometry, or off axis dispersion, of gratings. Here we demonstrate automated grating
efficiency measurements of four reflection gratings (300, 1200, 1800 and 3600 grooves per mm). Total measurement
time was less than 2 hrs at a maximum of 161 wavelengths per grating. This approach can reduce test times or assist
expand quality assurance, or design verification, programs. Automated measurements are performed in hours
demonstrating efficiency and ease-of-use advantages when compared to equivalent manually operated systems.
Keywords: UV-Vis-NIR, Spectrophotometer, Diffraction, Grating, Cary, Reflectance, Transmittance

1. INTRODUCTION
Diffraction gratings are optical components that facilitate separation and selection of specific wavelengths of light from
broadband sources such as incandescent lights or sunlight. A diffraction grating is found at the heart of every modern
spectrometer and yet, despite being used extensively commercially for over 60 yrs, practical and efficient
characterization has proven to be elusive.1
Part of the challenge can be attributed to the unique angular dependent geometry, or off axis, dispersion
associated with gratings. Shorter wavelengths are diffracted at smaller angles while longer wavelength’s have larger
diffraction angles (Figure 1.0).

Reflective Diffraction Grating Transmissive Diffraction Grating

Figure 1. Schematic presentation of the beam pathways in a reflective and transmissive diffraction grating.

Applied Optical Metrology II, edited by Erik Novak, James D. Trolinger, Proc. of SPIE
Vol. 10373, 103730O © 2017 SPIE · CCC code: 0277-786X/17/$18 · doi: 10.1117/12.2275526

Proc. of SPIE Vol. 10373 103730O-1

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 26 Sep 2022
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
 Diffraction gratings define, to a major point, performance parameters of a spectrophotometer such as the spectral
resolution and the dynamic range through better stray light control. Therefore, a well-controlled consistency in the
quality and performance of produced gratings is of paramount importance for the spectrophotometer manufacturing
industry.
An important characteristic of a grating is its diffraction efficiency which is defined as the ratio of the intensity of the
desired diffracted light into the desired order compared to the intensity of the incident light (absolute efficiency) or
compared to a reference mirror (relative efficiency). The diffraction efficiency is determined by the grating constant
(number of grooves per milimeter), the shape of the grooves, and by the regularity of the grooves. Therefore, the
diffraction efficiency is also good attribute to assess the quality of a grating. Furthermore, for a given grating the
diffraction efficiency is also wavelength dependent and changes with the angle of incidence (AOI).1
Grating diffraction efficiency curves are usually measured by producers using custom designed, in-house, built systems
which are often inflexible (e.g. only one AOI or diffraction angle) and require frequent user intervention.2 Analysis on
such systems often need long measurement times in order to achieve useable data quality. Generally, efficiency curves
are collected for all new master gratings but manufacturers do not routinely characterize or perform comprehensive
QA/QC on replicated gratings, simply because there is no commercial instrumentation available that allows fast, efficient
and accurate characterization of diffraction gratings.
Multi-angle Photometric Spectroscopy (MPS) was used to measure the reflectance of gratings across a range of
wavelengths and angles. A recent development by Agilent Technologies, the Cary 7000 Universal Measurement
Spectrophotometer (UMS) combines both reflection and transmission measurements from the same patch of a sample’s
surface in a single automated platform. Angles of incidence range 5 deg ≤ |i| ≤ 85 deg (i.e. angles on either side of
beam normal noted as +/-) under s or p polarized light.
The Cary 7000 UMS is the latest generation of high-end UV-Vis/NIR spectrophotometers from Agilent capable of
performing automated absolute reflection and transmission measurement at almost any angle over a wavelength range of
250 nm to 2500 nm. Automated and independent control of the detector position and sample angle allows transmission
and/reflection data to be collected in a single sequence at a pre-set series of measurement geometries without touching
the sample. An automated polarizer ensures accurate measurements at S, P, or any user-specified polarization angle.
We describe the use of MPS on the Cary 7000 UMS by changing grating angle with respect to the incident beam (i),
from 20 deg to 80 deg and the fixed detector position (d), user defined. This optical geometry replicates a typical
monochromator design configuration where the entrance aperture represents the entrance slit of the monochromator and
the detector the exit slit. Both s and p polarization efficiency profiles were collected within the range 250 nm to 2500 nm
automatically without the need for user intervention. This spectroscopic approach provides useful input into to quality
assurance programs while simultaneously reducing test costs and enhancing part throughput.
The accessory component of the Cary 7000 UMS, the Cary Universal Measurement Accessory (UMA), is available
separately to upgrade existing UV-Vis-NIR spectrophotometers from the Cary family, including the 4000, 5000, and
6000i.

2. EXPERIMENTAL DETAILS
To demonstrate the benefits in speed and accuracy of the Cary 7000 UMS for the routine characterization of diffraction
gratings we determined the reflection spectra of four reflection gratings (300, 1200, 1800 and 3600 grooves per mm).
The efficiency of a grating might be dependent on the polarization of the incoming light beam, therefore, spectra were
collected for both s- and p-polarized light.

Proc. of SPIE Vol. 10373 103730O-2

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 26 Sep 2022
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
 Figure 2. Left: 300 lines per mm grating mounted on Cube Beam Splitter Sample Holder. Right: Sample Holder placed onto
the sample stage in the UMA. Detector held at 10°.

As it can be seen from Figure 2 the gratings were mounted on the Agilent Cube Beam Splitter Holder and placed into the
UMA. The reflection characteristics of the gratings were analyzed for a fixed reflection angle. Therefore, the detector
angle was kept fixed at an angle of 10°. The angle at which light is reflected from the gratings is dependent on the
wavelength of the incident light. To ensure that the reflected light hits the detector at 10°, the angle of incidence needs to
be adapted for each wavelength. An Applications Development Language (ADL) code (given in the appendix) was used
to drive the sample grating to the desired angle of incidence for each analytic wavelength and to set the polarization
angle. The collects were performed applying the parameters given in Table 1.

Table 1. Agilent Cary 7000 UMS collection conditions used to acquire the %R spectra of the four different gratings.
Parameter Value
Angle of incidence Wavelength-dependent
Detector angle 10°
Wavelength range 250 – 2500 nm
Data interval 10 and 1 nm, respectively
Spectral bandwidth* 2 nm
Signal averaging time 0.5 sec
Polarization s-and p-polarization
Incident beam geometry 3°x1° (vertical x horizontal)
Baseline correction On
* Selection of correct spectral bandwidth (SBW) is critical. Using too large an SBW setting leads to reduced %R values caused by the
“refracted light cone” missing the detector.

3. RESULTS
In Figure 3 the reflection spectra of the four different diffraction gratings are shown. The spectra were collected with a
data interval of 1 nm and 10 nm. The use of the larger data interval reduces the number of collected data points and the
overall time needed to acquire (Table 2). However, the reflection spectra collected with 10 nm step size show a similar
same level of accuracy as the spectra collected with 1 nm steps.

Proc. of SPIE Vol. 10373 103730O-3

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 26 Sep 2022
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
 300 lines/mm, 2 nm spectral bandwidth, 1 nm interval. 300 lines/mm, 2 nm spectral bandwidth, 10 nm interval.
WI o

GO

20 WO Imm mm

.1.4t rideraelm0
Grav Avenue S&P Poi...on Tr
1000 1m :000 mou 1540 2000
wawkrqm 11.1 Wavelength (nm)

1200 lines/mm, 2 nm spectral bandwidth, 1 nm interval. 1200 lines/mm, 2 nm spectral bandwidth, 10 nm interval.

00

p MN.=
`___
-____
=I ...
w

400
Wave
a--__ .nu
.
eoo 000
wawienptnlnml
1000 ma unn

-_
-\ -
1800 lines/mm, 2 nm spectral bandwidth, 1 nm interval. 1800 lines/mm, 2 nm spectral bandwidth, 10 nm interval.

,MI--FEEM
' - -ma
Sra
W lines

PPn hl
so Pal kw .

eu EMI -

xu

400 ano
Wavelength (nm)
'm wa wo iono
Wavelength Om)

3600 lines/mm, 2 nm spectral bandwidth, 1 nm interval. 3600 lines/mm, 2 nm spectral bandwidth, 10 nm interval.

700
3600 lines per mm
Red P- Polarizati
eo
°°P9rwanmmn
Red
Gray S- Polarization Green Awng94PPolarization
80 Blue Average S: ' Polarization
W

0
60 90 \
40

ID

250 300 350 400 450 500

350 300 990 CW 450 5W
Wavelength (nm) Wavelength (am)

Figure 3. Left: Data Interval 1 nm SBW 2 nm Right: Data Interval 10 nm SBW 2 nm. Spectra collected with 10 nm step size
show the same level of accuracy as the spectra collected with 1 nm steps.

Proc. of SPIE Vol. 10373 103730O-4

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 26 Sep 2022
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
 The tabulated results (Table 2) demonstrate that broad wavelength range grating characterization and quality can be
determined in a matter of minutes without any user intervention during the short, fully automated collect. Longer
collects can also be executed, for example in over-night runs, when a great depth of spectral finesse is needed at wide
spectral range, narrow spectral resolution and small data intervals. Likewise, extremely rapid short wavelength range
peak efficiency scans could executed at only the grating blaze wavelength in a fraction of these times - providing almost
instantaneous feedback on grating quality.

Table 2. Comparison of Elapsed Collection Time between Data Interval of 1 nm and 10 nm. A larger Data
Interval significantly reduces collection times without compromise to the data quality.
Data interval = 1 nm Data interval = 10 nm
Grating Lines per
mm Total Elapsed Number of Data Total Elapsed Time Number of Data Points
Time Points
300 5 hrs 29 min 15 1601 34 min 38 sec 161
sec
1200 4 hrs 21 min 30 1251 28 min 57 sec 126
sec
1800 2 hrs 52 min 50 826 18 min 34 sec 83
sec
3600 1 hr 41 sec 290 7 min 57 sec 29

4. CONCLUSIONS
Rapid, automated characterization of diffraction gratings was performed using a Cary 7000 UMS. The UMS design
allows for multi-angle photometric spectroscopy (MPS) measurements. The MPS approach gives the user the flexibility
to characterize grating efficiency, either at a given angle of incidence (via sample stage rotation), or at a given
reflection/transmission angle (via detector rotation). The measurement takes place automatically, with minimal user
interaction, for s and p polarized conditions and can be completed in less than 2 hrs.

REFERENCES

[1] Optical Grating Evaluator: A Device for Detailed Measurement of Diffraction Grating Efficiencies in the Vacuum
Ultraviolet, D. J. Michels, T. L. Mikes, and W. R. Hunter Applied Optics, Vol. 13, Issue 5, pp. 1223-1229 (1974)
doi: 10.1364/AO.13.001223. https://www.osapublishing.org/ao/viewmedia.cfm?uri=ao-13-5-1223.
[2] Hutley, M. Ch.: Spectroscopic Properties. In Diffraction Gratings, 1st Edn. Academic Press, (1982).

Proc. of SPIE Vol. 10373 103730O-5

Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 26 Sep 2022
Terms of Use: https://www.spiedigitallibrary.org/terms-of-use