l size=-1 color=black> Below is a cache of http://nvl.nist.gov/pub/nistpubs/jres/103/1/j31tho.pdf. It's a snapshot of the page taken as our search engine crawled the Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
Ultraviolet Spectral Irradiance Scale Comparison: 210 nm to 300 nm
Volume 103, Number 1, JanuaryFebruary 1998
Journal of Research of the National Institute of Standards and Technology
[J. Res. Natl. Inst. Stand. Technol. 103, 1 (1998)]
Ultraviolet Spectral Irradiance Scale
Comparison: 210 nm to 300 nm
Volume 103
Number 1
JanuaryFebruary 1998
Ambler Thompson,
Edward A. Early, and
Thomas R. OBrian
National Institute of Standards
and Technology,
Gaithersburg, MD 20899-0001
Comparison of the irradiances from a num-
ber of ultraviolet spectral irradiance stan-
dards, based on different physical princi-
ples, showed agreement to within their
combined standard uncertainties as assigned
to them by NIST. The wavelength region
of the spectral irradiance comparison was
from 210 nm to 300 nm. The spectral ir-
radiance sources were: an electron storage
ring, 1000 W quartz-halogen lamps, deu-
terium arc lamps, and a windowless argon
miniarc.
Key words:
irradiance; spectral irradiance;
synchrotron radiation; ultraviolet radio-
metry; ultraviolet standards.
Accepted:
September 12, 1997
1.
Introduction
Many national programs and industrial applications
require accurate measurement of radiation in the ultravi-
olet (UV) spectral region. Examples of the important
applications of UV radiation and measurement technol-
ogy are: monitoring atmospheric ozone, UV curing of
polymers, water purification, and photolithography of
semiconductor materials. Artifacts or facilities for
transferring absolute UV irradiance are available from
the National Institute of Standards and Technology, and
are derived from different physical principles. This UV
measurement capability is maintained by three organi-
zational divisions within the Physics Laboratory at
NIST: the Optical Technology Division (OTD), the
Atomic Physics Division (APD), and the Electron and
Optical Physics Division (EOPD). The absolute detec-
tor-based scales are traceable to fundamental electrical
units using the OTDs High Accuracy Cryogenic
Radiometer [1]. The source-based scales are derived
from blackbody radiation in the OTDs Facility for Auto-
mated Spectroradiometric Calibration (FASCAL) [2,3],
from plasma sources (i.e., wall-stabilized hydrogen and
blackbody arcs) in the APD [4], and from electron
storage ring synchrotron radiation in the EOPDs
Synchrotron Ultraviolet Radiation Facility II (SURF II)
where spectral irradiance and radiance are calculable [5]
from the approximations of Schwinger [6]. SURF II is
currently undergoing a major upgrade which will lower
its overall uncertainties and incorporate the cryogenic
radiometer capabilities from the OTD [7].
An ultraviolet intercomparison was recently com-
pleted by the Comite´ Consultatif de Photometrie et Ra-
diometrie (CCPR) of the Comitte´ International des
Poids et Mesures (CIPM) between NIST, the National
Physical Laboratory (NPL) of the United Kingdom, and
the Physikalische Technische Bundesanstalt (PTB) of
Germany [8]. The intercomparison covered spectral
radiance and irradiance across the wavelength range
200 nm to 400 nm using deuterium and tungsten-
halogen lamps as transfer standards. This intercompari-
son produced anomalous results in which UV spectral
irradiance measurements performed by the different
laboratories disagreed with each other by 5 % to 10 %.
The NIST scales of UV spectral irradiance have been
compared previously: SURF II irradiance was compared
with tungsten-halogen standards [9] and with an argon
miniarc [10]. The spectral irradiance comparison of
SURF II with tungsten-halogen lamps was done at
two wavelengths (254 nm and 297 nm) and agreement
1 Volume 103, Number 1, JanuaryFebruary 1998
Journal of Research of the National Institute of Standards and Technology
between the spectral irradiances was on the order of
1 %, well within the relative combined standard uncer-
tainties. Comparison of SURF II to the argon miniarc at
214 nm found agreement to within 3 %, also within the
relative combined standard uncertainty of the compari-
son. Since the results from the latest CCPR intercom-
parison were at odds with previous measurements at
NIST, it seemed worth investigating whether the NIST
values of spectral irradiance from sources derived from
different physical principles were still in agreement.
Therefore the spectral irradiance sources at NIST were
compared to determine if unknown systematic effects in
the derivation or transfer of the NIST spectral irradiance
scales could contribute to the observed differences in
the CCPR intercomparison.
2.
Approach
The primary experimental considerations in compar-
ing the spectral irradiance values from NIST sources
were to: 1) carry out the irradiance comparison in air, 2)
design the transfer spectroradiometer to minimize
known differential source properties, 3) assume short
term stability of the spectroradiometer used to compare
source spectral irradiances, and 4) design the experi-
ment so as to minimize alignment errors and the need
for instrument repositioning. In this regard, the NIST
spectral irradiance sources are used to determine the
spectral responsivity of a spectroradiometer, in pre-
cisely the same manner as any user of standard sources.
The spectral responsivity of the instrument should be
invariant with the irradiance source, if the design of the
spectroradiometer has made it relatively insensitive to
geometric and radiometric differences in the standard
sources.
The basis for performing this type of evaluation is the
simplified measurement equation [11],
S (
0
) =
E ( )R (
0
, )d ,
(2.1)
where
is a wavelength in the range of a monochroma-
tor set at
0
, S (
0
) is the output signal of the photon
counting circuit, E ( ) is the spectral irradiance of the
source, and R (
0
,
) is the spectral irradiance respon-
sivity function of the instrument. Further, as shown in
Ref. [11] the spectral irradiance responsivity function is
given by
R (
0
,
) = R ( )z (
0
),
(2.2)
where R ( ) is the irradiance response function and
z (
0
) is the dimensionless slit-scattering function.
Combining Eqs. (2.1) and (2.2) yields
S (
0
) =
E ( )R ( )z (
0
)d ,
(2.3)
The spectral irradiance response function usually varies
slowly with wavelength, and indicates the sensitivity of
the instrument to light at a given wavelength. Con-
versely, the slit-scattering function is a rapidly varying
function of wavelength. Ideally it is independent of
0
and triangular in shape, and indicates the bandwidth
of the instrument. Given a source with a known spectral
irradiance E
s
( ), and assuming that E
s
( ) and R ( )
vary slowly over the wavelength range for which
z (
0
) is appreciable, Eq. (2.3) becomes
S (
0
) = E
s
(
0
)R (
0
)
z (
0
)d .
(2.4)
The product R (
0
)
z (
0
)d
is the spectral irradi-
ance responsivity, and reduces to R (
0
)d for a double
monochromator viewing standard sources, whose
spectral irradiance changes slowly relative to the instru-
ments bandpass. NIST sources with known spectral
irradiances were used to determine the spectral irradi-
ance responsivity of the spectroradiometer in this exper-
iment and the resultant spectral responsivities were
compared.
A number of NIST UV irradiance sources with differ-
ent physical bases for the assignment of spectral irradi-
ance were used in this comparison. The SURF II spec-
trometer calibration beamline, designated number 2 was
used as a spectral irradiance source based on synchro-
tron radiation and is hereafter referred to as SURF. Two
types of spectral irradiance standard lamps based on
blackbody radiation and calibrated in FASCAL were
used and are designated FEL and D
2
. The first type is a
modified 1000 W quartz-halogen FEL type lamp with
coiled-coil
filaments
calibrated
from
250 nm
to
2400 nm, while the second is a deuterium lamp cali-
brated from 200 nm to 350 nm. Two FEL lamps and
three D
2
lamps were used in the comparison. The minia-
ture argon arc source also known as an argon miniarc,
designated ArArc, was calibrated at NIST by the APD
just prior to the comparison. The comparison of the
ArArc, FEL and D
2
with SURF was carried out in air
since
the
ArArc,
FEL,
and
D
2
standards
are
calibrated in air and many UV applications require
in-air calibrations.
2 Volume 103, Number 1, JanuaryFebruary 1998
Journal of Research of the National Institute of Standards and Technology
A schematic diagram of the laboratory setup is shown
in Fig. 1. SURF was equipped with an ultra high vac-
uum (UHV) chamber section which contained a rotating
fused silica window, a limiting aperture, and a fused
quartz exit window between the UHV of the beamline
and the air atmosphere of the laboratory. Using only a
single, fixed exit window would lead to rapid and
nonuniform dar