Index of Refraction of Fused Quartz Glass for Ultraviolet, Visible, and ...

- -
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA
VOLUME 44, NUMBER 9 SEPTEMBER, 1954
Index of Refraction of Fused Quartz Glass for Ultraviolet, Visible,
and Infrared Wavelengths*t
WILLIAM S. RODNEY AND ROBERT J. SPINDLER
Optics Division, National Bureau of Standards, Washington, D. C.
(Received June 3, 1954)
The index of refraction of fused quartz glass was determined for 24 wavelengths from 0.34669 to 3.5078
microns, using the minimum deviation method. The whole range is covered with a single instrument. The
variation in index between samples was determined and no definite variations in dispersion were observed.
Possible relations between purity and index are discussed.
1. INTRODUCTION
THE physical and optical properties of fused
T quartz glass make its increased use in the optical
industry highly desirable whenever it can be readily
produced in optical quality. In composition it is
definitely reproducible as compared with optical
glasses, it is extremely stable, and its coefficient of
thermal expansion is remarkably low. Its transmission
limits, both in the ultraviolet and the infrared, are
nearly the same as for the crystalline material and in
addition it is free from double refraction.
The refractive index of fused quartz glass has been
investigated in the ultraviolet and visible regions by
Gifford and Shenstone and by Trommsdorff. These
data, together with some indices by other observers,
are referenced, compared, and analysed by Sosman. 1
For the infrared region some indices of fused quartz
have been computed from reflection data but experimentally
determined data seem limited to 3-decimalplace
values. 2 Very little precise and self-consistent
information has previously been obtained by use of
identical procedures for different wavelength regions
and for different specimens.
2. SAMPLES
Seven prisms of fused quartz glass were used in this
investigation. One was made over twenty years ago
from fused quartz glass produced by the General
Electric Company. Its indices in the visible region and
its degree of homogeneity were carefully investigated'
and it is used as a refractive-index standard for the
precise calibration of refractometers. A second prism
was made of domestic fused quartz glass of recent
origin, also produced by the General Electric Company.
Two other prisms A and B were made from the product
of the Heraeus Company in Germany and obtained
by Dr. Stanley S. Ballard who made them available
* The work described in this paper was carried out in part
under the sponsorship of the U. S. Air Force.
t The material contained in this paper was orally presented at
the fall meeting of the Optical Society of America, October 1952.
1 Robert B. Sosman, The Properties of Silica (Chemical Catalogue
Company, Inc., New York, 1927).
2 C. Mueller and H. Wittenhouse, Z. Physik 85, 559 (1933).
3 L. W. Tilton and A. Q. Tool. J. Research Natl. Bur. Standards
3, 619 (1929).
for this investigation. A fifth prism was made by
grinding and polishing windows on the periphery of a
6-inch disk recently manufactured by the Nieder
Fused Quartz Company, Babson Park, Massachusetts.
Another prism was made by placing windows on the
periphery of a cylinder of Heraeus fused quartz (sample
C) approximately four inches long and two inches in
diameter. The seventh sample was made available by
the Corning Glass Works, it is known to have extremely
good ultraviolet transmission characteristics but has a
stronger absorption band around 2.6 microns than the
other samples. All seven specimens were of near
optical quality and free of noticeable striae and bubbles,
except the disk which has considerable striae. Even in
this case, however, precise data were obtainable.
3. INSTRUMENTS
The Watts precision spectrometer was used for the
visible region and the Gaertner precision spectrometer
for the ultraviolet and infrared. Descriptions of these
instruments, with photographs, are given in previous
publications by the authors. 4 ' 5 On both instruments
the minimum-deviation method, with its desirable
features of high accuracy and simplicity of index
computations, was used.
The spectra used in these measurements were the
mercury and cadmium emissions for wavelengths out to
677
approximately two and one-half microns and the
absorption bands of polystyrene beyond this region to
approximately three and one-half microns.
A more detailed discussion of the instruments and
procedure will appear in the Journal of Research of the
National Bureau of Standards.
4. DATA
On the Heraeus prisms A and B, index determinations
were made at or very near 240 and 31C. The average
temperature of the room was controlled within 4h0.2 0 C
and was determined frequently by means of a thermometer
having its bulb near the prism. Temperature
coefficients were then computed and small corrections
4 Wm. S. Rodney and R. J. Spindler, J. Research Natl. Bur.
Standards 49, 253 (1952).
6 Wm. S. Rodney and R. J. Spindler, J. Opt. Soc. Am. 42, 431
(1952).

678
W. S. RODNEY AND R. J. SPINDLER
Vol. 44
TABLE I. Computed indices of refraction of fused quartz glass (Heraeus B) and residuals (o-c)X 105 for temperature of 24 0 C.
Wavelength Computed Heraeus Heraeus Heraeus Nieder G.E. G.E. Corning
Source (microns) index Eq. 1 sample A sample B sample C disk sample sample 2 silica glass
A-c B-c C-c N-c Gl-c G2-c S-c
cd 0.34669 1.47757 -1 +22 -5
Hg 0.36117 1.47522 +6 0 +6 +31 +23 -6
do 0.365015 1.47465 +5 -3 +6 +26 +21 -9
do 0.404656 1.46971 +6 -2 +21 +8 +31 +21 -6
do 0.435835 1.46677 +7 0 +23 +6 +33 +21 -5
do 0.546074 1.46014 +7 +1 +24 +8 +34 +22 -4
do 0.578012 1.45887 +5 +25
Cd 0.643847 1.45676 +9 0 +25 +9 +34 +24 -5
Hg 1.01398 1.45030 +9 0 +10 +35 +23 -6
do 1.12866 1.44893 +1
do 1.36728 1.44622 +1
do 1.39506 1.44590 -1
do 1.52952 1.44434 +9 0 +10 +35 +24 -7
do 1.6932 1.44234 +1
do 1.81307 1.44078 +1
do 1.97009 1.43861 +8 -1 +8 +36 +25 -8
do 2.24929 1.43431 +1
do 2.32542 1.43303 +9 -1 +7 +39 +28 -10
Polystyrene
absorption 3.2432 1.41326 -5 -5
do 3.2666 1.41263 0 -5
do 3.3033 1.41164 0 -1
do 3.3293 1.41092 0 -6
do 3.4188 1.40839 -12 +3
do 3.5078 1.40577 -4 -7
were made to adjust to exact temperature of 240 and
31'C. On the other prisms all index determinations
were near 24 0 C only.
The refractive indices of the Heraeus B prism for
24 0 C have been represented by means of the dispersion
formula'
0.008777808 84.06224
it2= 2.978645+ _ , (1)
2 -0.010609 96.00000-X 2
0
2,
a
.ZC
.05
.01,
0.2 0.5 1.0 2.0
WAVELENGTH,MICRONS
FIG. 1.
THE DISPERSION OF
FUSED QUARTZ GLASS
.02 \ II
GThis is a modification of a formula adjusted by Rudolph
Kingslake (private communication)'to fit some of our preliminary
and the goodness of fit of the formula is shown by the
small residuals, (B-c), as tabulated in Table I and
by their nearly random distribution. The predominance
of negative residuals for the regions 3 to 3.5g indicates
that further improvement can be made. However no
least squares solution was attempted because at
X=3.4188, where the maximum negative residual was
obtained, there is observational uncertainty because
of the very broad nature of this absorption band which
tends to make the wavelength of the minimum difficult
to determine. Note that a positive residual was obtained
for the Corning prism at this wavelength (last column
Table I).
In order to compare the indices for the various
prisms, the indices as computed by formula (1) have
been subtracted from all others. These differences are
listed in Table I. It will be noticed that the Heraeus,
Nieder, and Corning samples agree in general in
refractivity with the data of Trommsdorff and of
Gifford. Gifford's data were taken on the fused quartz
glass made by Gifford and Shenstone with particular
attempts at purity of material. It seems that the
Heraeus B and Corning samples may possibly be the
more free of traces of other substances, since the
common basic impurities would tend to make the index
higher. The possible effects of other impurities cannot
of course be altogether ignored. On this assumption
of relative purities it seems possible that formula (1)
and the first column of Table I express probable
data on this prism. It is similar to the formula that Rubens
used for the ordinary index of quartz [Ann. Phys. 45, 476
(1895)].

September 1954
REFRACTIVE
INDEX OF FUSE QUARTZ GLASS
679
values for nearly pure fused quartz glass. Further, it
seems advisable to add 2 or 3X1O-4 for probable
values of some specimens of such glass.
The dispersion of fused quartz glass was calculated
over the region measured and found to have its minimum
at about 1.5 as shown in Fig. 1. The region
between 2.4 and 3.2 is represented as a dashed line
since the strong absorption in this region make measurements
impossible. The region beyond 3.2 has dispersion
good enough to enable one to resolve the polystyrene
bands into six distinct parts.
The temperature coefficients were determined for
each of the lines measured. The precision is not adequate
to determine definitely the variation with wavelength.
The average value at the shorter wavelengths was
found to be 10X 10-6 and this may fall to an average of
about 4.OX 10-6 near 3.5yu. The values in the visible
region agree with those of Tilton 3 and others.
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA VOLUME 44, NUMBER 9 SEPTEMBER, 1954
Simple Device for Reducing Microphotometer Traces to Curves of Intensity
J. D. PURCELL
Naval Research Laboratory, Washington 25, D. C.
(Received June 14, 1954)
A mechanical device is described that performs automatically the laborious task of reducing a microphotometer
spectrum trace to a spectral intensity curve. In performing the conversion the microphotometer
trace, divided into sections, is contact printed onto sensitized material using a narrow slit light source and
moving the trace and photographic material at different rates of speed past the slit. The slit is parallel to
the wavelength axis and, in effect, scans along the density axis. The photographic material moves at constant
speed, while the motion of the trace is controlled by a cam whose shape is essentially the photographic characteristic
curve. The density curve is converted to a curve of log intensity in one step, and a second step is used
to convert to linear intensity. Other uses of the device are suggested.
INTRODUCTION
THE most tedious and time consuming operation
in photographic spectrophotometry is the conversion
of photographic densities into intensity values
by means of the photographic characteristic curve.
This problem is particularly acute where a large number
of intensity determinations must be made from a single
film or plate, as in the conversion of a microphotometer
tracing of a spectrum containing many lines, to a plot
of the intensity distribution in the spectrum.
The usual process of transforming microphotometer
tracings into intensity curves begins by tabulating
density values at as many wavelengths as are required
to reproduce the detail in the tracing. The density
values are next converted to intensities by referring
each one to the photographic characteristic curve for
the particular wavelength region involved. The characteristic
curves, of course, have to be determined for the
particular emulsion, wavelength range, processing
conditions, and so forth, which apply to the spectrum
photograph under study. Finally, the completed table
of intensity values is plotted against wavelength. This
procedure is quite tedious and time consuming in the
case of microphotometer tracings of very complex
spectra, and especially so when it is desirable to preserve
accurately the detail and line profiles. Although
none of these steps can be avoided, several instruments
have been developed for performing them more or less
automatically.
Mechanical devices have been described by Beals'
and by Weissler, Einarsson, and McClelland, 2 for
carrying out the reduction semi-automatically. In
both devices it was necessary for the operator continuously
to make a manual adjustment somewhat of the
nature of curve following. In the case of complicated
curves this manual operation would be quite tedious
and would have to be carried out at a rather slow speed
to avoid error.
Several microphotometers have been described which
read directly in intensity. The Williams and Hiltner
direct intensity microphotometer 3 employs two beams
of light which are used to compare the spectrogram
with a specially prepared two-dimensional calibration
spectrogram. Minnaert and Houtgast 4 have described
an attachment for the Moll microphotometer which
produces a direct-intensity trace. This device requires
an opaque mask whose edge is cut in the form of the
photographic characteristic curve. The signal from the
microphotometer photocell is fed into a galvanometer
which sweeps a narrow beam of light across this mask.
The light which passes the mask strikes a second photocell
which actuates the recorder. More recently Brown
and Birtley' have described an instrument which pro-
'C. S. Beals, J. Roy. Astron. Soc. Can. 38, 65 (1944).
2 Weissler, Einarsson, and McClelland, Rev. Sci. Instr. 23,
209 (1952).
3 R. C. Williams and A. Hiltner, Publ. Am. Astron. Soc. 10, 33
(1939).
4 M. Minnaert and J. Houtgast, Z. Astrophys. 15,354 (1938).
5 W. N. Brown and W. B. Birtley, Rev. Sci. Instr. 22, 67 (1951).