Date post: | 14-Apr-2018 |
Category: |
Documents |
Upload: | bbteenager |
View: | 237 times |
Download: | 0 times |
of 22
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
1/22
6. Spectral Techniques
Spectroscopy is the study of
spectra the quantitative
measurement of spectral lines:
their width, depth and shape.
We measure a line profile - a
graph of the specific intensity,
or flux density, of radiationreceived from a source as a
function of frequency.
Often the shape of the line profile is simply characterised by a measure of
its width (e.g. equivalent width see A2 Theoretical Astrophysics).
Spectroscopy is of fundamental importance to astrophysics because it
allows us to deduce many physical characteristics of planets, stars and
galaxies even though we observe them remotely, from enormous distances.
Intensity
frequency
Continuum
0Schematic diagram of an absorption line
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
2/22
From analysis of spectral lines we can learn about:-
Characteristic from
1. Chemical elements frequency
2. Chemical abundances intensity
3. Bulk velocity (i.e. velocity frequency
of atmosphere as a whole)
4. Temperature, pressure, line width
gravity
5. Spread of velocities line width
6. Magnetic and electric field fine structure in lines
(e.g. Zeeman splitting)
0
0
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
3/22
We can collect light in a narrow frequency range using a filter
Effectiveness measured by spectral resolving power,R
Examples: Dye filter
Interference filter
But we need or higher to be useful (i.e. to be
sensitive to Doppler shifts of a few km/s)
=
=
00R (6.1)
10010~ R
410~R
510~R
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
4/22
Much higher spectral resolving powers can be achieved using aPRISM or a DIFFRACTION GRATING
White light Red light
Blue light
White light
Red light
Blue light
Blue light
Red light
Note that a prism disperses blue
light more strongly than red light,while for a diffraction grating red
light is dispersed more strongly
than blue light.
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
5/22
a
For an infinite diffraction
grating with light incident atright angles (normal incidence),
the dispersed light has an
intensity maximum when the
path difference betweenadjacent light rays satisfies
na =sin (6.2)
sinaPath difference =
Constructive interference
a = spacing between lines of grating
n = order of the maximum
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
6/22
If the light is incident at an
oblique angle, , maximaoccur at path differences
which satisfy:-
( ) na =+ sinsin
a
)sin(sin +aTotal path difference =
(6.3)
To determine the angular dispersion
(i.e. the spread in angular deflection
corresponding to a given spread in
wavelength) we differentiate (6.2) or
(6.3) with respect to wavelength.
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
7/22
nd
da =
cosDifferentiating
Hence
cosa
n
d
d=
(6.4)
(requires to be in radians)
Angular dispersion
We can achieve a highangular dispersion via:
o high order
o small
o large
a
Suppose the dispersed light is focussed on a detector (e.g. a CCD) using a
lens or mirror of focal length f
Then
d
d
dfdfdx ==
Linear separation of lines on detectorActual separation of lines in spectrum
(6.5)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
8/22
nd
da =
cosDifferentiating
Hence
cosa
n
d
d= (6.4)
(requires to be in radians)
Angular dispersion
We can achieve a highangular dispersion via:
o high order
o small
o large
a
Linear Dispersion
Often also defined is the Reciprocal Linear Dispersion (RLD)
d
df
d
dx= (6.6)
dx
d=
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
9/22
Intensity
sin
a a2 a3aa2a30
Intensity profile from a monochromatic light source, of wavelength ,
diffracted by an infinite diffraction grating, is a series of infinitely thin
peaks equally spaced in
In practice, of course, any grating is finite (i.e. only a finite number of
lines is illuminated)
sin
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
10/22
Finite diffraction grating
Consider grating with rules.
For the aperture,
Phase difference
By principle of superposition, total wave amplitude diffracted throughan angle is (for an incident wave of unit amplitude)
(modelling the wave as complex simplifies the algebra)
1
2
3
4
5
N - 1
NN
thm
mamm == sin
2(6.7)
)1(2tot 1
++++=Niii eee K (6.8)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
11/22
iNiii eeee +++= K2tot
Multiplying through by
Subtracting (6.7) from (6.6)
i.e.
ie
(6.9)
iNi ee = 11 tot (6.10)
i
iN
e
e
= 1
1tot
(6.11)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
12/22
We can rewrite the RHS as
i.e.
using the result
And the Intensity in direction is then given by
( )
( )2/2/2/
2/2/2/
tot
iii
iNiNiN
eee
eee
= (6.12)
( )( )2/sin
2/sin)1(tot
Ne
Ni
= (6.13)
ixix
i eex
= 21sin
*
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
13/22
Thus
Primary Maxima occur at (same as for infinite grating)
But maxima are not infinitely narrow.
Width of maxima
Also N 2 secondary maxima in between
( ) [ ][ ]2/sinsin2/sinsin 2
2
0
kaNkaII =
Incident intensity
2=k = wave number
na =sin
aN
W
~
(6.14)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
14/22
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
15/22
Thus
e.g. For N = 5
( ) [ ][ ]2/sinsin2/sinsin 2
2
0
kaNkaII =
Incident intensity
2=k = wave number
sin
Intensity
a
a2
a
a2 0
(6.14)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
16/22
Thus
e.g. For N = 6
( ) [ ][ ]2/sinsin2/sinsin 2
2
0
kaNkaII =
Incident intensity2
=k = wave number
sin
Intensity
a
a2
a
a2 0
(6.14)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
17/22
The finite width of the primary maxima limits the resolving power of
a diffraction grating.
Consider two spectral lines,
of wavelength and
Lines are observed at order
light diffracted
through angles satisfyingsin
Intensity
1 22
1
n
a
n 11sin
=
a
n 22sin
= (6.15)
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
18/22
The finite width of the primary maxima limits the resolving power of
a diffraction grating.
We can only resolve the two
lines provided they are separatedby (at least) their width.
In other words, the resolution
limit is
Or
sin
Intensity
21
(6.17)
aNa
n =
nNR =
=
Can use an echelle grating to
achieve a high resolving power;
lines are blazed cut in a
special pattern to concentrate
light in high orders
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
19/22
The finite width of the primary maxima limits the resolving power of
a diffraction grating.
We can only resolve the two
lines provided they are separatedby (at least) their width.
In other words, the resolution
limit is
Or
sin
Intensity
21
(6.17)
aNa
n =
nNR =
=
Spectral resolving power dependsnot on ruling separation, but onthe total number of lines on the
grating and the order of the
maximum observed.
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
20/22
Design of a Slit Spectrometer
Key features of the design are summarised in the following diagram
Detector
Focussinglens
Grating
CollimatinglensSlitTelescope
aperture
The slit cuts out unwanted light. Its angular size at the collimating lens
defines the range of angles entering the grating
Diameter of collimating lens width of grating, so that little light is lost
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
21/22
Design of a Slit Spectrometer
Key features of the design are summarised in the following diagram
Detector
Focussinglens
Grating
CollimatinglensSlitTelescope
aperture
Grating response width = width of primary maxima =
Focussing lens also produces a diffraction pattern (see next section)
which smears out light, with width
Choose so that - i.e.
gratingDNa =
focusD
gratingfocus DD
7/30/2019 a2oa_sec6aaaaaaaaaaaaaaaaaaaaa
22/22
Design of a Slit Spectrometer
Key features of the design are summarised in the following diagram
Detector
Focussinglens
Grating
CollimatinglensSlitTelescope
aperture
Choose focal length of focussing lens so that width of diffraction peak
at the detector width of pixel on detector.
i.e. the diffraction maxima cover several pixels