Zeeman effect

Max-Planck-Institut für Kernphysik,

Saupfercheckweg 1, Heidelberg
22nd May 2006

Introduction
The most important experimental technique triggering the development of the mod-
ern atomic theory is spectroscopy, the observation of the characteristic wavelengths
of light that atoms can absorb and emit. A so-called wavelength dispersive element
such as a waterdrop, a prism or a grating, spatially separates different wavelengths
of light emitted from one source. If the light source emits white light, a band of all
colours will be found after the light passes the wavelength dispersive element (e.g.
rainbow). Observing the light from a lamp that contains only atoms of a certain
element, one finds distinct lines instead of the complete band on the detector. The
positioning of the lines on the detector is called a spectrum, and the corresponding
wavelengths are hence called ‘spectral lines’ of the element.
The additional information needed to be able to draw conclusions about the
atomic structure from spectral lines was provided by the experimental observation
of Rutherford that in an atom, electrons surround a heavy, positively charged nucleus
in the center, and by Planck’s explanation of the blackbody spectrum by quanta of
light, photons, whose energy Eph is connected with the wavelength λ by

hc
Eph = hν = , (1)
λ
where ν is the frequency, h is the Planck constant and c is the speed of light.
With this knowledge it became clear that electrons in an atom can occupy only
certain energy levels Ei , and that the transition of an electron from a higher energy
level E2 to a lower energy level E1 leads to the emission of a photon: E2 − E1 = Eph .
The energy needed to lift the electron to the higher, ‘excited’ energy level can be
provided by absorption of a photon of the very same energy or by collisions.
The increasing experimental precision reached with spectroscopy inspired scien-
tists to propose different atomic models, leading to the development of quantum

1
2

mechanics. With refinements to the quantum mechanical description of the atom,

relativistic quantum mechanics (necessary to explain the the splitting of energy levels
of same main quantum number n) and quantum electrodynamics (QED, necessary
to explain the splitting of levels with same quantum number j but different quan-
tum number l), theory is by now able to make highly accurate predictions about the
atomic structure.

Effects of an external magnetic field on the atomic

structure
If the atom is placed in an external magnetic field, the
I
atomic energy levels are split and the observed spectrum
changes. This was first observed by the dutch physicist
Pieter Zeeman and is hence called the Zeeman effect.
µl In an intuitive approach to explain the Zeeman effect,
consider an atom with an electron with charge e moving
Figure 1: Classical model of
with the speed v around the nucleus on a Bohr orbit with
an electron with angular mo-
mentum l inducing an angu- radius r. This orbiting electron can be described by a
lar magnetic moment µ~l current

v
I = −e · (2)
2πr
inducing an orbital magnetic moment

evr
~µl = I · A = I · πr2 n = n, (3)
2
where A = πr2 n is the area vector perpendicular to the orbit area πr2 . The
angular momentum of the electron with mass me is

l = r × p = me · r · v · n . (4)

An external magnetic field B interacts with the angular magnetic moment and,
hence, changes the electron’s potential energy:

e
∆Epot = −~µl · B = ·l·B (5)
2me
Using the direction of the magnetic field vector B as quantisation axis z and the
quantised angular momentum of the electron
 3

p
|l| = l(l + 1)~ with l = 0, 1, . . . , n − 1 and (6)
lz = ml · ~ with − l ≤ ml ≤ l , (7)

equation (5) can be simplified to

e·~
∆Epot = · m l · B = µB · m l · B , (8)
2me
where µB is called the Bohr magneton. The energy shift of the electron’s original
energy level by ∆Epot due to the coupling of the external magnetic field to the
electron’s angular magnetic moment leads to a splitting of the original level into the
previously degenerated 2l + 1 sublevels of same angular momentum l but different
magnetic quantum number ml .
The quantum mechanical way for deriving this energy splitting where also the
electron spin is taken into account starts off by treating the external magnetic field
as a perturbation ĤB of the unperturbed Hamilton Operator Ĥ0 . In general, an atom
with more than one electron has to be considered. In the case of an external mag-
netic field that is small as compared to the magnetic coupling energies between the i
electrons’ orbital and spin momenta li and si , respectively, one may use the so-called
LS-coupling approximation. This approximation is appropriate when the coupling
energy between the different electrons’ orbital momenta is large as compared to the
coupling of a single electron’s orbital momentum with the same electron’s spin mo-
mentum. Hence, all orbital and spin momenta are added seperately to total orbital
and total spin momenta L and S before determining the total angular momentum
J:

X p
L= li with |L| = L(L + 1)~ , (9)
i
X p
S= si with |S| = S(S + 1)~ , (10)
i
p
J=L+S with |J| = J(J + 1)~ , (11)
J z = MJ · ~ with −J ≤ MJ ≤ J . (12)

With these momenta introduced, the Hamiltonian of an atom in an external

magnetic field is
µB
Ĥ0 + ĤB = Ĥ0 − (J + S) · B , (13)
~
and the solution for the shift of the level energy becomes
4

B=0 B≠0 B=0 B≠0

+2
+1 2
L=2 P1/2 +1/2
0
gJ=2/3 -1/2
-1
-2
hν0 hν0
+1 2
+1/2
L=1 S1/2
0
gJ=2
-1 -1/2
μBB
hν hν hν hν
hν0 hν0 hν0 hν0

Figure 2: Left: Normal Zeeman effect for a d → p transition. The field splits the degenerate
MJ levels by ∆E = µB · B · MJ . Right: Anomalous Zeeman effect. Here, the levels are split by
∆E = µB · B · MJ · gJ . In both cases, transitions can occur if ∆mJ = 0,±1.

∆Epot = µB · B · MJ · gJ (14)
J(J + 1) + S(S + 1) − L(L + 1)
gJ = 1 + , (15)
2J(J + 1)
with the Landé factor gJ .
If all electron spins in the atom cancel each other, i.e. in the case of S = 0,
gJ = 1 is obtained and equation (14) becomes identical to equation (5). For his-
torical reasons, the splitting in absence of spin is called the normal Zeeman effect
whereas, since the electron spin was not yet discovered and the observed features
seen in the spectra were not understood, the more general case with S 6= 0 was
named the anomalous Zeeman effect. In the normal Zeeman effect, the size of the
energy splitting from the unperturbed level energy depends only on the magnetic
quantum number ml and is, hence, of the same size for all levels in the atom. In the
case of the anomalous Zeeman effect, the energy splitting depends on the quantum
numbers J, L and S, which is reflected by more complicated structures observed in
the corresponding spectra (see Fig. 2).
In a strong external field, the LS-coupling approximation is no longer appropriate.
The field-induced precessions are so rapid that one has to consider the total orbital
momentum L and spin S as they individually precess about B, that is, the effect of
B is effectively to decouple L from S, rendering J meaningless. This is called the
Paschen-Back effect.
 5

s
p
B
s-
s+

s
p

s-
s+

Figure 3: Isometric depiction of the polarisation of the different Zeeman components.

Selection rules and polarisation of emitted light

In transitions from an excited level i to a lower level k, a photon with the difference
energy Eph = ∆E = Ei − Ek is emitted. Still, as indicated in Fig. 2, not all energy
differences between existing levels are actually found in the spectra. The reason
is that not only the energy, but also momentum conservation, angular momentum
conservation and symmetry rules play a role on whether or not a transition can occur.
Only those transitions are allowed where the transition dipole matrix element
Z
Mik = e ψi∗ r ψk dV (16)

has at least one non-zero component

Z
(Mik )q = e ψi∗ q ψk dq , with q = x, y, z . (17)

Evaluating these three integrals yields that only those transitions have non-zero
components where the change of the magnetic quantum number fulfills ∆MJ =
MJ,i − MJ,k = 0, ±1. Furthermore, electric dipole transitions require that the change
of the total orbital momentum be ∆L = Li − Lk = ±1 and that the spin does not
change, i.e. ∆S = 0.
If the quantisation axis (i.e. the direction of the magnetic field vector) is the z-
axis, the matrix element component (Mik )z is the only non-zero one for ∆MJ = 0 (so-
called ‘π’-) transitions. Considering a dipole and its characteristics of emission, this
means the dipole is oscillating along the axis of the magnetic field vector and, hence,
does not emit any radiation in that direction. Also, since the electric field vector
is oscillating in only one direction, also the field vector of the emitted light wave
oscillates in this direction, meaning that the light emitted by ∆MJ = 0 transitions
is linearly polarised (see Fig. 3).
6

In the case of ∆MJ = ±1 (‘σ’-) transitions, (Mik )z is zero and both (Mik )x and
(Mik )y are non-zero and equal in magnitude, but phase-shifted by π/2. This phase
shift between the two superimposed dipoles leads to the observation of circularly
polarised light when viewing in z-direction (‘longitudinally’). Viewing the emitted
light in x- or y-direction (‘transversal’), only the projection of the superimposed
dipoles is seen, and thus the light from ∆MJ = ±1 transitions appears linearly
polarised when observed transversal to the magnetic field vector.

The experiment
The experiment consists of two parts. In the first part, you will observe the Zeeman
effect in both transversal and longitudinal direction, and study the polarisation of
the observed light. The second part of the experiment is dedicated to precision
spectroscopy using a grating spectrometer.
In both parts of the experiment, you will study the light emitted from a cadmium
(Cd) lamp. The electronic configuration of Cd is [Kr]4d10 5s2 . In this configuration,
all electronic spins compensate each other, such that S = 0. In the visible, the
spectrum of Cd consists of four lines, as listed in table 1.

Table 1: Spectral lines of Cd in the visible

Color violet blue green red
Wavelength [nm] 467.9 480.1 508.7 643.8

All measurements will be performed using the the red line at λ0 = 643.8 nm,
(ν0 = 465.7 THz) emitted by the electronic transition between the two excited
levels 1 D2 → 1 P1 in the fifth shell of Cd.

Part I: Spectroscopy of the Zeeman effect

The expected shift of the emitted wavelengths due to the external magnetic field
is well below 0.1 nm, hence a high resolution device has to be used in order to
observe the effect. In this experiment, a Lummer-Gehrcke plate is used (see Fig. 4).
The Lummer-Gehrcke plate is a quartz-glass plate with extremely plane parallel
surfaces. Light is coupled into the plate by a prism and then reflected back and forth
repeatedly inside of the plate close to the angle of total reflection. Thus, a small
fraction of the light intensity emerges at each reflection point and travels at an angle
of α ≈ 90◦ to the surface normal. The emerging rays are observed with a telescope
focused to infinity behind the Lummer-Gehrcke plate. Since the thickness of the
 7

Figure 4: Lummer-Gehrcke-Interferometer.

k
...
Δ2 3
α n1 2
β α 1
d n2
ε
Δ1

Figure 5: Left: Schematic to illustrate optical retardation using a glass plate. Right: Lummer-
Gehrcke plate.

plate is much smaller than its length and because of the high reflection coefficient,
many rays can interfere with each other, resulting in a very high resolution.
As shown in Fig. 5 (left side), the optical retardation between two light rays from
a glass plate, where one was reflected inside the plate once, is given by

d
∆ = 2n2 · − 2n1 · d · tan ǫ · sin α , (18)
cos ǫ
where d is the thickness of the plate, n1 and n2 are the refractive indices outside
and inside the plate, and α and ǫ are the angles of the light to the surface normal out-
side and inside the plate, respectively. Using sin α = nn21 sin ǫ and sin2 ǫ + cos2 ǫ = 1,
Eq. 18 becomes
q
∆ = 2n2 · d · cos ǫ = 2 · d · n22 − n21 sin2 α . (19)
In the case of the Lummer-Gehrcke plate, the light is coupled into the plate by
a prism to allow angles α ≈ 90◦ at the limit of total reflection. For a plate with
n2 = n used in air (n1 ≈ 1) Eq. 19 reduces to
√
∆ = ∆1 − ∆2 = 2d · n2 − 1 , (20)
were ∆i are indicated in the right part of Fig. 5. For constructive interference,
the optical retardation has to fulfill ∆ = m · λ, where k is an integer number and λ is
8

the wavelength of the interfering light. Obviously, whether or not constructive inter-
ference occurs when observing the Lummer-Gehrcke plate with a telescope depends
on the wavelength λ and on the exiting angle α, which again depends on the angle
β at which the light enters the Lummer-Gehrcke plate. For a given wavelength λ,
one obtains many orders k for the different incoming angles β at which the emerging
light beams interfere constructively (see Fig. 6).
One finds spatial overlap between the order k of wavelength λ + ∆λ and the order
(k + 1) of wavelength λ if

λ
k · (λ + ∆λ) = (k + 1) · λ → ∆λ = . (21)
k
With Eq. 20 one finds the free spectral range of the Lummer-Gehrcke plate to be

λ2
∆λ = √ . (22)
2d · n2 − 1
In first approximation, a small change in wavelength δλ < ∆λ leads to a shift

δa
δλ = · ∆λ , (23)
∆a
where δa is the distance between the line with wavelength λ+δλ from the position
of the line with wavelength λ, and ∆a is the distance of two orders k and k + 1 of
wavelength λ. Hence, for a small wavelength shift as it is induced e.g. by the Zeeman
effect, one obtains

δa λ2
δλ = · √ . (24)
∆a 2d · n2 − 1
The experimental setup is sketched in Fig. 7. The Cd lamp is placed between the
pole shoes of a magnet. Light emitted from the lamp enters the Lummer-Gehrcke
plate after passing a red filter (selecting the red spectral line). The magnet can be
rotated by 90◦ , such that the lamp can be observed both in transversal direction
to the magnetic field (as shown in Fig. 7) or, through a hole in the pole shoes, in
longitudinal direction of the magnetic field. The interference pattern generated by
the Lummer-Gehrcke plate is observed with a telescope. One can place a polarizer
and a λ/4 plate between the Lummer-Gehrcke plate and the telescope to observe
the different polarizations of the observed light rays.
With a height adjustment of the telescope, one can move a crosshair in the ocular
over the interference pattern. The position a of the crosshair is shown by a dial
gauge.
 9

Figure 6: Interference pattern for the Zeeman effect observed in the transverse configuration a)
without polarization filter, b) with polarization direction of the filter perpendicular to the magnetic
field and c) with polarization direction of the filter parallel to the magnetic field.

Figure 7: Sketch of the setup for studying the normal Zeeman effect (figure from Leybold, Physics
Leaflets). a) magnet pole pieces, b) Cd-lamp, c) red filter, d) Lummer-Gehrcke-plate housing, e)
telescope, f) ocular, g) height adjustment for telescope, h) arresting screw for column, i) arresting
screw for column base
10

Part I: Objectives
• Use the Hall probe teslameter and measure the magnetic field strength at the
position where the Cd lamp will be placed in dependence of the Voltage of the
magnet power supply. Use steps of 1 Volts and measure both increasing and
decreasing voltages (magnet hysteresis).

• Place the Cd lamp in the holder (avoid using any force!) and observe the
Zeeman effect in both transversal and longitudinal direction.

• Use the polarizer and λ/4 plate and confirm the different light polarizations
depending on the observation direction.

• In transversal observation direction, determine the positions a of all σ- and

π-lines for 12 − 15 interference orders at the voltages U = 6 V, U = 9 V, and
U = 12 V.

• Data analysis: With help of table 2, correct the measured magnetic field
strength values. Plot the corrected values of B versus the voltage, apply a lin-
ear fit. Use the fit result to determine the B-field strength at U = 6 V, U = 9 V,
and U = 12 V.

• Data analysis: for each magnetic field, plot the interference order k versus
the position a of the π-lines and find a polynomic function k = f (a) that
fits the data. The spacing of the different orders on the y-axis is f (∆a) =
f (a2 ) − f (a1 ) = 1. Determine the ratios δa/∆a with the fitted curve, calculate
δλ with Eq. 24 (Lummer-Gehrcke plate data: Refractive index n = 1.4567;
thickness d = 4.04 mm).

• Data analysis: Calculate µB for the three magnetic field strengths and compare
with the literature value of µB = 9.27400949 · 10−24 J/T.

Safety notes:
• Do not remove the cover of the Lummer-Gehrcke plate, it’s very expensive!

• When the magnet current is switched on, do not handle ferromagnetic objects
in the vicinity of the cadmium lamp.

• Never touch the Cd lamp with your bare hands. Any moisture/fat can lead to
friction forces in the lamp’s glass that may break it when it’s turned on, and
it becomes very hot.
 11

Table 2: Teslameter correction table

Increasing Voltage Decreasing Voltage
B measured B accurate B measured corrected B
0.006 0.004 0.965 0.885
0.079 0.075 0.876 0.866
0.159 0.155 0.893 0.826
0.225 0.223 0.840 0.786
0.304 0.291 0.801 0.745
0.382 0.362 0.740 0.682
0.444 0.425 0.689 0.624
0.518 0.495 0.617 0.571
0.568 0.557 0.542 0.507
0.666 0.637 0.485 0.446
0.722 0.688 0.422 0.390
0.803 0.723 0.335 0.326
0.848 0.766 0.280 0.252
0.889 0.807 0.218 0.199
0.907 0.867 0.140 0.130
0.942 0.870 0.074 0.066
0.965 0.885 0.006 0.004
12

Part II: Precision spectroscopy

While the Lummer-Gehrcke plate used in the first part of the experiment is highly
suited for measuring small wavelength differences, it cannot be used for wavelength
determination. As you will learn in the second part of the experiment, one can
perform highly accurate wavelength measurements using a grating spectrometer.

Czerny-Turner spectrometer
The Czerny-Turner (CT) spectrometer disperses light into discrete wavelengths by
means of a grating. The spectral image is viewed in the exit focal plane using a
two-dimensional CCD camera. A concave optical mirror (MC) is used to collimate
the radiation entering the spectrometer through an entrance slit. After diffraction
by the grating, the light is subsequently focused by a second concave mirror (MF)
onto the CCD camera, as shown in Fig. 8.

Focal plane Grating tower

MF
CCD camera MC

Entrance slit
Figure 8: Illustration of the Czerny-Turner spectrometer.

Grating properties
In a grating, the incident and reflection angles of a light beam with a wavelength λ
(see Fig. 9) are related by the following equation:

sinα + sinβk = knλ , (25)

where α is the incidence angle, βk is the diffraction angle relative to the grating
normal (NG ), k is the diffraction order and n is the groove density (1/d). When
k = 0, Eq. (25) is reduced to α = β0 , the specular reflection law. If the diffraction
 13

angle is fixed, the difference between α and βk , the so-called deviation angle (DV ),
is constant (see Fig. 9).
NG
NB

bk Incident
a light
Reflected wB
light
DV
LB
LA
d
wB

Figure 9: Scheme of the light reflexion in a grating. NB is the blaze normal and ωB defines the
blaze angle. The distance between two consecutive grooves is defined by d.

A reflection grating can concentrate most of the diffracted spectral radiation

into a single spectral order and, therefore, reduce the intensity of all other orders.
This redistribution of intensity among the orders depends on the angle between the
reflecting elements and the grating surface, which is called the blaze angle ωB (see
Fig. 9). In the visible spectrum, red light is deviated the least and violet the most.

Dispersion
The derivative of the diffraction angle over the wavelength is known as the angular
dispersion. It is a measure of the angular separation between beams of adjacent
wavelengths. An expression for the angular dispersion is derived by differentiating
Eq. (25) for a fixed incident angle α:

∂βk kn
= . (26)
∂λ cosβk
High dispersion can be achieved either by choosing a grating with a high groove
density (n), or by using a coarse grating in high diffraction order (k). The linear
wavelength dispersion at the exit focal plane of a spectroscopic instrument is usually
specified as reciprocal linear dispersion given in nm/mm. If the focal length of the
instrument is LB , then the reciprocal linear dispersion is given by:

∂λ kncosβk
D(λ) = = , (27)
∂p LB
14

where p represents the distance in mm. Since the size of the instrument depends
on the focal length of the optical system, by choosing a grating with a high groove
density, the instrument can be made more compact.
As shown in Eq. (27), the relation between the pixel position p on the CCD
camera and the real wavelength λ is given by the dispersion function D(λ). This
function depends on the wavelength. At least one reference point p0 in the spectrum
(a known wavelenght λ0 ) is required to calibrate the wavelength scale, if D(λ) is
known. Then the λ of any other line on the spectrum can be obtain by

Z p
λ = λ0 + D(λ)dp . (28)
p0

This is, in general, difficult and often only numerically solveable because the
dependence of the dispersion on the wavelength is trivial. If the dispersion function
is independent of λ, Eq. (28) is expressed as

λ = λ0 + D(p − p0 ) , (29)
and the desired wavelength λ can be determined.
In this experiment, you will use reference lines to find the dispersion function
of the spectrometer. You can approximate the dispersion with the positions of
the reference lines on the detector pi and their respective wavelengths λi with a
polynomial function λ(p)

λ(p) = A + B · p + C · p2 + . . . , (30)
where A, B, C... are free parameters.

Resolution
The resolution or chromatic resolving power of a grating describes its ability to
λ
separate adjacent spectral lines. The resolution is generally defined as R = ∆λ ,
where ∆λ is the difference in wavelength between two spectral lines with equal
intensity that are separable.
The limit of resolution of a grating is R = kN, where N is the total number of
grooves illuminated on the grating. A more practical expression for the resolution
is obtained using Eq. (25)

N d(sinα + sinβk )
R = kN = . (31)
λ
The spectrometer is equipped with a grating with 1800 groves/mm.
 15

CCD detector
A CCD camera is a solid state detector array made of silicon with a sensitive range
from 400 to 1100 nm. It converts the incoming light into electrons through the
photoelectric effect. These free electrons are stored in a rectangular array of imaging
elements called pixels defined by a grid of gate electrodes in the X and Y directions,
respectively. The CCD provides simultaneously information for both intensities and
positions projected along the height of the spectrograph image plane. The CCD
chip size used is 30 × 12 mm2 and is composed of 2048 × 512 pixels. Each pixel
has an area of 15 × 15 µm2 . The camera is cooled with liquid nitrogen (LN2) to
reduce thermal noise, and can integrate weak signals for hours without accumulating
disturbing background (less than 1 electron/pixel/hour). The typical readout noise
is less than 3 electrons/pixel at -140◦ C.

Calibration

Calibration lamp
In order to calibrate the spectrometer, i.e., to find the dispersion, reference lines
are needed. For this purpose a neon calibration lamp is used. The used references
lines, listed in Table 3, are taken from the spectral line database at NIST (National
Institute for Standards and Technology, http://www.nist.gov/).

Table 3: Reference lines used for the calibration taken from http://www.nist.gov/

Species λ (nm) Species λ (nm) Species λ (nm)

Ne I 585.24878 Ne I 621.72812 Ne I 667.82766
Ne I 588.18950 Ne I 626.64952 Ne I 671.70430
Ne I 594.48340 Ne I 630.47893 Ne I 692.94672
Ne I 597.55343 Ne I 633.44276 Ne I 702.40500
Ne I 602.99968 Ne I 638.29914 Ne I 703.24128
Ne I 607.43376 Ne I 640.22480 Ne I 717.39380
Ne I 609.61630 Ne I 650.65277 Ne I 724.51665
Ne I 614.30627 Ne I 653.28824 Ne I 743.88981
Ne I 616.35937 Ne I 659.89528
16

Part II: Objectives

The aim of this experiment is to measure the wavelength of the red spectral line of
Cd. Futhermore, another (very weak) line is emitted by the Cd lamp at ≈ 652 nm
wavelength. Using high precision wavelength determination, identify the element
emitting this line.

• Preparation: The bandwith observed on the detector is ≈ 20 nm. Conse-

quently, the grating has to be moved in order to project different parts of
the Ne and Cd spectra on the CCD detector. The grating in the spectrometer
converts wavelength/energy into a geometrical property, namely the dispersion
on the CCD detector. As a result, one has to be careful about the geometry
of the complete experimental setup (not only the spectrometer!) along the
light path. Think about possible error sources in the measurement and how
to circumvent them.

• Preparation: What is the expected shape of the spectral lines?

• Experiment: Take an overview spectrum of the light emitted by the Ne lamp

showing all Ne lines listed in table 3. The spectrometer calibration is set to
be off by more than 10 nm. Identify the Ne lines with the help of the reference
spectrum lying out at the practicum.

• Experiment: Take an overview spectrum of the light emitted by the Cd lamp

showing all four strong visible Cd lines.

• Experiment: Think of an optimal measurement strategy to minimize system-

atic errors (i.e. exposure time, calibration procedure, etc.). Discuss with the
experiment supervisor before proceeding with the experiment.

• Experiment: Obtain calibration spectra and the necessary spectra of the Cd

lamp for the high-precision measurements as discussed with the supervisor.

• Data Analysis: Determine wavelengths of the two red spectral lines emitted
by the Cd lamp with the help of the Ne calibration lines and Eq. 30. Identify
the unknown element with the help of the NIST Atomic Spectra Database
(http://physics.nist.gov/PhysRefData/ASD/lines form.html)
 17

Appendix: Spectrometer control program

Figure 10: Main window of the spectrometer control program SPECTRAMAX by the company
Jobin Yvon. You can zoom into spectra shown in the central part of the window by clicking and
holding the left mouse button and dragging the mouse, and then left-clicking on the rectangle you
created this way. You can return to viewing the full spectrum by clicking the button with the two
perpendicular arrows in the top button row. By clicking the button showing a list (second from
the right in the button row) you open the spectrometer control dialog shown in Fig. 11 on the next
page
18

Figure 11: Spectrometer control dialog. Choose the name of a data file before taking a spectrum
by clicking the button labelled ’Run’. CAUTION: do not use filenames with more than 8 characters!
The button ’CCD Areas’ opens the dialog shown in Fig. 12. The button ’CCD regions’ opens the
dialog shown in Fig. 13 and Fig. 14. The button ’Slits’ opens the dialog shown in Fig. 15. All
other controls are not needed for the experiment to be conducted.
 19

Figure 12: CCD areas dialog. Here you can define which CCD pixels you want to read out. The
checkbox ’linearization’ defines whether a spectrum is shown with a wavelength scale (linearization
on) or pixel scale (linearization off).

Figure 13: CCD regions dialog, Position tab. Here you can define which wavelength you want to
project to the center of the CCD camera and the exposure time (integration time, in seconds). The
spectrometer will move the grating once the button ’Run’ is clicked in the spectrometer control
dialog shown in Fig. 11.
20

Figure 14: CCD regions dialog, Range tab. If you want to take an overview spectrum which is
broader than the bandwidth covered in one position of the grating, use this dialog to define the
lower and upper wavelength and the exposure time. The spectrometer will obtain a number of
spectra (shown above the bandwidth selection as <exposures that will be made> / <exposures
needed to cover selected bandwidth>) and automatically merge them afterwards once the button
’Run’ is clicked in the spectrometer control dialog shown in Fig. 11.

Figure 15: Slits dialog. The spectrometer is equipped only with an entrance slit. Here you can
define the width of this slit (in mm). Note that depending on the intensity profile illuminating the
entrance slit, changing the slit width during the experiment can lead to systematic errors.