e/m

4. Electron Charge-to-Mass Ratio

4.1 Introduction

In this laboratory, we study the interaction of electrons and a magnetic field, with the specific pur-
e
pose to determine the electron charge-to-mass ratio m , sometimes referred to as the specific charge
of the electron or the mass-normalized charge. Throughout this manual, we use the designation
~ and magnetic induction or simply field for the field B.
magnetic field for the field H, ~

4.1.1 Scientific background

While the charge of the electron, −e 1 may be determined separately, as demonstrated in Milikan’s
oil drop experiment (cf lab MI), the electron mass me , because of its exceedingly low value2 , cannot
be found in a direct experiment with satisfactory precision. The mass may be determined, though,
if the charge-to-mass ratio is known. For convenience, we write in this text m for the electron mass.
In the beginning of the era of modern physics, in 1897, J.J. Thomson3 with others performed
measurements convincingly demonstrating that the previously observed cathode rays 4 were actually
particles. These were later identified as electrons5 , and Thomson is now generally credited with
their discovery. The 2006 CODATA6 recommended value is e/m = 1.758820150(44) × 1011 C/kg.
CODATA refers to this as the electron charge-to-mass quotient, but the term ratio is still in common
use.
Methods to determine the charge-to-mass ratio often rely on the observation of the deflection of a
charged particle in external magnetic or electric fields. A well known application of this principle
is the mass spectrometer. The same principle can be used to extract information in experiments
with rays of particles using the Wilson cloud chamber.

1
We adopt here the convention to designate e as the elementary positive charge to avoid confusion over its sign;
CODATA 2011: e = 1.602176565(35) × 10−19 C.
2
CODATA 2010 gives me = 9.10938291(40) × 1031 kg
3
Sir Joseph John Thomson, OM, FRS (1856-1940), British physicist. Nobel laureate 1906.
4
First observed in 1869 by German physicist Johann Hittorf, and named Kathodenstrahlen in 1876 by Eugen
Goldstein
5
The original name is retained, e.g. in the word Cathode Ray Tube (CRT).
6
Committee on Data for Science and Technology

i
ii 4. Electron Charge-to-Mass Ratio

4.1.2 Tubes and electron guns

The 19th century experiments with cathode rays involved glass containers of different shapes and
designs, but generally called tubes, of which an important example is the Crookes tube7 . These
tubes could be evacuated to a low pressure, and contained two separate electrodes connected to
an external electric circuit: a negative electrode, the cathode, either cold or externally heated; and
the electrically positive anode. Electrons produced at the cathode will be accelerated towards the
anode by the electrostatic field.
Cathode rays were so named because they were early on observed to emanate from the negative
electrode. The ’rays’ themselves are invisible, but may be detected indirectly, either as a glowing
light from the glass wall behind the anode, or, if the tube has a rarefied gas-filling, as a trace of
light along the path of the rays, the colour of which depends on the atomic species of gas. In the
first case, the visible light is a greenish fluorescence from the atoms of the glass wall being hit by
the rays; in the latter, the light indicating the trace of the electrons is actually light from the gas
atoms returning to a lower state after having been excited by the electron beam.
The production of the electron beam is of secondary interest in this lab, but nevertheless we need
some understanding of the physics behind. A dedicated electron gun (also called electron emitter)
is employed to generate an electron beam, see Fig.4.1. The cathode of the gun is an electrically
heated metal filament fed by a separate external source of standard 6.3 V. Electrons are emitted
from the surface of the cathode through thermionic emission, meaning that the thermal energy
overcomes the binding energy of the electrons in the cathode, which is often coated with carbides
or borides of transition metals to diminish the exit potential.
The free electrons are attracted by the anode and accelerated so that a collimated beam of electrons
with approximately the same speed leaves through a small hole (aperture) in the anode. Often
additional electrodes are provided for different purposes. Most common is the Wehnelt cylinder 8 ,
which functions as an electrostatic lens, concentrating (focussing) the beam before it arrives at the
anode. The cylinder is placed between the cathode and the anode. It is held biased to a variable,
slightly negative voltage relative to the cathode, in order for it to repel electrons leaving the cathode
in directions not parallel to the axis of the gun.
In the present laboratory, two different sets of equipment will be used: one with a magnetic field
perpendicular to the beam of electrons; another with the magnetic field in parallel to the beam.

4.1.3 Beam parallel to a magnetic field

In this experiment, also called the longitudinal arrangement of field and beam, we put to use a
cathode ray tube of a type previously commonplace in electronic oscilloscopes, shown in Fig. 4.4
inside the coil that produces the field parallel to the axis of the tube. The electron gun is mounted
at the narrow end of the tube, which opens in a shape resembling a circular cone towards the
wider, approximately flat end, coated with a fluorescent material. The CRT is provided with pairs
7
Sir William Crookes, OM, FRS (17 June 1832 – 4 April 1919) chemist and physicist.
8
Invented 1902 to 1903 by German physicist Arthur Rudolph Berthold Wehnelt.

Laboratory Manuals for Physics Majors - Course PHY112/122

4.2. THEORY iii

Abbildung 4.1: Electron gun in cathode ray tube installation. The electron beam leaves to the left.
Clearly seen is the cathode with three holes - there are three separate cathodes incorporated in this
design - the Wehnelt cylinder and the anode with its single hole. The wedge shaped structure to the
left is a pair of electrostatic deflection plates. The scale of the picture is such that the horizontal
section shown has an actual length of about 5 cm.(from Wikipedia: willemdd )

of electrostatic plates, allowing the beam to be deflected by transversal electrostatic fields, though
only horizontal deflection is used in this lab. The trace of the beam may be observed on the screen
as it strikes the fluorescent coating.

4.1.4 Beam perpendicular to a magnetic field

Alternatively, this is also the transversal arrangement of beam and field. The tube used here is
sometimes known as the Perrin tube9 . The electron beam comes from a gun mounted in a neck
of a spherical tube, and directed through the centre to become available for experiments. Since
the pressure in the tube is so low that almost no interaction between the gas molecules and the
electrons takes place, there is no visible trace of the beam, except for the screen inserted vertically
in the centre of the tube with an inclination of 15◦ to the beam. The beam leaves the anode through
a horizontal slit and is thus wide enough to excite a trace along the entire screen, which is covered
on the reverse with a fluorescent coating and has a printed cm-scale grid (graticule) on the front.

4.2 Theory

4.2.1 Accelerated electron beams

For simplicity, we let the electron have the positive charge e, and adjust the relationships accordin-
gly. With UA the accelerating voltage between cathode and anode of the electron gun, the speed of
the electron with charge e may be calculated from the conservation of energy: electrical potential
energy dispensed between cathode and anode must equal the kinetic energy gained by an electron
9
Jean Baptiste Perrin (30 September 1870 – 17 April 1942). French physicist and Nobel laureate in 1926.

Laboratory Manuals for Physics Majors - Course PHY112/122

iv 4. Electron Charge-to-Mass Ratio

as it leaves the gun through the anode aperture.

1
m · v 2 = e · UA . (4.1)
2
Solving for the speed v then gives: s
2 e · UA
v = . (4.2)
m
Within narrow limits, which depend on the statistical processes involved in the thermal emissi-
on at the cathode, and stray fields in the electron gun, all the electrons of the beam will have
approximately the same speed.

4.2.2 Motion of an electron in a homogeneous perpendicular magnetic field

In order for the electron beam to travel in a homogeneous field B, ~ the spherical tube is positioned
symmetrically between the coils of a Helmholtz pair, with their common axes running through
the centre of the tube, such that the field is perpendicular to the electron beam. A charge e with
~ is subject to the Lorentz force F~L = e · ~v × B.
velocity ~v moving in the field B, ~ The force is always
perpendicular to the propagation of the electron and thus no work is done and no kinetic energy
lost or gained.
For the special case of an electron moving at speed ~vs perpendicularly to the field lines of a
~ the magnitude of the Lorentz force is FL = e · vs · B constant. The path of the
homogenous field B,
electron becomes a circle with radius r:
m · vs2
= e · vs · B (4.3)
r
or explicitly
vs
r = . (4.4)
(e/m) · B
The time T required for one complete turn in the circle is accordingly:

2π · r 2π
T = = , (4.5)
vs (e/m) · B

which apparently is independent of the speed of the electron. To illustrate this, we may imagine
~ see Fig. 4.2. The electrons start at the common
three electrons in a plane perpendicular to a field B,
point P at time t = 0, but in different directions and with different speed. Equation 4.4 tells us that
the electrons follow circular paths with different radii and Eq. 4.5 that they all require the same
time T for a full turn in the circle, and meet again at point P at the same time t = T later. Since
F~L = e · ~v × B,
~ the Lorentz force vanishes for electrons moving strictly parallel to the field lines. In
this case there would be no interaction with the field that could be used to determine e/m.
If, however, the electron is moving with velocity ~v at some angle ϕ to the direction of a homogenous
field, then decomposing ~v into components parallel and perpendicular to the field allows us to device
a method for the determination of e/m. No Lorentz force is acting on the component vp = v · cos ϕ
parallel to the field lines, and consequently the component vp of the electron remains unaffected.

Laboratory Manuals for Physics Majors - Course PHY112/122

4.2. THEORY v

v1(t=0)

e3 e1
P
v3(t=0)

v2(t=0)
B e2

B perpendicular to the drawing plane

Abbildung 4.2: Circular traces of three electrons in a field B ~ perpendicular to the plane of the
paper, starting at point P in different directions with different speed.

In a plane perpendicular to the field, the tangential component of the motion with speed vs = v·sin ϕ
is subject to a Lorentz force that guides the electron on a circular path with a constant radius,
according to Eq. 4.4. Figure 4.3 shows that the combined motion is a helical path with axis parallel
to the field and with constant radius. In a distance d in the direction of the field, given by
2 π · vp
d = vp · T = (4.6)
(e/m) · B

the electron completes a full circular turn perpendicular to the field.

vs v0
r
ϕ
P P'
vp

~ starting at an angle ϕ at point P and

Abbildung 4.3: Path of an electron in a homogenous field B,
completing one turn of a circle.

Laboratory Manuals for Physics Majors - Course PHY112/122

vi 4. Electron Charge-to-Mass Ratio

4.3 Experimental

The experimental work requires skillful adjustments and careful reading of voltages and currents. In
some instances, to enhance precision, current is read from an analogue galvanometer with a narrow
strip of mirror below the tip of the pointer. The position of the pointer should be read when the
pointer obscures its own reflection. In this manner the reading becomes parallax free, and a major
source of error eliminated.

4.3.1 ~ parallel to the electron beam

Determination of e/m with a field B

In the actual set-up, a cathode ray tube is mounted on a horizontal rail that allows a coil of copper
wire winding (solenoid) to be slid over the tube, as shown in Fig. 4.4. A field of homogeneous
magnetic induction B ~ may thus be applied parallel to the axis of the tube.

electron trajectories

electrostatic screen
lenses deflection plates
cathode
576.4mm

The dimensions are as follows:

R
solenoid: l = 395 ± 1 mm,
1'204.8mm
- + d R = 39 ± 1 mm,
acceleration voltage UA solenoid electron path d = 75 ± 2 mm.
(electron energy)

Abbildung 4.4: Longitudinal section through the oscilloscope tube inside the coil producing a lon-
gitudinal field. The indirectly heated cathode is indicated together with the anode acceleration
voltage circuit.

The beam of electrons that leaves the electron gun travels on a straight line parallel to the axis of
~ The electrons leaving the electron gun
the tube, and is not influenced by the presence of a field B.
have the same speed v (within narrow limits) determined in Eq. 4.2 by the acceleration voltage
UA . The small aperture at the gun exit may be considered to be a common point from which the
electrons start, corresponding to the point P in Fig. 4.3.
With a pair of plates just after the anode, the beam may be deflected in a horizontal electrostatic
sinusoidal AC field, which becomes visible at the fluorescent screen as a straight line, symmetrical
about the centre point, and of a length proportional to the amplitude of the applied voltage. In
~ equivalent
effect, this allows the user to let the electron travel with a small angle ϕ relative to field B,
to the situation discussed above. The difference is only that, for practical reasons, we now have not

Laboratory Manuals for Physics Majors - Course PHY112/122

4.3. EXPERIMENTAL vii

a single constant angle ϕ that would correspond to single point on the fluorescent screen, but a time
dependent distribution of angles ±ϕ, corresponding to the line. The angles ϕ may be considered to
be small and therefore cos ϕ ≈ 1 and thus vp = v · cos ϕ ≈ v, independent of the exact value of ϕ.
As the field strength B gradually increases, the trace of the beam, the straight line on the screen,
graduately rotates and decreases in length, because the angular velocity is independent of the
electron speed and of the radius of the trace of the electron as can be seen from Eq. 4.5. All traces
will meet again in a single point (image of the effective electron source) provided that the field
strength B corresponds to an integer number of circles in the plane perpendicular to the magnetic
field lines. To put it differently, for the correct value of field strength B, such that Eq. 4.6 is satisfied,
we expect to see a single point on the screen. We may then replace vp in Eq. 4.6 with v from Eq. 4.2
and solve for the ratio e/m. This method was first used by H. Busch in 1922.10

8 π 2 · UA
e/m = . (4.7)
B 2 · d2
The control box of the CRT, marked PS zu KO-RÖHRE, has a few rather obvious controls. From
left to right: a red pilot light and a white toggle switch for the electric power, a black radio knob
for beam focus, a toggle switch marked Ablenkung for the fixed amplitude horizontal deflection of
the beam (50 Hz sinusoidal AC voltage applied to the horizontal plates), and below another black
knob for adjusting the intensity of the beam. The calculation of the solenoid field B is explained
in an appendix at the end of this manual.

• Ensure that the CRT is situated roughly in the center of the solenoid.

• Switch on the high-voltage supply (Heinzinger) for the cathode (electron acceleration). By
pressing Preset the current limit can be adjusted (2 mA). The voltage should be zero. Press
HV on and increase the voltage to about 400 V, switch on deflection. Adjust focus and
intensity, such that for B = 0 a sharp, horizontal trace is seen on the screen Limit the
intensity in order not to damage the fluorescent coating of the tube.

• Turn on the supply of the solenoid (Oltronix). Make certain that an external ammeter is used
for reading the current. From time to time check that the coil does not become excessively
hot (the outside should not feel hot to the touch of your hand). Increase the current and
describe how the trace on the screen changes.

• Increase the solenoid current and find the value for which deflected electrons are refocussed.
Determine the uncertainty of this subjective procedure. Make notes for the report about your
observations.

• Choose six values for the acceleration voltage ranging from 350 V to 600 V. For each value
determine the current needed to refocus the electron beam (again: only shortly apply high
currents and ensure that the current does not exceed 4 A!).

• Calculate for each current value the field strength B inside the solenoid using Eq. 4.15 and
plot B 2 as function of voltage UA .
10
H. Busch, Physik Zeitschrift 23, S. 438 ff. (1922); see as well H.V. Neher, Amer. J. Phys. 29, 471 (1961).

Laboratory Manuals for Physics Majors - Course PHY112/122

viii 4. Electron Charge-to-Mass Ratio

• Fit a line to the data points and determine e/m from its slope using Eq. 4.7. Calculate the
error on e/m.

4.3.2 ~ perpendicular to the electron beam

Determination of e/m using a field B

~ with speed
A beam of electrons that enters perpendicularly into a homogeneous field of induction B
v, describes a circle with radius Eq. 4.4. Eliminating the speed from Eq. 4.2 we obtain
2 UA
e/m = . (4.8)
r2 · B 2
The specific charge of the electron, e/m, is thus given by the radius of curvature of its circular path
r, the acceleration voltage UA and the magnetic induction B.
The experimental set-up is outlined in Fig. 4.5. A pair of Helmoltz coils (not shown) is positioned
such that a horizontal field acts perpendicular to the beam. The geometry and the pertinent field
calculations are explained in the appendix

b
UA
K A

0 B
a

r
K = cathode r−a
A = anode

Abbildung 4.5: Principles of the experiment in the Perrin tube, showing in the plane of the paper
the electron gun and a beam of electrons bent into a circular path by a perpendicular magnetic
~
induction B.

• Qualitatively investigate the independent influence of the acceleration voltage and the current
through the Helmholtz coils

• Set the acceleration voltage to 3000 V and adjust the coil current such that the trace goes
through the point of the graticule with co-ordinates (10,+2). The beam is assumed to pass
horizontally through the origin. Read the corresponding coil current from the mirror scale
galvanometer.

• Calculate the magnetic induction B for the midpoint between the coils, using Eq. 4.20 (ap-
pendix). Radius of the Helmholtz coils is R = 6.8 cm (estimate the error), number of turns
n = 320.

• Calculate the radius of curvature r of the trace using the co-ordinates a and b given in Fig. 4.5,
using the equation of the circle (r − a)2 + b2 = r2 . We may solve for the radius of curvature

Laboratory Manuals for Physics Majors - Course PHY112/122

4.3. EXPERIMENTAL ix

to obtain
a2 + b2
r= . (4.9)
2a
Estimate the primary errors and calculate the error of the radius using error propagation.

• The assumption that the beam passes through the origin of the graticule in parallel to the
grating lines is generally not true, which is liable to introduce a systematic error in the
measurement. It may be removed by reversing the direction of the field, thus making the
circle bend in the other direction. This operation does not, however, remove the random
errors in the reading of the co-ordinates, which has to accounted for properly.

• Repeat the previous determination of the radius of the circle in the new orientation: adjust
the current so that the trace passes through (10,-2) and read the current as before.

• Calculate the magnetic induction, again using Eq. 4.20, and the mean value of the current.
Then calculate the ratio e/m according to Eq. 4.8.

• Repeat the procedure above with an acceleration voltage of 4000 V, or, if this cannot be
attained, a value for which the reading on the voltmeter of the supply is stable.

• Calculate the error of the measurement of e/m, using estimated errors of the measured pa-
rameters and error propagation for Eq. 4.8.

• Calculate the mean of the two separately obtained values of the ratio e/m.

Laboratory Manuals for Physics Majors - Course PHY112/122

x 4. Electron Charge-to-Mass Ratio

4.4 Appendix

a) The magnetic induction of a solenoid.

~ is produced with a cylindrical

In the experiment using a longitudinal field, the magnetic induction B
solenoid. The magnetic field H~ in the coil may be calculated using the law of Ampère11 :

I N
~ · d~s =
X
H Ii (4.10)
Γ i

In words, this relationship tells us that the integral along a closed path of integration Γ in the
magnetic field, is equal to the sum over the N enclosed currents ni Ii . The integration is performed
P

with the infinitesimal path element of integration d~s. Inside an infinitely long solenoid the field is
approximately homogeneous and parallel to the axis, while the field vanishes outside the solenoid.

l
D C H=0
ds

H
A B

Abbildung 4.6: Longitudinal section of the solenoid to aid in the calculation of the magnetic field
~ Shown is a closed rectangular path of integration Γ : ABCD.
H.

To calculate the magnetic field H inside the solenoid, the path of integration Γ should be chosen
to render the most convenient evaluation, in this case the rectangle ABCD in Fig. 4.6. The path
is assumed to enclose N turns of wire, each carrying the same current I. On the right of Eq. 4.10
the sum becomes N · I. On the left we evaluate the path integral over the four linear segments, and
readily find that only AB contribute: the field for the segment CD is zero and the segments BC
and DA are perpendicular to the field so that the scalar product H ~ · d~s vanishes. The left side of
Eq. 4.10 thus becomes H · l, and the magnetic field H inside the infinitely long solenoid:

N ·I
H= (4.11)
l

A, exact calculation of the field along the symmetry axis in a distance x from one end of the solenoid
of finite length l and radius R yields:
!
N ·I x l−x
H= √ +p (4.12)
2l x2 + R 2 (l − x)2 + R2

11
André-Marie Ampère (1775 – 1836), French physicist and mathematician.

Laboratory Manuals for Physics Majors - Course PHY112/122

4.4. APPENDIX xi

1.0

0.8 d Abbildung 4.7: Magnetic field strength with

H(x)/HC

respect to HC in the center of the solenoid as

0.6 function of distance x from one end of the sole-
H(x)/HC
integration: noid. The grey shaded area denotes the trajec-
0.4 G=0.99 tory of the electrons and, hence, the range of
integration for the calculation of the correction
0.0 0.1 0.2 0.3
distance x from the end of the solenoid (m) factor G.

In the present case, the length d of electron trajectory is short compared to the length of the
solenoid and close to the center of the latter, as depicted by the grey shaded area in Fig. 4.7. The
field variations felt by the electrons are, thereby, small with respect to the mean field strength. Thus,
the inhomoheneity of the magnetic field may be lumped into an average field strength H ≈ G · HC .
Calculating HC from Eq. 4.12 and G by integration over the path length d yields finally:
N ·I
H = G · HC = G · √ with G = 0.99. (4.13)
l2
+ 4R2

~ in vacuum is then obtained as B

The magnetic induction B ~ = µ0 · H,
~ using the vacuum permeability
12
µ0 (also known as the magnetic constant) .

V·s V·s
µ0 = 4π · 10−7 = 1.256637061... × 10−6 (4.14)
A·m A·m
The magnetic induction B in vacuum on the symmetry axis of a solenoid eventually becomes:

µ0 · G · N · I
B= √ (4.15)
l2 + 4R2

b) The magnetic induction of a pair of Helmholtz coils

Two coils of electrical wire may be arranged such that a nearly uniform magnetic field is attained
in the volume between them, a principle developed by the German physicist Hermann Ludwig
Ferdinand von Helmholtz (August 31, 1821 – September 8, 1894). More specifically, a Helmholtz
pair consists of two identical thin, but not necessarily circular coils, positioned symmetrically on
each side of the experimental area along a common axis, and separated by a distance d equal to
the radius R of the coils. The two coils are electrically connected in a manner such that the same
electrical current runs through both, and that the field of each has the same direction, Fig. 4.8
The magnetic induction in the region between the two coils may be calculated using the law of
Biot13 and Savart14 :
~ = µ0 · I · d~s × ~r .
dB (4.16)
4π r3
12
Alternative SI-Units are H m−1 , or N A−2 , or T m/A, or Wb/(A m)
13
Jean-Baptiste Biot (1774 – 1862) French physicist, astronomer, and mathematician
14
Félix Savart (1791– 1841) French physicist, professor at Collège de France in 1836

Laboratory Manuals for Physics Majors - Course PHY112/122

xii 4. Electron Charge-to-Mass Ratio

Helmholtz coils

R R
I I

α z
z=0 α
r dB
ds
d=R

Abbildung 4.8: The relevant geometry for a set-up with a Helmholtz pair of radius R and distance
d = R. The z axis defines the symmetry of the pair; the origin is in the centre of the left coil.

Finding a general solution valid for the entire volume is a rather involved procedure. We therefore
restrict the calculation to the component along the symmetry axis.
I ds
dBz = µ0 · · · cos α (4.17)
4 π r2
From figure 4.8 we find the relations r2 = R2 + z 2 and cos α = R/r. Integrating over the N turns
of the left coil, we obtain the contribution to the field from the left coil at position z on the axis of
symmetry:
N ·I R µ0 · N · I R2
Bl = µ 0 · · 2πR · 2 = · (4.18)
4π (R + z 2 )3/2 2 (R2 + z 2 )3/2
The analogous calculation of the contribution Br from the right coil in the same system of co-
ordinates requires that we replace z with d − z = R − z. The total field is then found by a
summation of the two contributions:
 
µ0 · N · I  R2 R2 
Bz = · 2 +  , (4.19)
 
2  (R + z 2 )3/2 (R2 + (R − z)2 )3/2 

which is shown in Figure 4.9 as a function of z.

Bz

B l + Br
Bl Br Abbildung 4.9: Magnitude of the field compo-
nent Bz between a Helmholtz pair, on the axis
of symmetry, according to Eq. 4.19. The fields
z
d=R Br and Bl are the contributions from the right
and the left coil, respectively.

For the midpoint on the symmetry axis between the coils, cf Fig. 4.8, we find, with z = d/2 = R/2
in Eq. 4.19:
µ0 · N · I 8
Bz (z = R/2) = · √ . (4.20)
R 5 5

Laboratory Manuals for Physics Majors - Course PHY112/122