Brownian motion

and molecular size

Counting and sizing molecules



Literature Study
K.L. Planken
May-July 2003


Supervisor:
A.P. Philipse







van t Hoff Laboratory for Physical and Colloid Chemistry




Summary
This literature review discusses the methods used for the first reliable estimate of
molecular size by Loschmidt in 1865 and the first size estimate of molecules using
Avogadros number obtained from measurements on Brownian motion by Perrin in
1908. The subject of this literature review is introduced by a brief description on the
evolved atomistic theories and the development of the kinetic gas theory as well as the
ideal gas law. The estimates by Loschmidt and Perrin rely on the kinetic gas theory,
but Perrin had a value for Avogadros number and did not need to define a
condensation coefficient like Loschmidt. Perrin employed Einsteins explanation of
Brownian motion for his measurements to obtain Avogadros number. The discovery
of this motion by Brown in 1827, Einsteins explanation, and Perrins experimental
study of phenomena due to this motion are explained in more detail. The estimate of
Avogadros number from measurements on Brownian together with Loschmidts size
estimate, which eventually leads to a value for Avogadros number, are compared to
the timing and accuracy of other determinations.




























Contents
Page

1. Introduction 4

2. A brief history

2.1 Small particles, molecules, and atoms 6
2.2 Earliest determination of molecular size 12

3. Robert Brown (1773-1858)

3.1 Biography 16
3.2 Discovery of Brownian motion 17

4. Albert Einstein (1879-1955)

4.1 Biography 21
4.2 Theory of Brownian motion 22

5. Jean Baptiste Perrin (1870-1942)

5.1 Biography 27
5.2 Perrins experiments 28
5.3 Statistical equilibrium in a column 31
5.4 Displacement in a given time 34
5.5 Values for Avogadros number found by Perrin 36

6. Closing 37

References 40

4
1. Introduction
Nowadays chemists and physicists are familiar with quantities as Avogadros
number and dimensions of molecules and atoms, as well as phenomena like Brownian
motion. As a consequence we do not ask ourselves how these quantities were
determined or measured after the underlying phenomena were discovered and
explained by theories. Maybe more striking is that we do not realise why these
quantities have the well-known values, and how the correctness of these values can be
proofed. In other words we do not doubt their validity despite our lack of awareness
of their origin. This is one of the reasons why this literature study deals with the
determination of Avogadros number and the magnitude of molecular or atomic size.
The counting and sizing of atoms and molecules in relation with Brownian motion is
the central subject in this literature study.
To introduce this study, a brief description of the genesis of the atom and molecule
will be given to illustrate how the concept of the atom did evolve. The concept of the
existence of atoms is crucial for the development of the kinetic theory for gases and
the explanation of the ideal gas law. The description of the kinetic theory for gases in
turn cannot be neglected because of the strong analogy of colloidal systems and gases.
Phenomena as Brownian motion are directly related to the laws evolving from the
kinetic theory for gases. In addition, as will be discussed, without the kinetic theory of
gases the first estimation of molecular size by Loschmidt could not have been
established. Once the molar volume is defined, Avogadros number becomes crucial
for an accurate determination of molecular size or mass. This number, which is the
number of molecules or atoms in a mole, can be determined in several ways.
How was the first experimentally accurate determination of Avogadros number by
measurements on Brownian motion accomplished, and how was the size of molecules
and atoms estimated using this number? The estimation using the determined value of
Avogadros number by Perrin to estimate the size of molecules is of course different
from the first size estimation by Loschmidt. In what sense are these estimations
different? The methods have similarities as will be pointed out in this literature study.
To answer most of these questions, the events that led to the understanding of
Brownian motion and the determination of sizes of molecules or atoms by
measurements on Brownian motion are described. The discovery and observation of
Brownian motion by Brown, theoretical prediction or description by Einstein, and
5
experimental proof and application by Perrin are major topics in this description.
Short biographies of Brown, Einstein, and Perrin are included to give an impression
of these people. The estimates of the value of Avogadros number by Perrin and by
Loschmidt are compared to other determinations of Avogadros number.
There is a lot of literature available dealing with the above-described subjects, and
an attempt to give a complete description and overview of the events in the history of
physical-chemistry science that were necessary for its development is not realistic.
6
2. A brief history

2.1 Small particles, molecules, and atoms [1-5]
The first scholars in ancient Greece, a few thousands year ago, speculated about
the external manifestations and the internal structure of matter. Democritus of Abdera
(approximately 460 BC-370 BC), a Greek philosopher, was teaching that everything
that exists, like earth, water, fire, plants, animals, and even the body and soul of man,
consists of tiny unchanging elements, which he called atoms ( = atomos =
indivisible). Apparently, Democritus had worked out the theory and ideas of
Leucippus (approximately 470 BC). Democritus ascribed all phenomena and changes
occurring in the world to their motion. The ancient atomists were purely speculative
because their conclusions were not confirmed by experimental observations. In
contrast to the opinion of Democritus that matter was not continuous but consisted of
atoms of various shapes moving spontaneously in a vacuum, Aristotles teaching was
that matter was infinitely divisible and that the vacuum and motion, supposed by
Democritus, did not exist. Aristotles point of view was later supported by the church,
and dominated the Middle Ages. Nevertheless, Democritus theory was not
completely abandoned, as seen by the extension of the theory by Epicurus (350-275
BC), and the return of the theory in writings of Lucretius (approximately 100-55 BC).
According to Lucretius atoms exist, but cannot be seen. Instead the existence of atoms
is manifested in the vivid motion of dust. This phenomenon is often by many people
misinterpreted and confused with the Brownian motion. The vivid motion of dust
particles in sunlight is caused by convection in contrast to the Brownian motion,
which is completely different as will be discussed later.
The ancient Greeks were not able to connect their concepts with evidences. It took
approximately 2000 years before the subsequent followers of the first atomists were
able to place the concept of discontinuity and indivisibility of matter upon a more
evidential basis. In fact, from the atomistic ideas of Pierre Gassendi (1592-1655) up to
the beginning of the twentieth century, the existence of atoms was still only a
theoretical concept, and this concept was kept alive to describe as many phenomena
as possible in a uniform way. Gassendi was able to explain many physical phenomena
based on the assumption that gases were composed of rigid moving atoms made from
similar substance but different in size and form. Usually Pierre Gassendi is considered
to be the father of the kinetic theory of gases. In 1658, Hooke expanded Gassendis
7
theory and explained the elasticity of gases, which is due to the impact of hard
independent gas particles on the material that surrounds the gas, according to Hooke.
The next advance was due to Newton, who proposed that the pressure of a gas was the
result of the repulsive forces between the molecules. This was later proven to be only
a secondary effect and subsequently Daniel Bernouilli (1700-1782) came up with a
major improvement to the theory. He described the air, in his Hydrodynamics, as an
elastic fluid, with particles, which are extraordinary fast moving in various directions,
which generates pressure by the impact on colliding to each other and to the wall that
surrounds them. With this model he derived the same law as discovered by Boyle
(1662) and Mariotte (1676). This law states that at a fixed temperature the product of
pressure and volume for a fixed amount of gas is constant (P 1/V). Almost a
century later Charless law (1787) or Gay-Lussacs law (1802) was established, which
states that the pressure of any gas in a fixed volume is directly proportional to the
absolute temperature of the gas. This law implies that the cubical coefficient of
expansion is the same for all gases. Combination of the laws of Boyle and Mariotte
and Gay-Lussac yields the ideal gas law. The discovery of the ideal gas law and the
theoretical explanation supported the existence of atoms. The next important step was
made by Dalton, which will be described in the following section.
John Dalton (1766-1844) created a theory of chemical reactions based on the
existence of atoms, which are indestructible and non-creatable. These atoms combine
and separate in definite proportions in chemical reactions. This idea made it possible
to deduce several laws as the conservation of matter (Lavoisier 1789), definite
proportions of chemical combination (Proust 1799), and multiple proportions of
chemical combination (Dalton 1789). These laws in combination with the fact that the
ratios of the volumes of gases that combine chemically are simple integers (Gay-
Lussac) and Avogadros hypothesis were essential in the discovery of the
stoichiometry of chemical reactions. Avogadro stated that equal volumes of gases
contain, at standard temperature and pressure, equal numbers of these molecules or
atoms. But Avogadro was not able to proof this, and therefore his statement became
known as Avogadros hypothesis. Without a value for Avogadros number only
relative atomic masses and no estimation for molecular size or volume from the ideal
gas law could be obtained. In fact, Amadeo Avogadro (1776-1856) solved the (last)
discrepancy assuming that the elementary particles of all matter are not indivisible
8
atoms as Dalton had assumed, but particles that are compounded of atoms in the form
of molecules. Due to Stanislao Cannizzaro (1826-1910) the molar volume of a gas
become defined, without any awareness of a value for Avogadros number [6]. By
1860 Cannizzaro defined the molar volume of gases by assuming that the atomic
weight for hydrogen is 1.0, and that hydrogen gas, as well as oxygen gas (assumed by
Avogadro), is made up of diatomic molecules. He also assumed that Avogadros
hypothesis was correct. The molecular formula of water is therefore H
2
O. From the
combining weight of oxygen in water, which is 8.0, it follows that the atomic weight
of oxygen is 16.0, and the molecular weight of O
2
is 32.0. If equal volumes of all
gases contain equal numbers of molecules or atoms (Avogadros hypothesis), then the
molecular weight M of a gas is proportional to the density D of the gas, which can be
described by the equation:

M kD = (2.1)

with k the proportionality constant which is the molar volume. Using hydrogen and
oxygen, Cannizzaro obtained the average value for the molar volume of a gas of
22.4 L, see table 1. In this way the mole is defined as the number of grams equal to its
molecular weight on Cannizzaros scale, which strongly resembles the one we use
today (with some improvements in accuracy). The situation with respect to what was
known and what was not known in the 19
th
century is schematically shown in
figure 1.

Table 1: The molar volume of a gas at standard temperature (273 K) and
pressure (1 atm.), which is implied by assuming that the atomic weight of H is 1.0
using equation 2.1.
Gas Density
g L
-1

Molecular weight
g mole
-1

Constant k
L mole
-1

H
2
0.0894 2.0 22.37
O
2
1.427 32.0 22.42
Average value: 22.4

9


Figure 1: Sketch of the situation in the 19
th
century. Only relative atomic
weights were known, as well as the stoichometry of numerous chemical
reactions. Absolute molecular weights (M
w
), diameter (d), and a value for
Avogadros number were not known in the 19
th
century.

Having discussed some important developments that supported the existence of
atoms, the development of the kinetic theory of gases will be described, which was
developed, amongst others, by Rudolf Julius Emanuel Clausius (1822-1888). His new
gas theory, which he called Kinetic, assumes that the gas consists of molecules
moving rectilinearly with a constant (equal) velocity that changes by collisions with
other molecules or an impenetrable wall. The gas pressure is explained by the impacts
of molecules on the vessels wall, which is mathematically expressed by the volume
of the gas and the internal energy. James Clerk Maxwell (1831-1879) suggested with
his hypothesis that the molecules or atoms of a gas do not have equal velocities, and
he introduced the distribution-law for velocities. With this formalism, a number of
impressive results were achieved, like the derivation of the ideal gas law
(P V = constant T), which already had been found empirically by Carnot (1824)
and Clapeyron (1834), and earlier by Boyle and Mariotte and Gay-Lussac.
As we saw in the previous section the kinetic theory of gases is still hypothetical
concept. The discovery of the Brownian motion, which was not related to the kinetic
theory at all, is in fact an experimental proof for the validity of the kinetic theory. The
thermal motion of particles, which are much larger than atoms or molecules, is called
Brownian motion after the botanist R. Brown. Brown discovered (in 1827) and
described the motion of small particles observed in his extensive studies on small
materials. This is the reason why Browns name became inextricably linked to the
10
incessant agitation of minute suspended particles (colloids, diameter of 1-1000 nm).
The discovery that higher temperature also led to more rapid Brownian motion lead to
the suggestion in 1877 that its cause lay in the "thermal molecular motion in the liquid
environment". The idea that molecules of a liquid or a gas are constantly in motion,
colliding with each other and bouncing back and forth, is a prominent part of the
kinetic theory of matter in explanation of heat phenomena. According to the theory,
the temperature of a substance is proportional to the average kinetic energy with
which the molecules of the substance are moving or vibrating. The average kinetic
energy can be described by the equation:

2
1
3
2
b
k T
mv
m
= , (2.2)

with m the particle mass, v the velocity, k
b
the Boltzmann constant, and T the absolute
temperature. Colloidal systems have strong analogies to ideal gases and therefore to
atoms and molecules because of the thermal motion. To make the description of the
development of the kinetic theory complete the final summation of the gas molecular
kinetic theory, which was the work of Ludwig Eduard Boltzmann (1844-1906), will
be described in the next section.
In the 1870s Boltzmann published a series of papers. He showed that the Second
Law of Thermodynamics, which concerns energy distribution, could be explained by
applying the laws of mechanics and the theory of probability to the motions of the
atoms. He made clear that the second law is essentially statistical and that a system
approaches a state of thermodynamic equilibrium (equal energy distribution
throughout) because equilibrium is overwhelmingly the most probable state in which
matter occurs. During these investigations Boltzmann formulated the general law for
the distribution of energy among the various parts of a system at a specific
temperature. He also derived the theorem of equipartition of energy, known as the
Maxwell-Boltzmann law. This law states that the average amount of energy used for
each different direction of motion of an atom is the same, with the root-mean-square
speed being equal to the root of (3 KT/m). The equipartition of thermal energy is in
fact the origin of Brownian motion. Boltzmann derived the equation for the change of
the distribution of atoms due to collisions and laid the foundations and much of the
structure of statistical mechanics. His work on statistical mechanics was strongly
11
attacked by many scientists that did not believe in the atomic theory and by scientists
who wanted to base all of physical science on energy considerations only. Scientists
who did not fully grasp the statistical nature of his reasoning misunderstood his ideas.
Boltzmanns conclusions were finally supported by the series of discoveries in atomic
physics that began shortly before 1900 and by the explanation of fluctuation
phenomena, as Brownian motion. These phenomena in the physical world could be
explained only by statistical mechanics.
In the previous sections features of the kinetic theory of gases are described.
Remarkable in these developments is that the ideal gas law in fact already proves the
equipartition of energy, because the molecular energy is proportional to the product of
volume and pressure. At a given density and temperature, the pressure is determined,
so for each kind of particle the average energy must be equal.
Until 1908 the notion of the particle structure of matter was nothing but a hopeful
hypothesis. The confirmation on the discrete or grained nature of matter (or the
existence of atoms and molecules) by the discovery, observation, theoretical
prediction, and experimental proof, for which atomic hypothesis sought in vain for
thousands years, is in fact established by Brown, Einstein, and Perrin.
Ultimately, after Perrins experiments, it became possible to estimate the size of
molecules and atoms with the determined value of Avogadros number. Nevertheless
there are earlier estimates of this value, but Perrin was the first who was able to
determine an accurate value by measurements on Brownian motion. To determine
Avogadros number it is first necessary to determine the characteristic size or mass of
a molecule, or the size or mass of colloids as Perrin did. Early estimates of values of
Avogadros number were made by calculations that estimated the number of
molecules in a given volume, based on estimates of molecular diameter, and mean
free path (Loschmidt 1865, described in section 2.2). In 1873 Maxwell estimated that
Loschmidt's number (the number of molecules in 1 cm
3
of gas at standard temperature
and pressure) was 1.910
19
cm
-3
. Given that a mole of ideal gas occupies 22,400 cm
3

at normal temperature and pressure, Maxwells result can be converted to a value of
4.410
23
mol
-1
for Avogadros number. Slightly later, Kelvin obtained an estimate of
the number of molecules in a gas from the scattering of light, and his data would give
a value of 510
23
mol
-1
.
Examples of later estimates were radioactive methods, used by a number of
investigators to directly count the number of alpha particles emitted from radium and
12
uranium. Rutherford and Geiger estimated, approximately somewhat later than 1908,
that Avogadro's number was 6.210
23
mol
-1
. Thin films of sodium oleate assumed to
be monolayers on the surface of water led Lecomte du Nouy (1924) to estimate the
size of a molecule, and from that the number of molecules in a mole as 6.00410
23

mol
-1
in 1924.
Modern methods to determine Avogadro's number rely on the use of x-ray
crystallography to get precise dimensions in crystals. These can produce extremely
precise values of Avogadro's number (N
AV
= 6.022136710
23
mol
-1
), with a relative
uncertainty of less than 1.010
-6
.
The earliest estimation of the size of molecules without having a value for N
AV
is
described in the following section. The subsequent chapters after this description
describe the discovery and theoretical description to obtain from the application of
Brownian motion a value for N
AV
, which enables an estimation of molecular size.

2.2 Earliest estimation of molecular size [7, 8]
Johann Josef Loschmidt made the earliest determination from experimental data of
the size of molecules and atoms in 1865, but how did Loschmidt estimate the
molecular size? To answer this question it is important to realise what model he used
to determine the sizes of atoms. At the time of Loschmidt it was known that in gases
molecules are separated by distances that are very large compared to their diameters,
and it was accepted to neglect the volume of the particles and to describe them as
point masses. The velocity of the molecules, which are constant in motion, is
governed by the temperature. The model Loschmidt used describes molecules or
atoms, which are translating in a straight line without interaction with each other,
except at the moment of collision. This model was developed by Herapath (1821),
Waterston (1845), Joule (1848), Rankin and Krnig (1856), and later on by Clausius
and Maxwell (1857). The studies of these workers allowed precise and lucid
explanation of the principal properties of gases, such as atmospheric pressure, heat
conductivity, and the propagation of sound. But also the numerical determination of
important constants as the average velocity of the molecules for different gases at
different temperatures, and the relationship of the total kinetic energy of a gas
corresponding to that velocity. Both concepts were established by the work of
Clausius and extended through the mean free path of air molecules according to
Maxwell and O.E. Meyer. Loschmidt was able to obtain, with this theory, a value for
13
another constant namely the magnitude of the diameter of air molecules, assuming
that the molecules are spherical and do have a volume. For this purpose he applied an
equation due to Maxwell, and slightly modified by Clausius. This equation, which
strongly resembles the equation for the total surface of the spheres of impact
(compare 2.2 with 5.3), has the form:

2
4
1
3
N Ls = , (2.2)

with N the number of air molecules in a unit volume, L the mean free path between
collisions, and s the molecular diameter. This expression is rearranged to give the
relationship between molecular volumes:

( )
2
1
5
1
3
4
Ls
N

= ,
(2.3)

Loschmidt assumes that Avogadros hypothesis is valid, and therefore he states that
gaseous phases contain equal numbers of molecules in a unit volume at equal
temperatures and pressures. Because N is the number of molecules, 1/N is the
(molecular) volume occupied by each molecule if the molecules are equally
distributed. The quantity ( L s
2
)/4 describes the volume of a cylinder whose base is
the circular cross section with diameter s and whose height is the mean free path of
the molecule. The cylinder represents the total volume that the molecule successively
occupies as it travels along an average path. This cylinder is 5 1/3 times greater than
the molecular volume (equally true for all gases). The volume of a single molecule is
( s
3
)/6, and 1 is the unit volume that N molecules occupy. The volume that N
molecules occupy with their masses when not in motion, is (Ns
3
)/6. Loschmidt now
defined the ratio of (Ns
3
)/6 to unity as the condensation coefficient of the gas. The
equation for the molecular diameter has then the form:

8 s L = (2.4)

This equation shows that the diameter of a gas molecule is equal to eight times the
14
mean free path multiplied by the condensation coefficient. The much smaller volume
that a specified number of molecules occupy in the liquid state than in a gas is
indicated by the condensation coefficient. The molecules are assumed to be spherical
in the estimation of the condensation coefficient, the small empty spaces between
them when they are in the liquid state, was not taken into account in the estimation of
the condensation coefficient. The volume of a specified number of spherical
molecules occupied depends on the packing fraction which in turn depends on the
symmetry of the layers in a packing and on the temperature for fluids who expand
upon temperature raising. Loschmidt treated the packing fraction as a ratio inversely
expressed compared to the nowadays definition of packing fraction. For spheres the
close-packed lattice of identical spheres is 0.740, whereas Loschmidt used the value
of 0.855 noted as 1:1.17. However this had no important effect on his estimation of
the size of molecules. The observed condensation behaviour, according to Loschmidt,
is accurate enough when dealing with the question of general magnitudes. In the
original article of Loschmidt specific volumes for almost all liquids for which
observations were available, and which only contain oxygen, hydrogen, nitrogen,
carbon, sulfur, chlorine, bromine and iodine. With a few exceptions, these were taken
from the work of H. Kopp.
To estimate the size of an air molecule Loschmidt calculated the condensation
coefficient of air (using empirical relations), and obtained a value of 0.000866, using
the packing fraction in his notation of 1:15 (0.87) for air that air is 770 times lighter
than water in the gaseous from. He used the average value of 140 millionths of a
millimetre for the mean free path for molecules in air obtained from the work of
Maxwell, O.E. Meyer, and (experimental) work by Bessel. The latter scientists
obtained values for the mean free path from the intrinsic viscosity of air determined
by Stokes. On this basis Loschmidt calculated the size of an air molecule as:

8 0.000866 0.00014 mm 9.69 A s = =


(2.5)

This result is of the right order of magnitude and is less than a factor ten away from
the todays accepted values of collision diameters.
The estimation of the size of molecules enabled Loschmidt to calculate the number
of molecules in a cubic centimetre (~ 1.910
19
cm
-3
). In fact this number can be
15
computed to obtain a value for Avogadros number. Because Loschmidt devised a
method to obtain a value for the number of molecules in a certain volume,
Avogadros number is called Loschmidts number in German speaking countries.
16
3. Robert Brown (1773-1858)
3.1 Biography [9]

Figure 2: Robert Brown [10]

Robert Brown (figure 2) was born in Montrose, Scotland, in December 21, 1773.
His father was a Scottish Episcopalian minister who had a strong and independent
mind. Maybe Robert Brown inherited a similar intellectual strength, but he did not
inherit his fathers Christian dogmatism. Robert Brown was educated at the Marishal
College, Aberdeen, and after which he studied medicine at the University of
Edinburgh. In 1795, by the age of twenty-one he had joined the Fifeshire Regiment of
Fencibles as Ensign and Surgeons Mate, which was soon posted to Ireland. Robert
Brown used much of his time to study. On an ordinary day he usually studied German
grammar before breakfast, and after the meal he worked on botanical documents until
lunchtime. In the afternoon he would see patients and in the evening, if he was not
socialising or out to dine, he would continue his scientific work until midnight.
In October 1798, the young officer Brown was in London to recruit for the
regiment. He was introduced to the eminent botanist Sir Joseph Banks as a
Scotchman, fit to pursue an object with constancy and a cold mind. Within two
years, Banks was planning a voyage of discovery to the new territories that are now
known as Australia. Banks chose the botanist Robert Brown to join the voyage. On
December 8, 1801 they arrived in Australia, which was in that time called the new
territories of New Holland. They arrived on King George Sound, the south-western
corner of the great subcontinent. Within three weeks Robert Brown had collected
17
more than 500 species of plants, almost all of them were unknown to western science.
After staying for three months in Port Jackson and ten months on the island Tasmania
they returned to England in October 1805 with vast collections of drawings and notes,
many zoological specimens, and with nearly 4,000 different species of plants. Brown
devoted the next five years to describe 2,200 of the species, of which approximately
1,700 were previously unknown. He nominated 140 new genera. From 1806 till 1822
Robert Brown served the Linnean Society of London as a Clerk, Librarian, and
Housekeeper. He took on Banks home and collections in Soho Square when Sir
Joseph Banks died on June 19, 1820. The specimens of these collections were
transferred to the British Museum on the condition that Robert Brown remained as the
curator for life of these collections. This was an important event in the establishment
of the great London collections, which have since become so important in the world
of taxonomy.
Robert Brown had been elected to the fellowship of the Royal Society in 1810, and
became a Fellow of the Linnean Society in 1822. He was president of the Linnean
Society from 1849 to1853. He died in London on June 10, 1858, just a week before
Darwin received Wallaces paper on the theory of survival of the fittest. The date of
Browns death ultimately led to the availability of a free date at which Darwin might
present his own lecture on the theory of evolution to the Linnean Society.

3.2 Discovery of Brownian motion
Although Brown practised medicine as surgeon for five years, he later abandoned
this and turned his efforts towards botanical science [11]. His investigation concerned
the precise nature and structure of pollen, and to inquire the mode of action of pollen
on the pistillum in phaenogamous plants. The general notion of the sexuality of
flowering plants had been accepted from the end of the seventeenth century, when
scientists such as Nehemiah Grew [12] and Camerarius [13] had shown that
pollination was essential to the production of seeds. Despite the system of
classification on the basis of the sexuality of plants due to Linnaeus, the mechanism
of fertilisation remained uncertain and had to await the development of the compound
achromatic microscope objective by Selligue and Amici in the 1820s [14]. Brown,
however, performed his investigations with a simple microscope, all with one and the
same lens of which the focal length was about 0.8 mm [15] which has a magnification
of approximately 400 [16]. Brown obtained this double convex lens from Mr. Banks.
18
Mr. Dollond made a simple pocket microscope of it [16], with very a delicate
adjustment. The instrument with which Robert Brown studied Brownian Movement
and which he used in his work to identify the nucleus of the living cell [11] is shown
in figure 3. This instrument is preserved at the Linnean Society in London. It is made
of brass and is mounted onto the lid of the box in which it can be stored. For many
years it was believed that Brown could not have seen the nucleus with such a
primitive instrument.


Figure 3: Browns microscope (see text above)
with a flower of the Clarkia pulchella [17].

To give greater consistency, and to bring the subject as much as possible within the
reach of general observation, Brown continued to employ throughout his whole
inquiry the same lens, despite his possession of two more lenses of much higher
power, and which he also used to make his investigations. Before he started his first
investigations on particles contained in the pollen of the plant Clarkia pulchella,
Brown had already an impression from an earlier observation of the particles
contained in pollen. These particles were not uniform in size but were spherical,
cylindrical to oblong shaped. His aim was to investigate the mechanism of
fertilisation by tracking oblong shaped particles, which could be easily distinguished
from the mainly spherically shaped particles. Brown took grains of pollen from the
Clarkia pulchella before they were bursted. The particles were of unusual large size,
19
according to Brown, varying from 5 to 6 m, and cylindrical to oblong shaped. A
remarkable thing is that Brown estimated the dimensions of the particles he studied
down to approximately 1 m with a surprisingly accuracy [18]. While he was
examining the form of these particles immersed in water he observed that many of the
particles were evidently in motion. Brown also mentioned that the form of particles
was changing, this could be due to the rotational diffusion of cylindrical or oblong
shaped particles.
Remarkable is that many articles concerning Brownian motion contain a plainly
erroneous reference by stating that the pollen themselves were observed to execute the
motion, although in fact these are much too large to show the erratic motion [18]. The
pollen grains themselves are approximately 10
-2
cm. The root-mean-square
displacement of these particles in water at 298 K over a time period of 25 hours,
assuming that the particles are spherical and that the stokes-drag holds, is only 19 m.
The time to travel the same distance by particles obtained from the pollen grains
investigated by Brown, which have a diameter of approximately 5 m, is
approximately an half hour. In contrast, the smallest particles observed by Brown
were approximately 0.8 m. Particles of this dimension dispersed in water at
298 K, have a root-mean-square displacement of 19 mm in less than 12 minutes [19].
After frequently repeated observation Brown concluded that the motion arose
neither from currents in the fluid, nor from the gradual evaporation of solvent, but that
the motion belonged to the particles themselves. He was convinced that it was not life
itself at work as a voluntary motion of living micro-organisms (animalcules)
observed by earlier microscopists like Antoni van Leeuwenhoek [19]. Brown
investigated grains of pollen from the same plant immediately after bursting. From the
pollen he obtained similar sub-cylindrical particles in reduced numbers, which were
mixed with other particles of much smaller size and apparently spherical. These
smaller particles, which Brown termed molecules showed very rapid oscillatory
motion when dispersed in a fluid. Darwin found in a later study on Brownian motion
that smaller particles move faster [20]. This was an early indication that the viscous
drag force is proportional to the inverse of the dimension of the particle. Brown
extended his investigations and observed the vivid motion of pollen grains of many
other plants. His next step was to determine whether this motion is restricted to
material of living material. As he said: Having found motion in the particles of the
20
pollen of all living plants which I had examined, I was led next to inquire whether this
property continued after death of the plant, and for what length of time it retained
[15]. He found from studies on Mosses and Equiseta, which had been dried upwards
of one hundred years, that also particles of non-living material were subjected to the
erratic motion when immersed in a fluid. Brown also examined various products of
organic bodies, particularly gum resins, substances of vegetable origin, and even pit-
coal. In all these bodies he found his so-called active molecules in abundance. He
inferred that these molecules were not limited to organic bodies, nor even to their
products. The next object of inquiry was to ascertain that molecules existed in
mineral bodies. Brown examined window-glass, minerals (most remarkable are
travertine, stalactites, lava, obsidian, pumice, volcanic ashes, and meteorites) and
metals (such as manganese, nickel, plumbago, bismuth, antimony, and arsenic), rocks
of all ages, and even a fragment of a Sphinx to be sure that it is not living material. He
obtained from these materials molecules agreeing in size, form, and motion with those
he already had seen.
Brown observed the thermal motion of small particles and contributed in this way
to new insights in atomic and molecular existence. But brown was not able to explain
the vivid motion of small particles, although he was sure that the motion of particles
was not a consequence of exterior driving forces. He reported several experiments to
exclude external driving forces. One such experiment to exclude evaporation of fluid
was the immersing of fluid droplets in almond oil [19]. Many scientists were not
convinced by his experiments and they still suspected that the Brownian motion arises
from particle interactions, air currents, mechanical vibrations, surface tension effects,
electricity or magnetism, and local convection in the fluid brought about by minute
temperature differences [16]. A satisfactory explanation was not possible until 50
years later the kinetic molecular theory of matter became established. Thermal motion
and the existence of atoms or molecules became firmly established allmost 80 years
later, when Einstein formulated his theory for diffusion and the so-called Brownian
motion which was experimentally confirmed by experiments of Perrin.



21
4. Albert Einstein (1879-1955)
4.1 Biography [21, 22]


Figure 4: Albert Einstein [23].

Albert Einstein (figure 4) was born on March 14, 1879, in Ulm, Wrttemberg,
Germany. After public school in Mnchen and in Aarau (Switzerland), Einstein
studied mathematics and physics at the Swiss Polytechnic Institute in Zurich. From
1902 to 1909, he worked as an examiner at the Swiss Patent Office in Bern,
Switzerland. Away from work, he continued his discussions on scientific matters with
colleagues including his first wife Mileva Maric.
Maric and Einstein entered the Swiss Polytechnic Institute in Zurich in 1896 to
study mathematics and physics. Both graduated in 1900. Einstein married Mileva
Maric on January 6, 1903. They had two sons, and one daughter.
Einstein became a Swiss citizen in 1905. Einstein became professor of theoretical
physics at the University of Zurich in Switzerland in 1904. In the years 1911 and
1912, he had the same position at the German University in Prague. Einstein was
elected to the Prussian Academy of Science in Berlin in 1913. When he accepted the
professorship of physics at the University of Berlin in 1914, he once more assumed
the German citizenship. In the same year, he became director of the Kaiser Wilhelm
Physical Institute in Berlin. He occupied both positions until 1933. In1921 Einstein
became a Nobel-Prize laureate for his achievements in Theoretical Physics, and
especially for his discovery of the law of the photoelectric effect.
22
Einstein and his relatives left Germany and immigrated to USA on December 9,
1930 after the rise of NAZI regime in Germany. In 1933, the Nazi government took
his property and deprived him of his positions and citizenship. Einstein and his
relatives arrived in California in early 1931. Later he was invited to join the Institute
of Advanced Study in Princeton, New Jersey, and he accepted the position. He lived
and worked there until his death. In 1940, Einstein became an American citizen.
Albert Einstein died in Princeton, New Jersey on April 18, 1955.
Einstein was one of the fathers of the atomic age, and he was one of the great
scientists of all time. In the 1905, the annus mirabilus, Einstein contributed three
papers to the German scientific journal Annalen der Physik. Each of the papers
became the basis of a new branch of physics. The first paper that was published was
about the photoelectric effect [24]. In the third published paper, Einstein formulates
the Special Theory of Relativity. He established the law of mass and energy
equivalence through his famous formula E = m c [25]. The second paper on statistical
mechanics, which was published in May 1905, describes the derivation of the
diffusion coefficient and the expression for the root-mean-square displacement of
species [26]. This paper concerned the movements of tiny particles floating in a liquid
or gas.

4.2 Theory of Brownian motion
As already described, Einstein produced his statistical mechanical quantitative
theory of Brownian motion in 1905. Almost at the same time similar studies were
carried out on Brownian motion by the Polish physicist Marian Ritter von Smolan-
Smoluchowski, who independently discovered the same results. But Smoluchowski
used somewhat different methods than Einstein. If Smoluchowski had been only an
outstanding theoretical physicist and not a fine experimentalist as well, he would
probably have been the first to publish a quantitative theory of Brownian motion [22].
The existence of atoms, in the time Einstein wrote his paper, was still uncertain.
Einstein even had doubts about the validity of either applicability of the classical
thermodynamics on small particles or the molecular kinetic theory of heat. This doubt
is illustrated by one of the first sentences Einstein wrote in his paper: Wenn sich die
hier zu behandelnde Bewegung samt den fr sie zu erwartenden Gesetzmigkeiten
wirklich beobachten lt, so ist die klassische Thermodynamik schon fr
mikroskopisch unterscheidbare Rume nicht mehr als genau gltig anzusehen und es
23
ist dann eine exakte Bestimmung der wahren Atomgre mglich. Erwiese sich
umgekehrt die Voraussage dieser Bewegung als unzutreffend, so wre damit ein
schwerwiegendes Argument gegen die molekularkinetischen Auffassung der Wrme
gegeben [26]. Einstein wrote later that his major aim was to formulate statistical
thermodynamics and to find facts that would guarantee, as much as possible, the
existence of atoms of definite size. In the midst of this work, he discovered that,
according to atomistic theory, there would have to be an observable movement of
suspended microscopic particles. Einstein claimed that he couldnt connect the
observations concerning Brownian motion because these are very imprecise according
to Einstein. He wrote in his paper: Es ist mglich, da die hier zu behandelnden
Bewegungen mit der sogenannten Brownschen Molekularbewegung identisch sind;
die mir erreichbaren Angaben ber die letztere sind jedoch so ungenau, da ich mir
hierber kein Urteil bilden knnte [26].
In his paper ber die von der molekularkinetischen Theorie der Wrme
geforderte Bewegung von in ruhenden Flssigkeiten suspendierten Teilchen [26],
Einstein first shows that van t Hoffs law for the osmotic pressure (1884) is a direct
consequence of the molecular kinetic theory of heat, and also holds for diluted
microscopic particles. He proceeds with the derivation of the diffusion coefficient for
spherical particles. To derive this diffusion coefficient he defines a force balance in
which a certain force K, depending on the position of the particle, balances the motion
of particles due the osmotic pressure. This balance has the form:

0
p
Kv
x

(4.1)

with the number density of particles, x the position of the particle, and p the osmotic
pressure which is defined by (van t Hoffs law):

pV RTz

= . (4.2)

In this equation V* is a unit of volume, R is the gas constant, T the absolute
temperature, and z the amount of non-electrolyte gram-molecules. In the next step of
this derivation Einstein concerned spherical particles. He was thereby able to define
24
the velocity due to K of a single spherical particle with radius P, in a fluid with
viscosity k:

6
K
kP
. (4.3)

Equation 4.3 contains the Stokes friction coefficient for a sphere, derived by the Irish
physicist-mathematician Sir George Gabriel Stokes (1819-1903) [27]. With equation
4.3 Einstein formulated the amount of particles passing a cross-section unit per unit of
time:

6
vK
kP
. (4.4)

This flux of particles is balanced by the opposite flux due to diffusion. The form of
this balance is, according to Einstein:

0
6
vK v
D
kP x

=

(4.5)

From equation 4.1 and 4.5, the expression describing the Stokes-Einstein diffusion
coefficient can be obtained (using van t Hoffs law and eliminating K):

1
6
RT
D
N kP
= , (4.6)

with N the number of molecules in a gram-molecule.
After obtaining expression 4.6, Einstein proceed his paper with the derivation of
the root-mean-square displacement of suspendierten Teilchen (suspended particles)
in a fluid. For simplicity the detailed derivation will not be given, but may be found in
[26]. In the derivation, Einstein makes use of the expression of the probability finding
a particle at a certain time t and position x with respect to x at t = 0 (Gaussian
distribution), and of Ficks second law (describes the relation of particle concentration
25
and diffusion). The expression obtained in this derivation from which the root-mean-
square displacement is obtained has the form:

( ) ( )
-
, 0 and , f x t f x t dx n
+

= =

,
(4.7)

with f (x,t) the function which describes the amount of particles per unit volume, and
n the number of particles. From expression 4.7 together with Ficks law:

2
2
f f
D
t x

=

, (4.8)

with D the diffusion coefficient, and t the time, Einstein obtained the following
expression:

( )
2
4
,
4
x
Dt
n e
f x t
D t

= .
(4.9)

From expression 4.9 the root-mean-square displacement can be calculated:

2
2 x Dt = . (4.10)

This expression holds for the displacement along the x-axis, but can be easily
converted to 3-dimensional space, which has the form:

( )
2
, , 6 r x y z Dt = . (4.11)

Expression 4.10 and 4.11 hold for ideal particles, because in the derivation inter-
particle interactions are neglected.
Einstein created with his paper the basis for modern colloid science without
knowing it at the time he wrote it. With equations 4.6 and 4.11 it became possible to
estimate molecular dimensions if diffusion or displacement of colloids would be
26
experimentally accessible. Up to this time it seemed not possible to perform
measurements on Brownian motion, and especially to interpret these measurements
until Einstein wrote his paper. Brownian motion is a typical phenomenon on which
one cannot measure anything significantly without knowing the theory, which tells
you what to measure [1].
He concludes his paper by mentioning his hope for a confirmation by a scientist to
answer the for the theory of heat important question, according to his own words:
Mge es bald einem Forscher gelingen, die hier aufgeworfene, fr die theorie der
Wrme wichtige frage zu entscheiden!

27
5. Jean Baptiste Perrin (1870-1942)
5.1 Biography [28-30]

Figure 5: Jean Baptiste Perrin [28].

Jean Baptiste Perrin (figure 5) was born in Lille, France, on September 30, in 1870.
Perrin was educated at the cole Normale Suprieure in Paris, where he later became
an assistant in physics from 1894 till 1897. In the same time he started his research on
cathode rays and X-rays. In 1895 he established that cathode rays are negatively
charged particles (electrons). His attempt to determine the mass of these particles was
soon anticipated by the work of J.J. Thomson. Perrin received his degree of docteur
s sciences" in 1897 for a thesis on cathode and Rntgen rays. He was, in the same
year appointed to readership in physical chemistry at the Sorbonne University of
Paris. In 1898 he joined the faculty of the University of Paris (1898) where he became
professor of physical chemistry in 1910, a post which he held till 1940. Perrin was an
officer in the engineer corps during the First World War in 1914-1918. When the
Germans invaded France in 1940 he escaped to the U.S.A., where he died in 1942 on
April 7. After the War, in 1948, his remains were transferred to France by the
battleship Jeanne d'Arc, and he was buried in the Panthon.
Perrin was the creator of the Centre National de la Recherche Scientifique, an
organisation offering to most promising French scientists a career outside the
University, whose scientific talents would otherwise be lost. In addition, he founded
the Palais de la Dcouverte (Palace of discovery). Perrin was also responsible for the
establishment of the Institut d'Astrophysique, in Paris, and for the construction of the
28
large Observatoire de Haute Provence. Without his prestige and his power of
persuasion the Institut de Biologie Physico-Chimique would never have come into
being.
His earliest work on cathode rays and their nature was proved by him to be that of
negatively charged particles. He also studied the effect of the action of X-rays on the
conductivity of gases. In addition, he worked on fluorescence, the disintegration of
radium, and the emission and transmission of sound. Perrin is best known for his work
and study on colloids and, in particular, on the so-called Brownian movement, which
he started in 1908. His results in this field confirmed Einstein's mathematical analysis,
in which it was shown that colloidal particles should obey the gas laws, and thereby
enable to calculate Avogadro's number N
AV
, the number of molecules per gram
molecule of a gas. His observations also enabled him to estimate the size of water
molecules and atoms as well as their quantity in a given volume. This was the first
time the size of atoms and molecules could be reliably calculated from actual visual
observations. Perrin's work helped to raise atoms from the status of useful
hypothetical objects to observable entities whose reality could no longer be denied. In
1926, Perrin was rewarded with a Nobel Prize for his work on the discontinuous
structure of matter and the discovery of measurements on the equilibrium
sedimentation.

5.2 Perrins experiments [30, 31]
The introduction of the ultramicroscope in 1903 aided quantitative studies by
making visible small colloidal particles whose greater activity could be measured
more easily. Several important measurements of this kind were made from 1905 to
1911. At this time there were still people, among them V. Henri, who strongly
doubted the correctness of Einsteins theory because they, next to other facts, assumed
that this theory rested upon some unsupported hypotheses. Perrins reaction was, in
his own words: I am convinced by this of how limited at the bottom is our faith in
theories: we regard them as instruments useful in discovery rather than actual
demonstrations of fact. In the early years of the 20
th
century, Jean-Baptiste Perrin
was successful in verifying Einstein's analysis. As a matter of fact, after the
completion of the first series of measurements on displacements it became clear that
Einsteins formula is accurate. His work established the physical theory of Brownian
29
motion and more or less ended the scepticism about the existence of atoms and
molecules as actual physical entities.
For his studies on Brownian motion Perrin used various colloidal systems, most of
them were emulsions. Attempts using the usual colloidal solutions like arsenic
sulphide and ferric hydroxide were unsuccessful. Perrin was able to obtain emulsions,
which were composed of gamboge and mastic. Gamboge, also spelled Camboge, is a
hard, brittle gum resin prepared from dried vegetable latex from various Southeast
Asian trees (genus Garcinia). Gamboge is used as a colour vehicle and in medicine. It
is orange to brown in colour and when powdered turns bright yellow. Artists use it as
a pigment and as a colouring matter for varnishes. In medicine and veterinary
medicine it is a drastic cathartic, and it strongly irritates the skin [32]. Mastic, also
spelled Mastich, is an aromatic resin, obtained as a soft exudation from incisions in
mastic trees. It is used chiefly to make pale varnishes for protecting metals and
paintings. When dispersed in bodied (thickened by heating) linseed oil, mastic is
known as megilp and is used as a colour vehicle. Mastic is also used as an adhesive in
dental work [33]. Perrin prepared gamboge water emulsions, by completely dissolving
the gamboge upon addition of alcohol. A yellow emulsion composed of tiny gamboge
spheres was obtained by adding water to the alcohol gamboge solution. Mastic
appeared to be the most suitable resin to obtain a stable colloidal system using the
same preparation method. Mastic alcohol dispersions give milky emulsions,
composed of a colourless, transparent, and glassy substance when water is added.
The emulsions Perrin obtained contain grains of various sizes (polydisperse), and
Perrin realised that uniform emulsions of equal sized grains were required for his
measurements. Perrin used centrifugation to separate the unequal sized grains to
prepare emulsions of uniform sized grains. This process resembles fractional
distillation. After several months of repeated centrifugation a few tenths of a gram of
grains was obtained, having the same diameters approximately equal to the diameter
Perrin wished to obtain, out of a kilogram of grains. With these emulsions of low
polydispersity (best emulsions had a polydispersity within one percent) Perrin aimed
to determine Avogadros number. The well-known measurements Perrin performed,
are the determination of the vertical distribution of grains, the translatory
displacements, rotational diffusion, and diffusion measurements. As will be pointed
out in the description of measurements on the vertical distribution of grains and on the
translatory displacements, much of the system, as the density of the grain and the
30
radius of the grains, has to be known to be able to determine Avogadros number.
Perrin determined the density of the grains as well as the radius of the grains, each by
three different methods. One determination of the density was done by using the
specific gravity bottle method. The masses of water and emulsion that fill the same
bottle are measured. In addition the mass of the suspended resin is determined after
dissication in the oven. A viscous liquid is obtained by drying an emulsion of
gamboge at 110C, which undergoes no further loss of weight on elongated storage in
the oven. Other density determinations were the determination of the density of the
glassy substance obtained from a dried emulsion (this substance is probably identical
with the material of the grains), and energetic centrifugation of an emulsion after
adding potassium bromide (density gradient centrifugation). To determine the volume
of the grains Perrin did direct measurements of the radius of the grains, and he
determined the radius by an application of Stokes law (measurement of the
sedimentation velocity from which the radius can be calculated using the Stokes
friction factor). An example of the direct measurement of the radius is shown in
figure 6. Perrin considerably minimised the error of the determined radius of single
grains due to the diffraction that occurs in magnified images, by measuring the length
or surface area of respectively a row or area of a known number of grains. He
obtained these rows and areas by nearly complete evaporation of the fluid. The
capillary forces cause the grains to run together and thereby forming rows or areas,
see figure 6.

Figure 6: Microscopic image of one of Perrins suspensions, dried
on a microscope slide to determine the radius of the particles [34].

The result of the radius determination of his best emulsion was 0.373 m, which is the
31
average value of 50 rows of 6 to 7 grains. In comparison, Perrin obtained for the same
emulsion a radius of 0.369 m by measuring the area of 2000 grains distributed over
10
-5
square centimetres.

5.3 Statistical equilibrium in a column
It is well known that the air is more rarefied in the mountains than at sea level, and
that, in general terms, any vertical column of gas is compressed under its own weight.
Perrin used the relation between altitude and barometric indications, which were
worked out by Laplace. He realised that in the emulsions he studied, equilibrium is
established between the opposing effects of gravity, which pulls the particles
downwards, and the Brownian movement, which tends to scatter them. In fact this
equilibrium distribution is nothing but a Boltzmann distribution (barometric height
distribution). By introducing the gravitational characteristic length it can be easily
shown that the number density decrease upon elevation of a gas of small gram
molecular weight is less, compared to a gas of larger gram molecular weight. The
characteristic gravitational length is:

g
l
B
k T
mg
= ,
(5.1)

with l
g
the length over which the density decreases by 37%, g the gravitational
constant, k
B
the Boltzmann constant, T the absolute temperature, and m the mass of a
gas molecule. Perrin illustrated this phenomenon by a schematic representation of
three columns shown in figure 7.

Figure 7: Three gigantic vertical columns each filled with same number of molecules of
(from left to right) hydrogen, helium, and oxygen. The largest column is 300 kilometres [31].
32
The originality of Perrins approach lay in his insight to a possible analogy
between the behaviour of gas molecules and colloid particles, which would lead to a
similarity in experimental phenomena and allow a clear derivation of fundamental
values of kinetic molecular theory. Leon Gouy and others also suggested this, but
their experimental attempts were not successful. Perrin suspected that the phenomena
of the height distribution of gasses and the implied arguments were also applicable to
emulsions (colloids) if they obey the gas laws. As Perrin said: We shall thus be able
to use the weight of the particle (grains), which is measurable, as an intermediary or
connecting link between masses on our usual scale of magnitude and the masses of
the molecules. The equation Perrin used to determine the value of Avogadros
number has the form:

( )
0
ln
AV
RT n
N
V D d gH n

=

(5.2)

with n the number density of grains at height H, n
0
the number density at height
H = 0, D and d the mass density of respectively the grains and liquid, T the absolute
temperature, R the gas constant, N
AV
Avogadros number, V the volume of a grain,
and g the gravitational constant. Perrin had overcome the difficulty of counting grains
at various height levels using a delicate method. His first attempts were to make
instantaneous photographs to count the grains from these images. But an intense light
is needed because of the magnification and short time of exposure, and consequently
he obtained never good images of grains with a diameter of half a micron. He
therefore reduced the field of vision by placing in the focal plane a diaphragm
consisting of an opaque disc of foil having a very small round hole pierced in it by a
needle. The visible field becomes very restricted, and enables to estimate by eye at
once the exact number of grains seen at any given moment. The number of grains at
regular time intervals at a certain height level was established by placing a shutter in
the path of the rays that illuminate the preparation. A cross section in which the
different levels are indicated of an emulsion as studied by Perrin is shown in figure 8.
Note that the height distribution is similar to the distribution shown in figure 7.
Perrin took typical cross readings in a cell at equidistant planes within 100 m. At
each level 200 readings were taken and averaged. One careful measurement reported
33
in Les Atomes was a counting of grains (radius of 0.212 m) at different levels to a
total amount of 13,000 grains. Perrin calculated Avogadros number using equation
5.2 and the determined number densities at different levels as well as the determined
density and volume of the grains.
A B C
Figure 8: (A) Cross section of an emulsion of resin in water. The
diameter of the grains is about one micron [34]. (B) Spheres of gamboge
(a = 0.29 m) with levels that are 10 m apart. (C) Spheres of mastic
(a = 0.52 m) at levels that are 12 m apart [30, 31, 35].

The influence of temperature was also investigated, and Perrin reports that the
distribution is expanded when the temperature is increased. This effect proved,
according to Perrin, that Gay-Lussacs law applies to emulsions.
In the context of exact determinations of molecular magnitudes Perrin said, By
studying emulsions we are really able to weigh the atoms and not merely estimate
their weights approximately. Once Avogadros number is determined the mass and
or volume of atoms and molecules can be calculated. The mass of a hydrogen
calculated by Perrin, with a value of 6810
22
mol
-1
for Avogadros number from the
equilibrium measurements, is 1.4710
-24
g, which is close to the actual value of
1.6610
-24
g.
Perrin was able to calculate the diameter of the molecules sphere of impact, using
Avogadros number and Clausius equation, corrected by Maxwell, for the total
surface of the spheres of impact of N molecules in a gram molecule (mole):
34
2
' 2
v
ND
L
= ,
(5.3)

with D the radius of the sphere, v the volume of a gram molecule (molar volume), and
L the free path (the mean free path L is equal to 2 times the free path L). In turn,
the free path can be calculated according to Maxwells equation for the viscosity of a
gas, which is experimentally accessible:

1
3
GLd = , (5.4)

with the viscosity, G the mean molecular velocity, and d the density. Some of the
diameters calculated in this way by Perrin are shown in table 2. The calculated
diameters do not have the same degree of precision, particularly for polyatomic
molecules, in comparison with calculated masses because it is assumed that the atoms
or molecules are spherical (definition of diameter of impact or radius of protection).

Table 2: Results of calculated collision diameters.
Substance Diameter (m) v.d. Waals diameter (m) [36]
Hydrogen 2.110
-10
2.410
-10

Oxygen 2.710
-10
2.810
-10

Nitrogen 2.810
-10
3.210
-10

Chlorine 4.110
-10
3.610
-10


5.4 Displacement in a given time
Another method Perrin used to determine Avogadros number was to measure the
root-mean-square displacement of grains during certain successive equal time
intervals. The Brownian movement is entirely irregular in all directions and this fact is
the basis of Einsteins theory. Perrin accomplished the measurement of the
displacement of grains using a camera lucida with known magnification. The
positions occupied by a grain after successive time intervals were determined as
shown in figure 9. Figure 9A shows three diagrams obtained by tracing the horizontal
projections of the lines joining consecutive positions occupied by the same mastic
grain with a radius of 0.53 m, marked every 30 seconds. The displacement in time in
35
one direction can be readily obtained by projection of the positions along any axis and
will be proportional to the time. If the positions were to be marked at intervals of time
100 times shorter, a segment shown in figure 9A would be replaced by a polygonal
contour as shown in figure 9B, relatively just as complicated as the whole figure.
The expression Perrin used to determine Avogadros number is a combination of
Einsteins equations for the root-mean-square displacement in one dimension
(equation 4.10) and the expression for the diffusion coefficient (equation 4.6), and has
the form:

2
3
t RT
N
a x
=
(5.4)

with N Avogadros number, t the time (seconds), <x
2
> the mean square displacement,
R the gas constant, a the radius of the grain, and the viscosity of the fluid. Perrin
showed that Stokes law also holds for microscopic spheres, and the use of equation
5.4 is therefore applicable to the emulsions he used to determine the mean
displacement of grains. Typical numbers of displacements determined to get an
average value are ranging from 100 to 1,500.

A B
Figure 9: In A, three diagrams of root-mean-square displacements of the same mastic
grain (radius is 0.53 m) during successive time series of 30 seconds [37]. A magnification
of the root-mean-square displacement during one time interval is shown in B [31].
36
5.5 Values for Avogadros number found by Perrin
Other methods to determine Avogadros number Perrin used were measurements
on rotational diffusion and diffusion measurements. The rotational diffusion
measurements were done with grains with a radius up to 50 m containing inclusions
of water. Due to these inclusions the rotational diffusion of the grains is visible.
Diffusion measurements were done with large grains in a vessel with an object
glass on top to which the grains are absorbed. The diffusion coefficient is deduced
from the increase of grains absorbed to the object glass.
Perrin and his workers performed several experiments to determine Avogadros
number, and varied as much as they could the variables as temperature, radii of
grains, viscosity and density of the dispersion medium. The values found for
Avogadros number, obtained by the different methods are shown in table 3.

Table 3: Values found for Avogadros number.
Avogadros number
(N
AV
/10
22
mol
-1
)
Deduced from:
68.2 Vertical distribution
68.8 Translatory displacement
65 Rotational diffusion
69 Diffusion measurement

By 1914 measurements on Brownian motion had led to an estimation of the value
of Avogadros number of 6.0310
23
mol
-1
, which is close to the nowadays well-known
and accepted value of 6.022136710
23
mol
-1
. The various values for Avogadros
number found by Loschmidt, Perrin and others will be compared and discussed in the
next section.

37
6. Closing
Loschmidt was the first who estimated (in 1865) the molecular size of the right
order of magnitude as shown in figure 10. Perrin was the first who accurately
determined N
AV
from measurements on Brownian motion of colloids. Other
measurements to determine Avogadros number were accomplished using other
methods. Before 1924, data was derived mainly from atomic or molecular movement
in gases or fluids. Between 1924 and 1951 values for Avogadros number were
mainly obtained from data of X-ray wavelengths and data from 1965 until the present
day are derived from X-ray/crystal density measurements. As can be clearly seen
from the values for Avogadros number (left y-axis) and the relative uncertainties of
these numbers (right y-axis) in figure 10, the accuracy of the values found for
Avogadros number is strongly increased in time. Perrins determined value of
Avogadros number is remarkable accurate taken into account that he obtained this
value from measurements on Brownian motion.
1860 1880 1900 1920 1940 1960 1980 2000 2020
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
1E-8
1E-7
1E-6
1E-5
1E-4
1E-3
0.01
0.1
1
10
Loschmidt
Perrin
N
AV
/ 10
23
mol
-1
N
A
V

/

1
0
2
3

m
o
l
-
1
Year
Relative Uncertainty

R
e
l
a
t
i
v
e

U
n
c
e
r
t
a
i
n
t
y

Figure 10 [38]: Progress in the determination of N
AV
(solid circles, left y-axis)) from
1865 to 2001, together with the relative uncertainty (open squares, right y-axis). The
solid triangle represents the value for N
AV
determined by Loschmidt (1865), and the
solid star (1908) the value for N
AV
determined by Perrin.

38
The first estimation of the molecular size of air molecules by Loschmidt as well
as the estimation of molecular dimensions by Perrin is based on an equation that
describes the relation between the total surface of the spheres of impact and the mean
free path. In fact for both estimations another equation is used, which describes the
relation between the viscosity of a gas and the mean free path. Loschmidt makes use
of the equation for the total surface of the spheres of impact together with a value for
the mean free path. Without having a value for the number of molecules in a cubic
centimetre he is able to calculate the collision diameter for air molecules by
defining a condensation coefficient. This condensation coefficient, which reflects
packing fractions, is determined from empirical relations. Once Loschmidt obtained a
value for the diameter of air molecules he was able to calculate the number of
molecules in a cubic centimetre. Because Loschmidt uses a rather large value for the
mean free path, he overestimated the diameter of air molecules and consequently
underestimated Avogadros number. In contrast, Perrin uses also the equation for the
total surface of the spheres of impact but he calculated the mean free path from the
viscosity of the gas, which is experimental accessible, using Maxwells equation. To
calculate collision diameters for molecules Perrin only needed a value for Avogadros
number, which he determined from measurements on Brownian motion as height
distributions, translatory displacements, diffusion and rotational diffusion
measurements using gamboge and mastic colloidal systems. Because Perrin
overestimated Avogadros number, his estimations of collision diameters are
somewhat too small.
As we saw in the previous section Brownian motion led to the size estimation of
molecules. The discovery of the Brownian motion may be seen as a milestone in
history for chemical and physical science, maybe in particular for colloid science. It is
easy for anyone working in experimental colloid science nowadays to become rather
blas about Brownian motion because it is seen straightforwardly in the optical
microscope or in the fluctuating speckle pattern produced by the scattering of a laser
beam through a colloidal solution or suspension. The discovery of the Brownian
motion actually led eventually to some great achievements of modern condensed
physics. Among the achievements are the establishment of statistical mechanics, the
confirmation of the reality of the molecular and atomic nature of matter, and the
demonstration of the importance of thermal forces in physics of everyday phenomena.
The fundamental picture of reality, provided by the study of colloidal systems among
39
others by Perrin, goes far beyond the magnitude of the size range of colloidal
particles. Colloid science had and has its impact on the chemistry of atoms and
molecules because of the analogy of physical laws they obey. Brown, Einstein, and
Perrin are in fact the people who did the crucial work and are responsible for the
events that led to the understanding of Brownian motion. The events that led to the
understanding of Brownian motion, seen afterwards, can be divided in five stages
which are: the discovery and observation of Brownian motion, theoretical prediction
or description, experimental proof, and application. It took almost 100 years before
Brownian motion was understood sufficiently to relate it to N
AV
and molecular size.
As all major discoveries in science, the discovery and understanding of Brownian
motion did not occur in vacuum, but instead rely on the observations and findings in
the work of many others scientists.




40
References
[1] Philipse, A.P., Dansen op een Delta-piek, over colloiden en andere
kleinigheden, in Van 't Hoff Laboratorium voor Fysische en Colloidchemie.
1997, Universiteit Utrecht: Utrecht. p. 21.

[2] Pullman, B., The Atom in the History of Human Thought. 1998, New York:
Oxford University Press. 403.

[3] Lindberg, D.C., The Beginnings of Western Science: the European scientific
tradition in philosophical, religious, and institutional context, 600 B.C. to A.D.
1450. 1992, Chigaco: The University of Chigaco Press. 455.

[4] Lacina, A., Atom-from hypothesis to certainty. Phys. Educ. 1999, 34, 379-402.

[5] Haw, M.D., Colloidal suspensions, Brownian motion, molecular reality: a short
history. J. Phys.: Condens. Matter 2002, 14, 7769-7779.

[6] Dickerson, R.E., Gray, H.B., and Haight, J.G.P., Chemical Principles. Third
Edition ed. 1973, California: The Benjamin/Cummings Publishing Company,
Inc. 943.

[7] Porterfield, W.W. and Kruse, W., Loschmidt and the Discovery of the Small.
Journal of Chemical Education 1995, 72, 870-875.

[8] Loschmidt, J., Zur Grsse der Luftmolecle. Sitzingsberichte der Kaiserlichen
Akademie der Wissenschaften in Wien. 1865, 52, 395-407.

[9] Ford, B.J., Brownian Movement in Clarkia Pollen: A Reprise of the First
Observations. The Microscope 1992, 40, 235-241.

[10] Courtesy Edgar Fahs Smith Collection. University of Pennsylvania Library.
Facts On File, Inc. Science Online. www.factsonfile.com.

[11] Brown, R., The organs and Mode of Fecundation in Orchideae and
Asclepiadeae. Transaction of the Linnean Society 1833, 16, 710-713.

[12] Grew, N., The Anatomy of Plants. 1682, London. 171-2.

[13] Camerarius, R., Epistola de sexu plantarum. 1694, Tubingen: Cited by von
Sachs, J. "History of Botany, 1530-1860" translated by Garnsey, H.E.F.
rev.ed.,Clarendon Press, Oxford, 1906, p.387.

[14] Lister, J.J., Phil. Trans. 1830, 187-9.

[15] Brown, R., A Brief Account of Microscopical Observations Made in the Months
of June, July, and August 1827, on the Paricles contained in the Pollen of
Plants; and on the General existence of active Molecules in Organic and
Inorganic Bodies. Edinburgh New Philosophocal Journal 1828, 5, 358-371.
41
[16] Muncke, H.H., ber Robert Brown's mikroskopischer Beobachtungen, ber
Gefrierpunkt des absoluten Alkohols, und ber eine sonderbare Erscheinung an
der Coulomb'schen Drehwaage. Poggendorfs Annalen der Physik 1829, 17,
159-165.

[17] http://www.sciences.demon.co.uk/wbbrownb.htm

[18] Layton, D., The original Observations of Brownian Motion. Journal of
Chemical Education 1965, 42, 367-368.

[19] Brown, R., Additional Remarks on Active Molecules. Edinburgh Journal
Science 1829, 1, 314-319.

[20] Sloan, P.R., Darwin, Vital Matter and the Transformation of Species. J. History
of Biology 1986, 19, 369-445.

[21] Michelmore, P., Einstein, profile of the man. 1962, New York: Dodd, Mead.
269.

[22] Pais, A., Subtle is the Lord....: the science and the life of Albert einstein. 1982,
New York: Oxford University Press. 552.

[23] http://www.nobel.se/physics/laureates/1921/

[24] Einstein, A., ber einen die Erzeugung und Verwandlung des Lichtes
betreffenden heuristischen Gesichtspunkt. Annalen der Physik 1905, 17, 149.

[25] Einstein, A., Zur Elektrodynamik bewegter Krper. Annalen der Physik 1905,
17, 275.

[26] Einstein, A., ber die von der molekularkinetischen Theorie der Wrme
geforderte Bewegung von in ruhenden Flssigkeiten suspendierten Teilchen.
Annalen der Physik 1905, 17, 223.

[27] Stokes, G.C., On the effect of the internal friction of fluids on the motion of
pendulums. Proc. Cambridge Philos. Soc. 1851, 9, 8-106.

[28] http://www.britannica.com/eb/article?eu=60809&tocid=0&query=perrin&-ct=

[29] http://www.nobel.se/physics/laureates/1926/perrin-bio.html

[30] Nye, M.J., Molecular Reality, A perspective on the Scientific Work of Jean
Perrin. 1972, New York: American Elsevier Inc. 201.

[31] Perrin, J., Atoms. 4th ed. 1916, London: Constable & Company LTD. 211.

[32] http://www.britannica.com/eb/article?eu=36646&tocid=0&query=-
gamboge&ct=

[33] http://www.britannica.com/eb/article?eu=52598&tocid=0&query=mastic&ct=
42
[34] Randriamasy, F., Les grandes experiences de Jean Perrin, les preuves
experimentales de la realite moleculaire. Revue du Palais de la decouverte
1992, 20, 18-29.

[35] Perrin, J., Mouvement Brownien et Ralit Molculaire. Annales de Chimie et
de Physique. 1909, 18, 57.

[36] Verkerk, G., Broens, J.B., Kranendonk, W., Puijl van der, F.J., Sikkema, J.L.,
and Stam, C.W., Binas, informatieboek VWO/Havo voor het onderwijs in de
natuurwetenschappen. 1992, Groningen: Wolters-Noordhof bv.

[37] Perrin, J., Mouvement Brownien et Ralit Molculaire. Annales de Chimie et
de Physique. 1909, 18, 81.

[38] Brown, R. and Milton, M., Towards an improved determination of the
Avogadro constant. 2003, National Physical Laboratory: Middlesex.