Milestones_-_2008_-_Spin

Prévia do material em texto

“It was not simply out of a spirit of contradiction that I 
exposed a light source to a magnetic field,” so said Pieter 
Zeeman, on receiving the 1902 Nobel Prize in Physics 
with Hendrik Lorentz.
The effect of a magnetic field on light had been 
studied years before, most notably by Michael Faraday: in 
1845, he showed that when light passes through certain 
materials immersed in a magnetic field, the plane in 
which the light oscillates is rotated — now known as 
Faraday rotation, this was the first experimental evidence 
of a connection between light and electromagnetism. 
Later in his life, Faraday wondered whether a magnetic 
field could have a direct influence on a light source — 
specifically, on the light emitted by atoms or molecules 
when excited in a flame. This investigation was the last 
experiment recorded in his laboratory notebook, but the 
result was negative: on 12 March 1862, Faraday wrote 
that there was “not the slightest effect demonstrable”. 
Possibly referring to this experiment, James Clerk 
Maxwell stated in 1870, on the subject of light-emitting 
particles, that “no force in nature can alter even very 
slightly either their mass or their period of oscillation”. 
This statement, “coming from the mouth of the founder 
of the electromagnetic light theory and spoken with such 
intensity”, greatly troubled Zeeman.
Motivated by his own studies of the magneto-optic 
Kerr effect (a phenomenon closely related to the Faraday 
effect, but measured for light reflected from a magnetized 
medium), Zeeman decided in 1896 to revisit Faraday’s last 
experiment. Two recent technical advances proved useful. 
The first was the invention, by Henry Augustus Rowland, 
of diffraction gratings consisting of mirrors ruled with a 
large number of parallel lines; Rowland’s gratings offered 
unprecedented spectral resolution. The second was the 
steady development of photography, which brought with 
it the possibility of capturing spectra and analysing 
them later. 
Zeeman was soon able to show that, for a sodium 
flame placed in an electromagnet, there was a 
broadening of the sodium D-lines when the magnet was 
switched on. The effect was reportedly far from 
conspicuous; however, firm support was provided by the 
observation of characteristic polarization effects. This 
confirmed predictions made by Lorentz — immediately 
after seeing the first results produced by Zeeman, he had 
proposed a model of electrons vibrating within the light-
emitting particles. In 1897, Zeeman reported much 
clearer splittings in the blue line of cadmium. Although 
Lorentz’s theory did explain the most elementary 
splitting patterns, it was not long before numerous 
exceptions were found, as more elements were studied in 
greater detail.
“Nature gives us all, including Professor Lorentz, 
surprises”, concluded Zeeman. He might have been 
surprised to know that it would take nearly three decades 
for this ‘anomalous Zeeman effect’ to be explained fully 
— when it was realized that the splitting is a 
consequence of spin (Milestone 3).
Andreas Trabesinger, Senior Editor, Nature Physics
ORIGINAL RESEARCH PAPERS Zeeman, P. Over den invloed eener 
magnetisatie op den aard van het door een stof uitgezonden licht. Versl. Kon. 
Akad. Wetensch. Amsterdam 5, 181–184, 242–248 (1896) | Zeeman, P. Over 
doubletten en tripletten in het spektrum, teweeggebracht door uitwendige 
magnetische krachten. Versl. Kon. Akad. Wetensch. Amsterdam 6, 13–18, 
99–102, 260–262 (1897)
fuRtHER REAdING Zeeman, P. The effect of magnetisation on the nature of 
light emitted by a substance. Nature 55, 347 (1897) | Zeeman, P. Light radiation 
in a magnetic field, Nobel Lecture, 2 May, 1903. Nobelprize.org [online], 
<http://nobelprize.org/nobel_prizes/physics/laureates/1902/ 
zeeman-lecture.html> (1903) | Rayleigh. Pieter Zeeman. 1865–1943. 
Obit. Not. Fellows R. Soc. 4, 591–595 (1944)
 m I L E S tO N E 1
Physics is set spinning Razvanj
p 
| D
re
am
st
im
e.
co
m
…it would 
take nearly 
three decades 
for this 
‘anomalous 
Zeeman 
effect’ to be 
explained…
Nature MILeStONeS | Spin	 March 2008 | S5
 MILESTONES
© 2008 Nature Publishing Group 
 
http://nobelprize.org/nobel_prizes/physics/laureates/1902/zeeman-lecture.html
http://nobelprize.org/nobel_prizes/physics/laureates/1902/zeeman-lecture.html
the fact that quantities such as energy, mass and charge come in 
discrete and indivisible amounts, or quanta, is now so fundamental to 
our understanding of the universe that it is often taken for granted. 
It is difficult to appreciate fully just how counterintuitive an idea it 
was to suggest that the angular momentum of an object should take 
on similarly quantized values. the roots of this idea lay in the 1913 
description by Niels Bohr of the structure of the atom, later refined 
by arnold Sommerfeld in 1916, which suggested that not only do 
electrons exist in orbitals of well-defined size and shape, but that 
the orientation of these orbitals is strictly defined — a characteristic 
known as space quantization. In 1920, most physicists, including 
Max Born, who was one of the architects of quantum mechanics, 
considered the idea to be more a mathematical abstraction than a 
concrete physical reality. Yet, in 1922, Otto Stern and Walter Gerlach 
demonstrated such a reality beyond all reasonable doubt.
their experiment involved passing a collimated beam of 
silver atoms through an inhomogeneous magnetic field and onto 
a glass slide where the deposits formed a pattern. classical models 
suggested that the electron orbitals around the nucleus of these atoms 
should be randomly and continuously distributed, and that a single, 
broad and continuous spot of silver should form in the centre of the 
slide. the Bohr–Sommerfeld model, by contrast, predicted that space 
quantization of these orbitals should cause the beam to be split into 
several discrete parts in the inhomogeneous field, forming discrete lines 
of silver deposits on the slide.
Despite its elegant simplicity, the experiment almost never 
happened. For one thing, for there to be any observable splitting, the 
alignment of the beam and the centre of the magnetic field had to be 
just right. More prosaically, in the middle of a worsening economic 
depression, funding the construction of the experiment proved to be 
almost as difficult. thankfully, perseverance on the part of Stern and 
Gerlach, and a cheque for several hundred dollars provided by henry 
Goldman (co-founder of the investment firm Goldman Sachs), allowed 
them to observe the splitting predicted by quantum theory — a result 
 m I L E S tO N E 2
Answers	on	a	postcard
By 1920, physicists were still 
struggling to make sense of the splitting 
of atomic spectral lines in a magnetic 
field, discovered by Pieter Zeeman 
(Milestone 1). With the Bohr atom of 1913 
had come the notion of quantized orbits 
for electrons around the atomic nucleus. 
Perhaps, it was thought, the Zeeman 
splitting arose from the interaction 
between the angular momenta of the 
‘core’ of electrons in closed shells and of 
the ‘radiant electron’ in the outermost 
unclosed shell. Yet calculations made 
on this basis — notably by Arnold 
Sommerfeld and Alfred Landé, using 
an Ersatzmodell — failed to match 
experimental data. The helium atom 
posed a particular problem: which of 
its two electrons should be considered 
‘core’ and which ‘radiant’? Also, no one 
could explain why, if an atom were in 
its ground state, all its electrons were 
not bound into the innermost shell. In 
1924, Landé gave up the struggle as 
“impossible once and for all”.
Wolfgang Pauli, however, was 
undeterred. He dropped the notion 
of an interaction between core 
and radiant electron and proposed 
instead that line splitting arose as a 
consequence of an intrinsic property 
of the electron: “eine klassisch nicht 
beschreibbare Art von Zweideutigkeit” 
— a classically indescribable two-
valuedness — as he wrote in the first 
of his two 1925 Zeitschrift fürPhysik 
papers. 
 m I L E S tO N E 3
The spinning electron
Ralph Kronig, a young physicist in 
Landé’s laboratory, suggested to Pauli 
that this might be imagined as the 
rotation of an electron about its own axis 
with one half-unit of angular momentum 
— in other words, spin. Pauli disliked the 
idea: it was still ‘classically indescribable’ 
and, moreover, the calculations of level 
splitting by Kronig were a factor of two 
out from measured values. Kronig did 
not publish, but later that year George 
Uhlenbeck and Samuel Goudsmit did 
— the same idea, beset by the same 
problems. Pauli still didn’t like it.
Nevertheless, it was a short step from 
the ‘two-valuedness’ of an intrinsic 
electron quantum number for Pauli to 
realize why all electrons in an atom 
are not bound in the innermost state. 
Guided by comments made by Edmund 
C. Stoner, Pauli saw that, if the four 
numbers used to describe line splitting 
(denoted n, k, m and j) were thought of 
as quantum numbers of the electron, 
then once an electron existed in some 
state defined by particular values of 
n, k, m and j, no other electron could 
enter that state. Using this ‘exclusion 
principle’, Pauli could derive the exact 
shell structure of the atom — two 
electrons to close the innermost shell, 
eight to close the next, and so on.
The deal was sealed early in 1926, 
with the publication by Llewellyn H. 
Thomas of what became known as the 
Pauli and Bohr watch a 
spinning top. Photograph 
by Erik Gustafson, courtesy 
of AIP Emilio Segrè Visual 
Archives, Margrethe Bohr 
Collection.
Postcard from Gerlach to Bohr. Image courtesy of Niels Bohr Archive, Copenhagen.
Milestones
S6 | March 2008 	www.nature.com/milestones/spin
© 2008 Nature Publishing Group 
 
The idea of the spinning electron, as proposed by Samuel Goudsmit and George 
Uhlenbeck in 1925 (Milestone 3), and incorporated into the formalism of quantum 
mechanics by Wolfgang Pauli, was a solution of expediency. Yet this contrivance 
threw up a more fundamental question: as a 25-year-old postdoctoral fellow at the 
University of Cambridge formulated the problem in 1928, why should nature have 
chosen this particular model for the electron, instead of being satisfied with a point 
charge?
The young postdoc’s name was Paul Dirac, and in two papers published in the 
Proceedings of the Royal Society of London, he set out to explain why. Using the clear, 
austere prose and adroit mathematics that were his hallmarks, he showed how spin 
emerged as a natural consequence of the correct application of special relativity to 
the quantum mechanics of the electron.
Dirac was able to remove the nonlinearities in space and time derivatives that 
had confounded other attempts to marry those two great new physical theories. 
The corollary of his logic was that the wavefunction of the electron must have four 
components, and must be operated on by four-dimensional matrices. These matrices 
required an additional degree of freedom beyond position and momentum in the 
physical description of the electron. Inspection revealed them to be extensions 
of the two-dimensional spin matrices introduced by Pauli in his earlier ad hoc 
treatment. Applied to an electron in an electromagnetic field, the new formalism 
delivered the exact value of the magnetic moment assumed in the spinning electron 
model.
What had emerged was an equation that, in its author’s words, “governs most of 
physics and the whole of chemistry”. Dirac was a famously modest man, and was not 
wont to exaggerate. The Dirac equation is still today the best description not just of 
the electron, but of all spin-1/2 particles — including all the quarks and leptons from 
which matter is made.
When asked what had led him to his formula, Dirac replied simply “I found it 
beautiful”. His equation is indeed a powerful example of the deep and mysterious 
connection between the language of mathematics and the expressions of the 
physical world.
Yet, however much beauty might be indicative of rightness, a physical theory 
is judged on its predictive power. The Dirac equation did not disappoint. The 
interpretation of two of its four solutions was clear: they were the two spin states 
of the electron. But the other two solutions seemed to require particles exactly like 
electrons, but with a positive charge.
Dirac did not immediately and explicitly state the now-obvious conclusion — out 
of “pure cowardice”, he explained later. But when, in 1932, Carl Anderson confirmed 
the existence of the positron, Dirac’s fame was assured. He shared the 1933 Nobel 
Prize in Physics — its second-youngest-ever recipient — and his equation went 
on to become the bedrock of quantum electrodynamics, the quantum field theory 
of the electromagnetic interaction. Following his death in 1984, a stone was set 
into the floor of Westminster Abbey in London. It was inscribed with his name and 
iγ · дΨ = mΨ — the shortest and sweetest rendering of his extraordinary brainchild. 
Richard Webb, Senior Editor, Nature News & Views
ORIGINAL RESEARCH PAPERS Dirac, P. A. M. The quantum theory of the electron. Proc. R. Soc. Lond. A 
117, 610–624 (1928); ibid. 118, 351–361 (1928)
fuRtHER REAdING Dirac, P. A. M. The Principles of Quantum Mechanics (Int. Ser. Monograph. Phys. 27) 
4th edn (Oxford Univ. Press, Oxford, UK,	1982) | Pais, A., Jacob, M., Olive, D. I. & Atiyah, M. F. Paul Dirac: The 
Man and his Work (Cambridge Univ. Press, Cambridge, UK, 1998)
 m I L E S tO N E 4
A relative success
that is now most famously recorded on a postcard from Gerlach 
to Bohr congratulating him on the success of his theory.
although the Stern–Gerlach experiment categorically 
disproved classical models of the atom, it was also inconsist-
ent with the Bohr–Sommerfeld model. In fact, the observed 
splitting of the silver beam had nothing to do with the orbital 
angular momentum, but was due to the spin angular momen-
tum of the unpaired electron in the atomic structure of silver 
— something that was not appreciated until years later, follow-
ing the introduction of the idea of electron spin by Wolfgang 
Pauli (Milestone 3). 
Not content with realizing what is perhaps the clearest and 
most direct demonstration of the quantum nature of atoms, 
Stern went on to demonstrate and measure the quantized spin 
of the proton — together with the size of its magnetic moment 
— for which he was awarded the 1943 Nobel Prize in Physics.
Ed Gerstner, Senior Editor, Nature Physics
ORIGINAL RESEARCH PAPERS 
Gerlach, W. & Stern, O. Der experimentelle Nachweis der Richtungsquantelung im 
Magnetfeld. Z. Phys. 9, 349–352 (1922) | Frisch, R. & Stern, O. Über die magnetische 
Ablenkung von Wasserstoff-Molekülen und das magnetische Moment des Protons. I. 
Z. Phys. 85, 4–16 (1933) | Estermann, I. & Stern, O. Über die magnetische Ablenkung 
von Wasserstoff-Molekülen und das magnetische Moment des Protons. II. 
Z. Phys. 85, 17–24 (1933)
fuRtHER REAdING Friedrich, B. & Herschbach, D. Stern and Gerlach: how a bad 
cigar helped reorient atomic physics. Phys. Today 56 (12), 53–59 (2003)
Pauli’s ‘two-
valuedness’ 
was indeed due 
to the spin of 
the electron.
‘Thomas factor’ — the missing factor 
of two. Thomas’ classical analysis 
finally won over Pauli the perfectionist 
(Paul Dirac would soon supply the 
full quantum relativistic formalism; 
Milestone 4). Pauli’s ‘two-valuedness’ 
was indeed due to the spin of the 
electron. Probably the great man was 
cheered, having written to Kronig in May 
1925, “At the moment physics is again 
terribly confused. In any case, it is too 
difficult for me, and I wish I had been a 
movie comedian or something of the sort 
and had never heard of physics.”
Alison Wright, Chief Editor, 
Nature Physics
ORIGINAL RESEARCH PAPERS Stoner, E. C. 
The distribution of electrons among atomic levels. 
Phil. Mag. 48, 719–736 (1924) | Pauli, W. Über den 
Einfluß der Geschwindigkeitsabhängigkeit der 
Elektronenmasse auf den Zeemaneffekt. Z. Phys. 
31, 373–385 (1925)| Pauli, W. Über den 
Zusammenhang des Abschlusses der 
Elektronengruppen im Atom mit der 
Komplexstruktur der Spektren. Z. Phys.	31, 
765–783 (1925) | Uhlenbeck, G. E. & Goudsmit, S. A. 
Ersetzung der Hypothese vom unmechanischen 
Zwang durch eine Forderung bezüglich des 
inneren Verhaltens jedes einzelnen Elektrons. 
Naturwiss. 13, 953–954 (1925) | Thomas, L. H. 
The motion of the spinning electron. Nature 117, 
514 (1926)
fuRtHER REAdING Pauli, W. Exclusion 
principle and quantum mechanics, Nobel 
Lecture, 13 December, 1946. Nobelprize.org 
[online], <http://nobelprize.org/nobel_prizes/
physics/laureates/1945/pauli-lecture.html> 
(1946) | Tomonaga, S. The Story of Spin (Univ. 
Chicago Press, Chicago, 1997)
Im
ag
e 
co
ur
te
sy
 o
f L
id
a 
Lo
pe
s C
ar
do
zo
 K
in
de
rs
le
y.
Nature MILeStONeS | Spin	 March 2008 | S7
© 2008 Nature Publishing Group 
 
Since the earliest observation 
that a lump of lodestone 
attracts iron, around 
600 Bc, magnetism has 
attracted numerous 
philosophers and 
physicists eager to 
understand its secrets. 
Yet the physical explana-
tion came only with the 
development of quantum 
mechanics in the 1920s 
— after all, magnetism is a 
purely quantum effect arising 
from the ‘spin’ property of the 
electron. By convention, we say 
that the spin points either up or 
down. this means that there are two 
distinct spin states instead of one; 
hence, certain theories were out 
by a factor of two with respect to 
experiments.
Following a series of papers 
published in the years 1925 to 1927 
— during which quantum mechan-
ics was developed, interpreted and 
applied to atoms with more than 
one electron outside a closed shell 
— Werner heisenberg solved the 
mystery of ferromagnetism using the 
concept of spin plus the exclusion 
principle formulated by Wolfgang 
Pauli, which states that two electrons 
with the same energy and momentum 
cannot occupy the same quantum 
state. In other words, two electrons 
with the same energy but different 
spins can lie in the same orbital. the 
 m I L E S tO N E 5
An attractive theory
In 1932, Werner Heisenberg mused on 
an odd fact. The proton and the neutron 
(which had been discovered only earlier 
that same year by James Chadwick) had 
almost exactly the same mass. Despite 
their different charges, they also 
responded identically to the forces that 
dominate within the atomic nucleus.
To Heisenberg’s nose, this had a whiff 
of an uncovered symmetry about it. 
Appropriating the mathematics that 
Wolfgang Pauli had used to describe 
spin (Milestone 3), he postulated that 
the proton and neutron were two 
states of the same particle, the nucleon. 
These states differed only in a quantity 
analogous to spin — the ‘isotopic spin’, 
or isospin as it came to be known. The 
nuclear force conserved isospin, which 
accounted for the similarities between 
protons and neutrons. Other forces, 
such as electromagnetism, broke 
isospin symmetry, which explained the 
nucleons’ differences. 
As Eugene Wigner wrote of the isospin 
concept in a 1937 paper, “no such states 
are known to be of any importance 
[…] [but they] will turn out to be very 
useful”. In that paper, he used isospin 
to predict correctly the energies of 
all nuclei up to atomic number 42; 
more recent work has extended that 
success to even heavier nuclei. Much 
like the quantum-relativistic prediction 
of a spinning electron by Paul Dirac 
(Milestone 4), isospin was an example of 
what Wigner would later, in a celebrated 
essay, describe as the “unreasonable 
effectiveness” of mathematics in 
predicting physical phenomena.
And how. In 1935, Hideki Yukawa 
modelled a nuclear force mediated by 
lighter particles exchanged between 
the nucleons. Isospin conservation 
demanded three such particles. 
Believers in unreasonable effectiveness 
could not have been surprised when, 
some 10 years later, three particles 
answering the description turned up in 
cosmic rays and in the first accelerator 
experiments: the two charged and one 
neutral pion. 
In 1954, Chen Ning Yang and Robert 
Mills took the ideas of Yukawa further 
to establish their principle of ‘gauge 
invariance’. This was the centrepiece of 
 m I L E S tO N E 6
Spin’s	nuclear	sibling
a generalized mathematical description 
of forces mediated by exchange particles 
of integer spin and isospin — the bedrock 
of current quantum field theories of the 
fundamental forces of nature.
Meanwhile, however, physics was 
entering the accelerator age, and the 
discovery of a seemingly unordered 
menagerie of particles similar to the 
pions was straining the foundation of 
isospin symmetry. It gradually became 
clear that isospin was not a fundamental 
symmetry, but just one corner of a 
larger edifice. In addition, the proton 
and neutron were not two states of the 
same particle. In fact, they were not 
elementary particles at all, but were 
made up of smaller entities — quarks.
facts that an electron has spin, as 
well as charge, and that two identical 
electrons must occupy different states, 
are the keys to the periodic table.
until then, the force aligning the 
electron spins could not be explained 
in terms of known interactions, none 
of which was strong enough. In the 
words of Paul Dirac: “the solution of 
this difficulty […] is provided by the 
exchange (austausch) interaction of 
the electrons, which arises owing to 
the electrons being indistinguishable 
one from another. two electrons may 
change places without our knowing 
it, and the proper allowance for the 
possibility of quantum jumps of this 
nature, which can be made in a treat-
ment of the problem by quantum 
mechanics, gives rise to the new kind 
of interaction. the energies involved, 
the so-called exchange energies, are 
quite large.”
By applying such an energy 
tax on indistinguishable particles, 
ST
O
C
KB
YT
E
Werner Heisenberg. Photograph by Friedrich Hund, 
courtesy of AIP Emilio Segrè Visual Archives.
Milestones
S8 | March 2008 	www.nature.com/milestones/spin
© 2008 Nature Publishing Group 
 
Once deprived of its original 
legitimacy, one might have expected 
spin’s nuclear sibling to disappear. 
In fact, the same mathematical 
description resurfaced within the 
new fundamental paradigm as ‘weak 
isospin’, which is a property of quarks 
that is conserved in weak interactions 
— a testament to how deeply 
embedded the language of spin seems 
to be in the workings of the world.
Richard Webb, Senior Editor, 
Nature News & Views
ORIGINAL RESEARCH PAPERS Heisenberg, W. 
Über den Bau der Atomkerne. Z. Phys. 77, 1–11 
(1932) | Yukawa, H. On the interaction of elemen-
tary particles. Proc. Phys. Math. Soc. Jap. 17, 48–57 
(1935) | Wigner, E. On the consequences of the 
symmetry of the nuclear Hamiltonian on the 
spectroscopy of nuclei. Phys. Rev. 51, 106–119 
(1937) | Yang, C. N. & Mills, R. L. Conservation 
of isotopic spin and isotopic gauge invariance. 
Phys. Rev. 96, 191–195 (1954) | Gell-Mann, M. 
Symmetries of baryons and mesons. Phys. Rev. 
125, 1067–1084 (1962)
fuRtHER REAdING Wigner, E. The unreason-
able effectiveness of mathematics in the natural 
sciences. Comm. Pure Appl. Math. 13, 1–14 (1960) | 
Robson, D. Isospin in nuclei. Science 179, 
133–139 (1973) | Warner, D. D., Bentley, M. A. & 
Van Isacker, P. The role of isospin symmetry in 
collective nuclear structure. Nature Phys. 2, 
311–318 (2006) | Anderson, R. & Joshi, G. C. 
Interpreting mathematics in physics: charting 
the applications of SU(2) in 20th century physics, 
arXiv.org [online] <http://arxiv.org/PS_cache/
physics/pdf/0605/0605012v1.pdf> (2006)
heisenberg proposed a model 
that counted up all the spins and 
included the exchange interaction 
between nearest neighbours only; for 
example, in a linear chain of spins, 
only two neighbouring spins would 
count, in a square lattice, four. When 
these were summed, heisenberg 
found a ground state (lowest-energy 
configuration) in which all the spins 
of the electronslined up in parallel 
— that is, a ferromagnetic state 
without the need for any external 
magnetic field. 
May Chiao, Senior Editor, 
Nature Physics
ORIGINAL RESEARCH PAPERS Heisenberg, W. 
Zur Theorie des Ferromagnetismus. Z. Phys. 49, 
619–636 (1928) | Dirac, P. A. M. Quantum 
mechanics in many-electron systems. Proc. R. Soc. 
Lond. A 123, 714–733 (1929)
fuRtHER REAdING Mott, N. & Peierls, R. 
Werner Heisenberg. 5 December 1901–1 
February 1976. Biogr. Mem. Fellows R. Soc. 23, 
212–251 (1977)
The Dirac equation is monumental in physics, encapsulating so beautifully, in 
relativistic terms, the behaviour of a spinning electron, or indeed of any 
particle that has half-integer spin (Milestone 4). Wolfgang Pauli — on 
whose ideas the concept of a spinning electron was based — was 
impressed by the mathematical ‘acrobatics’ of Paul Dirac in arriving at the 
succinct expression, published in 1928, but he was not, however, 
satisfied. Pauli questioned the reliance of Dirac’s theory on the exclusion 
principle and the emphasis on a half-unit of spin — why should nature 
permit only half units? 
With Victor Weisskopf, Pauli set about resurrecting the Klein–Gordon 
equation, which describes a particle that has zero spin, but which had 
been all but abandoned following the unsuccessful attempt by Erwin 
Schrödinger in 1926 to build it into a theory of quantum-wave mechanics. 
Weisskopf and Pauli, however, succeeded in quantizing the Klein–Gordon 
equation to obtain spin-0 particles of both negative and positive charge — just as 
Dirac had obtained spin-1/2 particles of negative and positive charge from his 
equation. These spin-0 particles, moreover, did not obey the exclusion principle.
Dirac’s ‘positive electron’ — the positron — was discovered (although 
not immediately recognized as such) by Carl Anderson in 1932, the same 
year that James Chadwick discovered the neutron; in 1935, Hideki 
Yukawa postulated the existence of the meson, and the muon was 
discovered in 1936. There was suddenly a growing family of particles 
to describe, alongside the electron, proton and photon. It was 
thinking about how to reconcile the Klein–Gordon and Dirac 
equations, and the existence of all these particles (how many 
more might be discovered?) that led Pauli to one of the most 
subtle concepts of modern physics — the spin–statistics 
theorem.
In his 1940 paper, Pauli identified a vital connection between spin and 
quantum statistics (in the 1920s, it had been realized that something more 
than the Maxwell–Boltzmann variety was needed at the quantum level). 
According to Pauli, particles of half-integer spin obey Fermi–Dirac statistics 
(and, hence, are now called ‘fermions’) and those of integer spin obey 
Bose–Einstein statistics (‘bosons’). Mathematically speaking, the quantization 
of fields with half-integer spin relies on ‘plus’ commutation relations, 
whereas that of fields with integer spin uses ‘minus’ commutation 
relations. Put another way, the wavefunction of a system of bosons is 
symmetric if any pair of bosons is interchanged, but is antisymmetric for 
interchanged particles in a system of fermions.
Subtle indeed, but from Pauli’s spin–statistics connection arises the 
exclusion principle for fermions, with its implications for atomic structure, 
and a ‘non-exclusion’ principle for bosons — many bosons can adopt the 
same quantum state at once, as happens in a Bose–Einstein condensate. 
Further particle discoveries since 1940 and the subsequent building of the 
‘standard model’ have also served to confirm that nature works with both 
integer and half-integer spins.
Alison Wright, Chief Editor, 
Nature Physics
ORIGINAL RESEARCH PAPERS Pauli, W. & Weisskopf, V. F. Über die Quantisierung der skalaren 
relativistischen Wellengleichung. Helv. Phys. Acta 7, 709–731 (1934) | Pauli, W. The connection between 
spin and statistics.	Phys. Rev. 58, 716–722 (1940)
fuRtHER REAdING Duck, I. & Sudarshan, E. C. G. Toward an understanding of the spin–statistics 
theorem. Am. J. Phys. 66, 284–303 (1998) | Tomonaga, S.	The Story of Spin (Univ. Chicago Press, 
Chicago, 1997)
 m I L E S tO N E 7
Vital	statistics
Mi
los
luz
 | D
rea
ms
tim
e.c
om
According to 
Pauli, particles of 
half-integer spin 
obey Fermi–Dirac 
statistics and those 
of integer spin obey 
Bose–Einstein 
statistics.
Milestones
Nature MILeStONeS | Spin	 March 2008 | S9
© 2008 Nature Publishing Group 
 
If, during the 1920s and 1930s, the 
atomic nucleus had seemed of interest 
to few besides the (mostly) gentleman 
scientists studying it, by the end of the 
Second World War its wider importance 
was abundantly clear. The coming of the 
nuclear age was an appropriate cue for 
the two papers that cleared the way for 
arguably the most widespread practical 
application of nuclear spin today: nuclear 
 m I L E S tO N E 8
New resonance
magnetic resonance (NMR) spectroscopy.
The 1946 work of Edward Mills Purcell 
at the Massachusetts Institute of 
Technology and Felix Bloch at Stanford 
University gave new relevance to one 
object of intense gentlemanly interest 
before the war: the Zeeman splitting of 
nuclear spin states in a magnetic field 
 (Milestone 1). The degree of splitting 
at a particular magnetic field strength 
depends on the gyromagnetic ratio of 
the nucleus. In NMR, a second, transverse 
field at the characteristic (typically radio) 
spin-transition frequency produces an 
absorption resonance — a powerful way 
to identify the nuclei present in a sample.
Purcell et al. brought protons (1H) in 
solid paraffin to resonance; Bloch et 
al. did the same in liquid water. The 
coincident timing was no accident: the 
development of radar technologies 
during the war, for which several 
of the researchers involved had 
won their spurs, had made sources 
of radiofrequency radiation freely 
available for the first time. 
The effect itself was not entirely 
new. In 1938, Isidor Rabi had used it to 
measure magnetic moments of both 
atomic species in a lithium chloride 
molecular beam, receiving the 1944 
Nobel Prize in Physics for that advance. 
Even earlier, the Dutch physicist 
Cornelis J. Gorter had looked for the 
resonance of 7Li in lithium fluoride and 
1H in alum, using a calorimetric method. 
Hampered by experimental vagaries 
and limited resources, he published 
a negative result. (In later years, on 
receiving a prize for his contributions to 
low-temperature physics, Gorter would 
muse on his strange ability to miss out 
on groundbreaking discoveries in this 
and other instances.)
The innovations offered by Bloch and 
Purcell’s approaches were the transition 
to real liquid and solid systems, and, in 
Bloch’s case, the use of an induction coil 
to pick up and sharpen the resonance 
signal. These opened the way for the 
use of NMR in all manner of contexts, 
including in living tissue — where 
it became the lynchpin of magnetic 
resonance imaging (Milestone 15). 
In 1944, although shielded in the 
relative obscurity of Kazan in the 
steppes of Tatarstan, the Soviet physicist 
Yevgeny Zavoisky published the first 
measurements of an analogous effect 
involving electron spins. Electron 
paramagnetic resonance depends on an 
atom possessing an unpaired electron, 
 m I L E S tO N E 9
From the compass to Apollo 
Long before the concept of spin had 
been realized, the phenomenon of 
magnetism was a source of fascina-
tion and curiosity. the first scientific 
record of magnetism was made by 
the Greek philosopher thales of 
Miletos, who, in the sixth century bc, 
studied the attraction of materials 
such as iron to loadstone (magnetite). 
the first magnetic device was, of 
course, the compass — probably 
invented by several cultures inde-
pendently and first documented in 
chinese literature in the eleventh 
century ad. 
Nearly a millennium later, and 
particularly since the 1950s, devices 
based on magnetism are once more 
proving significant in shaping our 
way of life. the magnetic tape, whichwas invented in 1878 by Oberlin 
Smith, was commercialized in the 
1930s by aeG and BaSF. In later 
decades, it was developed into, for 
example, videotape in 1951 and the 
magnetic stripes on credit cards in 
the 1960s. 
Following the invention of mod-
ern computers, technology similar 
to magnetic tape was the logical 
choice for long-term data storage. 
the first hard drive with a moveable 
head was built into the IBM 305 
computer, which shipped in 1956. 
Its large hard disks — 24 inches in 
diameter — had a storage density of 
2 kilobits per square inch. IBM was 
also a pioneer in the development of 
the removable floppy disk. the first 
floppy disk, which had a diameter 
of 8 inches and a storage capacity of 
about 80 kilobytes, dates from 1969. 
the cumbersome 8-inch format was 
soon brought down in size: the last 
popular format was the Sony 3.5-inch 
floppy.
Magnetism has also been the key 
to several other, historical, storage 
techniques. the earliest was the 
‘drum memory’ in the 1950s, which 
consisted of rotating circular metallic 
plates coated with a magnetic mate-
rial. Drum memory was superseded 
in the 1960s by the ‘core memory’ 
— a hand-woven grid of wires with 
small ferrite rings (the cores) at the 
intersections. complex current pulses 
First NMR signals from water. Image reprinted with 
permission from Bloch, F., Hansen, W. W. and 
Packard, M. Phys. Rev. 70, 474–485 (1946). 
Courtesy of the American Physical Society.
The coming 
of the nuclear 
age was an 
appropriate cue… 
for arguably the 
most widespread 
practical 
application of 
nuclear spin 
today.
Magnetic core random 
access memory. Image 
courtesy of H. J. Sommer III, 
Professor of Mechanical 
Engineering, Penn State 
University.
Milestones
S10 | March 2008 	www.nature.com/milestones/spin
© 2008 Nature Publishing Group 
 
http://en.wikipedia.org/wiki/Magnetic_core_memory
thereby limiting the range of its 
application, but making it useful for 
the detection and identification of free 
radicals. Zavoisky might also have been 
the first to see an NMR signal, but he 
did not follow it up, at least not with 
publications. Had the vicissitudes of the 
age been less, and the dissemination of 
scientific information easier, his claim 
might have been better heard in the 
West. As it was, the 1952 Nobel Prize in 
Physics went to Bloch and Purcell. 
Richard Webb, Senior Editor, 
Nature News & Views
ORIGINAL RESEARCH PAPERS Gorter, C. J. 
Negative results of an attempt to detect nuclear 
magnetic spins. Physica 9, 995–998 (1936) | 
Rabi, I. I., Zacharias, J. R., Millman, S. & Kusch, P. 
A new method of measuring nuclear magnetic 
moment. Phys. Rev. 53, 318 (1938) | Zavoisky, E. 
Relaxation of liquid solutions for perpendicular 
fields.	J. Phys. USSR 9, 211–216 (1945) | 
Purcell, E. M., Torrey, H. C. & Pound, R. V. 
Resonance absorption by nuclear magnetic 
moments in a solid.	Phys. Rev. 69, 37–38 (1946) | 
Bloch, F., Hansen, W. W. & Packard, M. Nuclear 
induction.	Phys. Rev. 69, 127 (1946) | Zavoisky, E. 
Spin magnetic resonance in the decimetre-wave 
region.	J. Phys. USSR 10, 197–198 (1946)
fuRtHER REAdING Gorter, C. J. Bad luck in 
attempts to make scientific discoveries. Phys. 
Today 20	(1), 76–81 (1967) | Kochelaev, B. I. & 
Yablokov, Y. V. The Beginning of Paramagnetic 
Resonance (World Scientific, Singapore, 1995)
Nuclear magnetic resonance (NMR) spectroscopy 
(Milestone 8) is one of the most powerful 
analytical techniques in modern chemistry — a 
window into the world of molecules that can 
provide information about their structures, 
dynamic behaviour and how they interact with 
one another.
Prior to the 1950s, the study of NMR was 
rooted firmly in the physics community. It was 
assumed that the frequency at which a given 
nucleus resonated depended only on the strength 
of the magnetic field in which it was placed. 
Physicists therefore anticipated that they could use 
the technique to measure — with unprecedented 
precision — the magnetic moments of different 
nuclei. When, in 1950, Warren Proctor and Fu 
Chun Yu set out to do this for 14N, something 
unexpected happened. 
For their experiments, they chose the 
compound ammonium nitrate (NH4NO3), which is 
highly soluble in water and contains two nitrogen 
nuclei per molecule — factors that were expected 
to improve the NMR signal. In what they 
described as a “surprising observation”, however, 
not one but two resonance frequencies were 
detected — one for the nitrogen nuclei in the 
ammonium (NH4+ ) ions and the other for those in 
the nitrate (NO3– ) ions.
This was the first reported observation of the 
phenomenon that soon became known as 
‘chemical shift’, in which the local chemical 
environment surrounding a nucleus influences the 
frequency at which it resonates. The implications 
of NMR for the structural analysis of organic 
compounds became apparent soon afterwards, 
when, in 1951, a group of researchers from 
Stanford University showed that different 1H 
nuclei in the same molecule resonate at different 
frequencies.
James Arnold, Srinivas Dharmatti and Martin 
Packard demonstrated the huge potential of 
NMR spectroscopy by applying the technique to 
ethanol (CH3CH2OH), a compound in which each 
molecule comprises three sets of non-equivalent 
1H nuclei. Using tiny sample volumes and placing 
them in the most uniform region within a 
magnetic field, they obtained a spectrum 
displaying three separate lines, corresponding 
to the resonant frequencies of the 1H nuclei in 
the CH3, CH2 and OH groups, respectively. 
Moreover, the relative intensities of the three 
signals corresponded with the number of 
protons in each different chemical environment. 
So it was possible not only to identify different 
molecular fragments but also to glean 
quantitative information about the number of 
equivalent nuclei in each. 
Later in 1951, Herbert Gutowsky and David 
McCall showed that different spin-active nuclei in 
the same molecule interact with one another, 
giving rise to fine structure in the NMR signals 
that encodes a wealth of information regarding 
molecular connectivity and structure.
It did not take the chemistry community long to 
embrace the technique for the spectroscopic 
analysis of compounds. Techniques using radio-
frequency pulses — rather than a continuous 
source — broadened the scope of NMR 
spectroscopy, and Fourier-transform methods of 
data processing notably improved the sensitivity of 
the method. The combination of these advances 
allowed the development of sophisticated multi-
dimensional NMR experiments that revolutionized 
the field (Milestone 16).
Stuart Cantrill, Chief Editor, 
Nature Chemistry
ORIGINAL RESEARCH PAPERS Proctor, W. G. & Yu, F. C. The 
dependence of a nuclear magnetic resonance frequency upon 
chemical compound. Phys. Rev. 77, 717 (1950) | Hahn, E. L. Spin 
echoes. Phys. Rev. 80, 580–594 (1950) | Arnold, J. T., Dharmatti, S. S. 
& Packard, M. E. Chemical effects on nuclear induction signals from 
organic compounds. J. Chem. Phys. 19, 507 (1951) | Gutowsky, H. S. & 
McCall, D. W. Nuclear magnetic resonance fine structure in liquids. 
Phys. Rev. 82, 748–749 (1951) | Carr, H. Y. & Purcell, E. M. Effects of 
diffusion on free precession in nuclear magnetic resonance experi-
ments. Phys. Rev. 94, 630–638 (1954) | Ernst, R. R. & Anderson, W. A. 
Application of Fourier transform spectroscopy to magnetic reso-
nance. Rev. Sci. Instrum. 37, 93–102 (1966) | Aue, W. P., Bartholdi, E. 
& Ernst, R. R. Two dimensional spectroscopy. Application to nuclear 
magnetic resonance. J. Chem. Phys. 64, 2229–2246 (1976)
fuRtHER REAdING 
Becker, E. D. Magnetic resonance: an account of some key discoveries 
and their consequences. Appl. Spectrosc. 50, 16A–28A (1996)
 m I L E S tO N E 1 0
A	shift	in	expectations
through the wires were able to read, 
as well as set, the magnetization of 
the cores. Despite being an intricate 
device, a core memory of two cubic 
feet, with a capacity of 4,096 words, 
wasused in the apollo guidance com-
puter, onboard the NaSa missions 
to the Moon. computer memory was 
miniaturized further in the late 1970s, 
for example using ‘bubble memory’ in 
which data storage is based on small 
magnetic domains on a thin film. 
Soon afterwards, hard drives became 
the dominant data-storage system for 
computers.
the history of magnetic devices 
illustrates well how, with a little 
inventiveness, the macroscopic 
manifestations of magnetism can be 
harvested to achieve amazing techno-
logical advances. however, it would 
take a more fundamental under-
standing of spin physics to achieve 
the next technological revolution in 
computing and information storage 
(Milestone 18).
Joerg Heber, Senior Editor, 
Nature Materials
Paolotoscani | Dreamstime.com
Milestones
Nature MILeStONeS | Spin	 March 2008 | S11
© 2008 Nature Publishing Group 
 
In 1935, albert einstein, Boris 
Podolsky and Nathan rosen ques-
tioned whether quantum mechanics 
fully describes ‘physical reality’. 
their paper, which was intended to 
illustrate that quantum mechanics 
is incomplete, sparked discussions 
that go deep into the philosophical 
aspects of ‘reality’ and how phys-
ics can describe it. Later that year, 
einstein confessed in a letter to 
erwin Schrödinger that he felt that 
“the main point was, so to speak, 
buried by erudition”, and began 
publishing his own versions of the 
‘incompleteness argument’. all of 
these accounts, and the original 
einstein–Podolsky–rosen paper, 
made their point using continuous 
variables — that is, position and 
momentum. however, the version 
that is most widely discussed in the 
modern literature, and also forms 
 m I L E S tO N E 1 1
Mind-boggling 
reality
Impurities are not always unwanted. 
With the right type and dose of impurity 
atoms, the bulk properties of a material 
can be tuned in a beneficial way, which 
is a technique made heavy use of, for 
example, in standard silicon technology. 
At the microscopic level, interesting 
questions arise about how an impurity 
atom interacts with its host.
In 1964, Jun Kondo resolved a 
long-standing question regarding 
the electrical resistance of magnetic 
impurity-doped metals. The mystery was 
this: the resistance of a metal should 
decrease with decreasing temperature, 
as atomic vibrations freeze out, so that 
conduction electrons can move more 
easily through the material; however, 
for magnetically doped metals, the 
resistance was found to increase again 
below a certain temperature. Kondo 
discovered that it is the intrinsic spin of 
magnetic impurity atoms that leads to 
this anomalous resistance. The amount 
of scattering that electrons experience 
at the impurities does not decrease but 
increases when the temperature goes 
down, and this leads to the observed 
minimum in total resistance. 
Not only did this finding explain a 
nagging problem but it also triggered 
a vast amount of theoretical follow-up 
work. The initial issue to tackle was 
that the effect seemed to yield infinite 
resistance as zero temperature is 
approached — clearly an unphysical 
result. It was soon found that this 
divergence of resistance is suppressed, 
below a certain temperature (the 
Kondo temperature), by the formation 
of a bound state between impurity and 
conduction electrons, in which electron 
spins line up to screen the spin of the 
impurity atom. A later development was 
 m I L E S tO N E 1 2
Odd one out
the extension of the theory to ‘Kondo 
lattices’ in which electrons interact not 
only with the odd magnetic impurity but 
rather with an array of localized spins, 
and are significantly slowed down by 
the strong interactions. This model can 
explain some of the unusual properties, 
such as anomalous superconductivity, 
displayed by so-called heavy-fermion 
compounds.
Another line of research is the Kondo 
effect in nanometre-sized structures, 
such as quantum dots, in which the 
interactions between a single magnetic 
impurity and its environment can be 
controlled. A quantum dot can be tuned 
to contain an odd number of electrons 
— that is, an unpaired spin. Below the 
Kondo temperature, this localized spin 
can form a bound state with the free 
electrons in the electrodes on either 
side of the quantum dot, similar to the 
classical Kondo effect. However, this 
‘Kondo resonance’ opens an additional 
pathway for electrons to flow through 
the quantum dot and, as a result, the 
resistance decreases, in contrast to the 
original effect.
Jun Kondo. Image courtesy of 
AIST, Tokyo.
the basis of many experimental 
investigations, presents the argu-
ment in a simpler and clearer form, 
in terms of discrete spin variables. 
It was penned by David Bohm and 
appeared originally in his 1951 book 
Quantum Theory; he developed the 
argument further, in the context 
of experimental proofs, with Yakir 
aharonov in 1957.
Bohm and aharonov considered a 
molecule made of two atoms — each 
having one half-unit of spin — 
combined such that the total spin of 
the molecule is zero. When the two 
atoms are separated, and, for one 
of the spins, the spin component is 
measured along a given direction, 
the same component is immediately 
known for the other spin — it is 
exactly the opposite, as the total spin 
still has to be zero. at first sight, it 
might not seem surprising that infor-
mation about the properties of the 
second particle of a composite system 
can be deduced without performing 
any measurement on it, and without 
any interaction between the two par-
ticles, if the initial condition restricts 
how the two particles behave with 
respect to each other.
For a quantum spin, however, the 
situation is more subtle. Quantum 
mechanics allows only one compo-
nent of the spin to have a definite 
value. If, for instance, the 
x component of the spin is known, 
then the components along the y 
and z axes must be indeterminate; 
the component that is definite is 
determined by its measurement. 
Yet, in this case of two separated 
spin-1/2 particles, an experimenter 
can decide at the last minute — long 
after the two constituents have been 
separated — along which direction 
the first spin is measured. and this 
choice has immediate consequences 
on which component of the second, 
unobserved, spin is definite. 
how can the second spin know 
what has been done to the first? Is 
there some kind of hidden interac-
tion that quantum theory does not 
account for? Does quantum mechan-
ics allow what einstein famously 
called “spooky action at a distance” 
(an idea he did not like)? einstein 
argued that if no action at a distance 
can instantaneously influence the 
second spin, then it must have had all 
its components well defined from the 
David Bohm. Image from 
Library of Congress, 
New York — Telegram and 
Sun Collection, courtesy of 
AIP Emilio Segrè Visual 
Archives.
S12 | March 2008 	www.nature.com/milestones/spin
© 2008 Nature Publishing Group 
 
The way that spin is woven into the 
very fabric of the Universe is writ 
large in the standard model of 
particle physics. In this model, 
which took shape in the 1970s and 
can explain the results of all 
particle-physics experiments to 
date, matter (and antimatter) is 
made of three families of quarks 
and leptons, which are all fermions, 
whereas the electromagnetic, strong 
and weak forces that act on these particles are 
carried by other particles, such as photons and 
gluons, which are all bosons.
Despite its success, the standard model is 
unsatisfactory for a number of reasons. First, 
although the electromagnetic and weak forces 
have been unified into a single force, a ‘grand 
unified theory’ that brings the strong interaction 
into the fold remains elusive. Second, the origins 
of mass are not fully understood. Third, gravity is 
not included.
Moreover, there are other, less obvious 
problems with the standard model. The two 
natural mass scales in nature are zero and the 
Planck mass, ~ 1019 GeV c−2. Neither photons nor 
gluons(which carry the electromagnetic and 
strong forces, respectively) have mass, but the W 
and Z bosons that are responsible for the weak 
force have masses of about 90 GeV c−2. Where 
does this mass scale come from?
This ‘hierarchy problem’ can be solved by fine-
tuning the model so that various quantum 
fluctuations cancel out, although many physicists 
are uncomfortable with this solution because 
some parameters must be fine-tuned to better 
than 1 part in 1015. However, a form of symmetry 
between fermions and bosons called 
supersymmetry offers a much more elegant 
solution because the quantum fluctuations 
caused by bosons are naturally cancelled 
out by those caused by fermions and 
vice versa.
Symmetry plays a central role in physics. The 
fact that the laws of physics are, for instance, 
symmetric in time (that is, they do not change 
with time) leads to the conservation of energy. 
These laws are also symmetric with respect to 
space, rotation and relative motion. Initially 
explored in the early 1970s, supersymmetry is a 
less obvious kind of symmetry, which, if it exists in 
nature, would mean that the laws of physics do 
not change when bosons are replaced by 
fermions, and fermions are replaced by 
bosons.
Although it is difficult to explain 
supersymmetry through analogies to classical 
physics, its consequences are dramatic — it 
predicts that every fundamental particle has a 
superpartner with half a unit of spin less. The 
electron, for instance, has a spin of a half, so its 
superpartner (which is known as a selectron) has 
zero spin. This means that the superpartner of a 
boson is always a fermion and vice versa.
Supersymmetry also plays a central role in 
theories that attempt to unify the forces in the 
standard model with gravity by treating 
fundamental particles as vibrating strings or 
membranes in 10-dimensional or 11-dimensional 
spacetimes. In these theories the gravitational force 
is carried by a spin-two boson called the graviton. 
Searching for supersymmetric particles will be a 
priority when the Large Hadron Collider comes into 
operation at CERN, the European particle-physics 
laboratory near Geneva, in 2008. 
Peter Rodgers, Chief Editor, 
Nature Nanotechnology
ORIGINAL RESEARCH PAPERS Golfand, Y. A. & Likhtman, E. P. 
Extension of the algebra of Poincaré group generators and viola-
tion of P invariance. JETP Lett. 13, 323–326 (1971) | Neveu, A. & 
Schwarz, J. H. Factorizable dual model of pions. Nucl. Phys. B 31, 
86–112 (1971) | Ramond, P. Dual theory for free fermions. Phys. 
Rev. D 3, 2415–2418 (1971) | Wess, J. & Zumino, B. Supergauge 
transformations in four dimensions. Nucl. Phys. B 70, 39–50 (1974) | 
Wess, J. & Zumino, B. A Lagrangian model invariant under super-
gauge transformations. Phys. Lett. B 49, 52–54 (1974)
fuRtHER REAdING Dimopoulos, S., Raby, S. & Wilczek, F. 
Supersymmetry and the scale of unification. Phys. Rev. D 24, 
1681–1683 (1981) | Almadi, U., de Boar, W. & Furstenau, H. 
Comparison of grand unified theories with electroweak and strong 
coupling constants measured at LEP. Phys. Lett. B 260, 447–455 
(1991) | Greene, B. The Elegant Universe (Vintage, London, 2000) | 
Kane, G. & Shifman, M. (eds) The Supersymmetric World: The 
Beginnings of the Theory (World Scientific, Singapore, 2000)
 m I L E S tO N E 1 3
Super		
symmetry
The ATLAS experiment under construction at the Large Hadron 
Collider. Image courtesy of CERN.
The ideas and methods developed 
by Kondo and his fellow theorists 
turned out to be relevant to a wide 
range of problems that involve strong 
interactions between particles. As a 
result, the ‘Kondo effect’ — which, in 
truth, comprises a range of phenomena 
to do with collective behaviour arising 
from localized magnetic impurities 
— is an active research topic today and 
one that still throws up surprises.
Liesbeth Venema, Senior Editor, Nature
ORIGINAL RESEARCH PAPERS Kondo, J. 
Resistance minimum in dilute magnetic alloys. 
Prog. Theor. Phys. 32, 37–49 (1964) | Anderson, 
P. W. A poor man’s derivation of scaling laws 
for the Kondo problem.	J. Phys. C 3, 2346–2441 
(1970) | Goldhaber-Gordon, D. et al. Kondo 
effect in a single-electron transistor. Nature 
391, 156–159 (1998) 
fuRtHER REAdING Wilson, K. G. The 
renormalization group: critical phenomena and 
the Kondo problem.	Rev. Mod. Phys. 47, 773–840 
(1975) | Tsvelik, A. M. & Wiegmann, P. B. Exact 
results in the theory of magnetic alloys.	Adv. 
Phys. 32, 453–713 (1983) | Kouwenhoven, L. & 
Glazman, L. Revival of the Kondo effect.
Phys. World 14(1), 33–38 (2001)
outset — hence, quantum mechanics 
must be incomplete. 
a decisive step came in 1964 
when John Bell, building on the 
Bohm–aharonov formulation in 
spin variables, showed that quantum 
mechanics makes predictions that 
contradict the local-realistic world 
view of einstein and do require 
action at a distance of some sort. 
Bell’s theorem has been put to the 
test many times since, and although 
there is, as yet, no single experiment 
that closes all possible loopholes, the 
weight of evidence does still favour 
quantum mechanics. 
Andreas Trabesinger, Senior Editor, 
Nature Physics
ORIGINAL RESEARCH PAPERS Einstein, A., 
Podolsky, B. & Rosen, N. Can quantum-
mechanical description of physical reality be 
considered complete? Phys. Rev. 47, 777–780 
(1935) | Bohm, D. Quantum Theory Ch. XXII 
(Prentice-Hall, Englewood Cliffs, New Jersey, 
1951) | Bohm, D. & Aharonov, Y. Discussion of 
experimental proof for the paradox of Einstein, 
Rosen, and Podolsky. Phys. Rev. 108, 1070–1076 
(1957) | Bell, J. S. On the Einstein Podolsky Rosen 
paradox. Physics 1, 195–200 (1964)
fuRtHER REAdING Bell, J. S. Bertlmann’s socks 
and the nature of reality. J. Phys. 42, 41–62 (1981) | 
Sauer, T. An Einstein manuscript on the EPR 
paradox for spin observables. Stud. Hist. Philos. 
Mod. Phys. 38, 879–887 (2007)
Nature MILeStONeS | Spin	 March 2008 | S13
© 2008 Nature Publishing Group 
 
the fact that a dirty metal can support 
an electric current flowing without 
resistance might sound exotic to 
some, but it is a textbook property of 
‘conventional’ superconductors. When 
spins are involved, however, the super-
conductors do become ‘exotic’.
the conventional mechanism 
of superconductivity was explained 
by John Bardeen, Leon cooper and 
robert Schrieffer (in what is known 
as the BcS theory) back in 1957. 
For decades, it was a mystery how 
electrons, which are classified as 
‘fermions’, could be forced into a single 
ground state that was more typical 
for ‘bosons’ (as when they undergo 
Bose–einstein condensation). 
Fermions cannot all pile into the same 
ground state because only two — one 
with its spin pointing upwards and 
the other pointing downwards — can 
occupy each quantum state; bosons do 
not heed such conventions. as it turns 
 m I L E S tO N E 1 4
Sticking	together
The scientific principles of magnetic 
resonance imaging (MRI) stem from those 
of nuclear magnetic resonance (NMR; 
Milestone 8); however, now, especially in 
the mind of the public, the latter lies very 
much in the shadow of the former. The 
technological jump from NMR spectrum 
to MR image began in the early 1970s, 
and subsequent developments have 
established MRI as a priceless technique 
in medical research and diagnostics.
MRI uses magnetic fields and 
radiowaves to produce, in a non-invasive 
manner, ‘tomographic’ images of a 
three-dimensional object. Paul Lauterbur 
introduced the scientific basis behind 
this mode of visualization as “image 
formation by induced local interactions”. 
His idea was to combine two magnetic 
fields, so that one induces an interaction 
whereas the other restricts this 
interaction to a localized region in space. 
He proposed the term ‘zeugmatography’ 
to describe the technique, from a Greek 
word meaning ‘that which is used for 
joining’; however, the name never 
became widely accepted.
Lauterbur’s pioneering experiment 
involved the imaging of a cross-section 
throughtwo glass tubes of ordinary 
water (H2O) attached to the inside 
wall of a larger tube of deuterated 
water (D2O). A two-dimensional image 
 m I L E S tO N E 1 5
From spectrum to snapshot
showing the location of the tubes of 
H2O was generated by combining four 
projections taken from different angles 
around the set-up.
Some scepticism surrounded this initial 
observation. It seemed counterintuitive 
that radiowaves could be used to image 
objects that were much smaller than 
their wavelength. Yet it was because the 
interactions were restricted to certain 
regions that, in fact, the technique 
became independent of wavelength. 
Lauterbur recognized the potential 
of the concept at the time of the first 
simple experiment. He believed it could 
be used to investigate complex systems 
and noted the possibility of visualizing 
biological tissues — in particular, 
distinguishing between malignant 
tumours and healthy tissue. Many of the 
early practical developments in MRI were 
made by Peter Mansfield, who discovered 
how to acquire images rapidly. In 
recognition of their contributions, 
Lauterbur and Mansfield shared the 2003 
Nobel Prize in Medicine.
The invention of MRI has not been 
without controversy. Others have 
claimed to have produced the first 
out, the only way for fermions to form 
a condensate is for them to pair up, 
with the help of the crystalline lattice: 
when one electron passes through the 
lattice, the positive ions are slightly 
attracted to the passing negative elec-
tron; if a second electron comes along, 
it will sense the deformed lattice and 
be attracted to the net positive charge, 
and hence to the original electron. 
this kind of lattice-assisted coupling 
is weak, but it is strong enough for the 
paired electrons to drop down to a 
collective ground state. 
Naturally, people considered 
whether such pairing could glue 
other fermions together, in particular 
3he. as 4he exhibits superfluidity 
— that is, the liquid can flow without 
viscosity below a certain temperature 
— it was believed (hoped) that 3he 
would do likewise. Yet, given that 4he 
is a boson and 3he is a fermion, it was 
not clear how 3he could condense 
until the BcS theory came along. 
however, the magnetic interactions 
between the 3he particles are strong, 
and so they cannot pair up as 
electrons do. Instead, the pairing glue 
must come from another source.
the first authors to propose 
ferromagnetic fluctuations of the 
spins as such a glue were a. Layzer 
and D. Fay. When one particle 
whizzes through the liquid, its spin 
(pointing upwards, for instance) 
attracts other spin-up particles and 
repels spin-down particles. the effec-
tive spin-up polarization can then 
attract another spin-up particle, lead-
ing to a spin-up–spin-up pair. When 
coupled in this way, the 3he atoms 
are able to form a superfluid.
amazingly, all the theoretical 
groundwork, including that by 
Philip anderson and Pierre Morel, 
and by roger Balian and richard 
Werthamer, was laid down before 
the experimental confirma-
tion of a superfluid state in 3he 
— that came in 1972 and earned its 
authors, Douglas Osheroff, robert 
richardson and David Lee, the Nobel 
prize in Physics in 1996.
although superfluidity was antici-
pated, measurements revealed several 
unique superfluid phases in 3he. 
Moreover, because of the non-zero 
spin of the pairs (in conventional 
superconductors the net spin is zero), 
Milestones
S14 | March 2008 	www.nature.com/milestones/spin
© 2008 Nature Publishing Group 
 
In the mid-1970s, major advances in solution 
nuclear magnetic resonance (NMR) spectroscopy 
set the stage for a revolutionary new application: 
solving the three-dimensional (3D) structures of 
proteins in the solution state. At the time, X-ray 
crystallography was already a well-established 
tool for the determination of protein crystal 
structures. Unlike crystallography, NMR does not 
require proteins to form diffracting crystals and 
this broadens the range of proteins that can be 
investigated. Furthermore, most proteins exist 
naturally in a solution state, or in contact with 
fluids, so knowledge of their properties in their 
native environment has physiological relevance.
A unique strength of NMR is its ability to 
supplement molecular structures with 
information on dynamic processes, which may be 
influenced, for example, by ligand binding in 
solution: even before structure determination by 
NMR became feasible, the technique was used to 
obtain information about protein dynamics. In 
1971, Adam Allerhand and colleagues 
demonstrated the existence of sub-nanosecond 
segmental motions in proteins, and by the middle 
of the decade, the groups of Brian Sykes, Robert J. 
P. Williams and Kurt Wüthrich presented evidence 
for lower-frequency motions in globular proteins. 
Two fundamental advances in the late 1970s 
set the scene for the development of NMR into a 
method for determining previously unknown 
protein structures (rather than refining 
incomplete structures). First, Richard Ernst, 
building on a breakthrough idea by Jean Jeener, 
demonstrated the principle of two-dimensional 
(2D) NMR spectroscopy. This technique, which 
also applies to other spectroscopies, allowed 
researchers to record not only chemical shifts 
(Milestone 10) but also the interactions between 
pairs of nuclear spins — it later won Ernst the 
1991 Nobel Prize in Chemistry. Second, Wüthrich 
discovered that the nuclear Overhauser effect 
could be exploited in NMR experiments with 
proteins, allowing the mapping of networks of 
near-by atom pairs that are not connected 
through covalent bonds. Beginning in 1976, 
Wüthrich and Ernst joined forces, and with 
Kuniaki Nagayama and Anil Kumar they 
developed a number of 2D NMR experiments, 
which became the basis for solving protein 
structures.
In 1982, Gerhard Wagner and Wüthrich 
published the sequence-specific assignments for 
a small protein, basic pancreatic trypsin inhibitor. 
Meanwhile, Werner Braun and Timothy Havel in 
the Wüthrich group were developing algorithms 
and software capable of calculating protein 
structures from NMR data. In 1985, Michael 
Williamson, Havel and Wüthrich reported the first 
solution-state protein structure — that of 
proteinase inhibitor IIA from bull seminal plasma. 
The results were met with disbelief. It was not 
until several structures solved initially using NMR 
were solved again using crystallography that the 
NMR technique was accepted. In 2002, Wüthrich 
was rewarded with the Nobel Prize in Chemistry. 
NMR has become a powerful technique for 
protein structure determination. Numerous 
advances made during the past two decades — 
including the development of three- and four-
dimensional spectroscopy, isotope labelling 
methods, and increases in magnetic field strength 
— have raised	the limit on the size of proteins that 
can be investigated. Currently, about 10% of 
structures being deposited in the Protein Data 
Bank are solved using NMR. Perhaps the most 
exciting frontier is the application of NMR to 
investigate protein dynamics, especially for large 
molecular machines, which will undoubtedly lead 
to new insights in biology.
Allison Doerr, Associate Editor, Nature Methods
ORIGINAL RESEARCH PAPERS Allerhand, A. et al. Conformation 
and segmental motion of native and denatured ribonuclease A in 
solution. Application of natural-abundance carbon-13 partially 
relaxed Fourier transform nuclear magnetic resonance. J. Am. Chem. 
Soc. 93, 544–546 (1971) | Wüthrich, K. & Wagner, G. NMR 
investigations of the dynamics of the aromatic amino acid residues in 
the basic pancreatic trypsin inhibitor. FEBS Lett. 50, 265–268 (1975) | 
Dobson, C. M., Moore, G. R. & Williams, R. J. P. Assignment of aromatic 
amino acid PMR resonances of horse ferricytochrome c. FEBS Lett. 51, 
60–65 (1975) | Snyder, G. H., Rowan, R. & Sykes, B. D. Complete 
tyrosine assignments in the high-field proton nuclear magnetic 
resonance spectrum of bovine pancreatictrypsin inhibitor selectively 
reduced and carboxamidomethylated at cystine 14-38. Biochemistry 
15, 2275–2283 (1976) | Aue, W. P., Bartholdi, E. & Ernst, R. R. 
Two-dimensional spectroscopy. Application to nuclear magnetic 
resonance. J. Chem. Phys. 64, 2229–2246 (1976) | Nagayama, K., 
Wüthrich, K., Bachmann, P. & Ernst, R.R. Two-dimensional J-resolved 
1H NMR spectroscopy for studies of biological macromolecules. 
Biochem. Biophys. Res. Commun. 78, 99–105 (1977) | Kumar, A., 
Ernst, R. R. & Wüthrich, K. A two-dimensional nuclear Overhauser 
enhancement (2D NOE) experiment for the elucidation of complete 
proton–proton cross-relaxation networks in biological 
macromolecules. Biochem. Biophys. Res. Commun. 95, 1–6 (1980) | 
Wagner, G. & Wüthrich, K. Sequential resonance assignments in 
protein 1H nuclear magnetic resonance spectra: basic pancreatic 
trypsin inhibitor. J. Mol. Biol. 155, 347–366 (1982) | Williamson, M. P., 
Havel, T. F. & Wüthrich, K. Solution conformation of proteinase 
inhibitor IIA from bull seminal plasma by 1H nuclear magnetic 
resonance and distance geometry. J. Mol. Biol. 182, 295–315 (1985)
 m I L E S tO N E 1 6
Solution for solution structures 
‘NMR image’, most notably Raymond 
Damadian, who had reported in 1971 the 
ability to distinguish between normal 
tissue and tumours using magnetic 
resonance. On the announcement of the 
prize, he fervently disputed the decision 
of the Nobel committee.
The unquestionable fact remains that, 
although the invention of MRI — with 
its marriage of magnetic fields — took 
scientists by surprise at its conception, 
in its more recent lifetime it has 
proved to be an invaluable tool for the 
medical world. 
Alison Stoddart, 
Associate Editor, Nature Materials
ORIGINAL RESEARCH PAPERS Damadian, R. 
Tumor detection by nuclear magnetic 
resonance. Science 171, 1151–1153 (1971) | 
Lauterbur, P. C. Image formation by induced 
local interactions: examples employing nuclear 
magnetic resonance. Nature 242, 
190–191 (1973) | Mansfield, P. & Grannell, P. K. 
NMR ‘diffraction’ in solids? J. Phys. C 6, 
L422–L426 (1973) | Mansfield, P., Garroway, A. N. 
& Grannell, P. K. Image formation in NMR by a 
selective irradiative process. J. Phys. C 7, 
L457–L462 (1974) | Mansfield, P. Multi-planar 
imaging formation using NMR spin echoes. 
J. Phys. C 10, L55–L58 (1977)
3he also yielded some unexpected 
properties. In 1987, a team working 
in Moscow discovered a pure spin 
supercurrent. unlike the supercur-
rent in a conventional superconduc-
tor that carries charge and mass, the 
spin supercurrent carries only spin 
and there is no mass flow. 3he is truly 
exotic, because of its spin.
May Chiao, Senior Editor, 
Nature Physics
ORIGINAL RESEARCH PAPERS Bardeen, J., 
Cooper, L. N. & Schrieffer, J. R. Microscopic theory 
of superconductivity. Phys. Rev. 106, 162–164 
(1957) | Anderson, P. W. & Morel, P. Generalized 
Bardeen–Cooper–Schrieffer states and the 
proposed low-temperature phase of liquid He3. 
Phys. Rev. 123, 1911–1934 (1961) | Balian, R. & 
Werthamer, N. R. Superconductivity with pairs in 
a relative p wave. Phys. Rev. 131, 1553–1564 
(1963) | Layzer, A. & Fay, D. Superconducting 
pairing tendency in nearly ferromagnetic 
systems. Int. J. Magn. 1, 135–141 (1971) | Osheroff, 
D. D., Richardson, R. C. & Lee, D. M. Evidence for a 
new phase of solid He3. Phys. Rev. Lett. 28, 
885–888 (1972) | Osheroff, D. D., Gully, W. J., 
Richardson, R. C. & Lee, D. M. New magnetic 
phenomena in liquid He3 below 3 mK. Phys. Rev. 
Lett. 29,	920–923 (1972) | Borovik–Romanov, A. S., 
Bun’kov, Yu. M., Dmitriev, V. V. & Mukharskii, 
Yu. M. Observation of phase slippage during the 
flow of a superfluid spin current in 3He-B. JETP 
Lett. 45,	124–128 (1987)
fuRtHER REAdING Leggett, A. J. A theoretical 
description of the new phases of liquid 3He 
Rev. Mod. Phys. 47, 331–414 (1975)
Milestones
Nature MILeStONeS | Spin	 March 2008 | S15
NMR has become a powerful 
technique for protein structure 
determination.
© 2008 Nature Publishing Group 
 
 m I L E S tO N E 1 8
A giant leap for 
electronics
Towards the end of the 1960s, scientists 
had begun exploring the technological 
potential of magnetism combined 
with semiconductor physics. Having 
succeeded in introducing small amounts 
of magnetic impurities into otherwise 
non-magnetic semiconductors, 
Robert Gałazka and colleagues 
presented, in 1978, remarkable data 
on II–VI compounds doped with 
manganese. In these ‘diluted magnetic 
semiconductors’ (DMSs), the low-
concentration defects (the manganese 
ions) did not compromise the quality of 
the material, meaning that its magneto-
optical and magneto-transport 
properties could be probed. At the same 
time, pronounced magnetic properties 
could be observed — such as the 
spin splitting of electronic or impurity 
bands.
A further breakthrough came in the 
early 1990s, with the advent of low-
temperature molecular-beam epitaxy. 
By growing semiconductor films under 
conditions that were far from thermal 
equilibrium, it became possible to 
introduce manganese impurities into 
III–V materials (which is more difficult 
under equilibrium conditions owing to 
the low solubility of manganese). Hideo 
Ohno and colleagues then demonstrated, 
in 1992, ferromagnetic order in the DMS 
(In,Mn)As — indium arsenide containing 
only 1.3% manganese — by measuring 
magneto-transport and, in particular, an 
anomalous Hall effect in the material. 
The work was followed up, in 1996, 
with proof of ferromagnetism in doped 
gallium arsenide — (Ga,Mn)As — at 
temperatures up to 110 K. As GaAs can be 
used in electronic devices that operate 
at room temperature, these studies 
established the basis for research into 
‘technologically relevant’ DMSs.
The 1990s also brought theoretical 
work by Tomasz Dietl, in collaboration 
with Ohno’s group, which explained the 
origin of ferromagnetism in (Ga,Mn)As 
using a model developed, by Clarence 
 m I L E S tO N E 1 7
Dilute for 
impact
In retrospect, it seems surprising that, 
although spin and charge are two of 
the most fundamental properties of 
electrons, the advantage that could 
be gained from combining them in a 
consumer device was only realized 
in the 1990s, when IBM introduced a 
new type of hard-disk drive that would 
revolutionize data storage. 
Crucial to this technological revolution 
was an earlier discovery made by the 
groups of Albert Fert and Peter Grünberg 
— for which the two won the 2007 
Nobel Prize in Physics. In 1988, they had 
observed a large change in the electrical 
resistance of thin metal layers as a 
function of an external magnetic field. 
Although bulk magnetoresistive effects 
had been known for more than a century 
(discovered by William Thomson, later to 
become Lord Kelvin), they were usually 
only moderate. However, in the studies 
led by Fert and Grünberg, the effect was 
much more pronounced. From the outset, 
these thin magnetic multilayer structures 
Zener in 1950, for ferromagnetism in 
transition metals. According to this 
model, the magnetic order originates 
from the delocalized holes that mediate 
the interaction between localized 
magnetic moments. The importance 
of the work was twofold. First, the 
carrier-mediated magnetic order 
suggested the possibility of controlling 
the ferromagnetism using electric 
fields — which was soon demonstrated 
— and, beyond that, the development 
of efficient spintronics devices. Second, 
the Dietl model provided an effective 
recipe for calculating the Curie 
temperature of other zinc-blende and 
wurzite semiconductors, to advance the 
search for a room-temperature DMS. In 
particular, Dietl showed that DMSs based 
on zinc oxide (ZnO) or gallium nitride 
(GaN) could have Curie temperatures as 
high as 300 K.
Investigations since then have 
indeed revealed room-temperature 
ferromagnetism in oxides and 
semiconductors that include ZnO or 
GaN. However, it is yet to be proved 
that the carrier-mediated mechanism 
proposed by Dietlis really at work 
showed magnetoresistance of up to 50%; 
the phenomenon was dubbed ‘giant’ 
magnetoresistance (GMR).
GMR is based on the spin-dependent 
scattering of electrons travelling across 
metallic thin films. In its most basic 
realization, a GMR device consists of two 
thin magnetic metal films, separated by 
a non-magnetic metal. If the magnetic 
layers have a different magnetic 
orientation with respect to each other, 
the electrons scatter strongly in the 
trilayer structure and the electrical 
resistance is high. However, once the 
magnetic orientation of the magnetic 
layers is aligned using an external 
magnetic field, electrons with spins 
antiparallel to that direction scatter 
much less, and move more easily between 
the magnetic and non-magnetic layers 
— hence, the electrical resistance is low.
This groundbreaking discovery quickly 
led to the use of GMR to miniaturize 
the recording heads of hard-disk drives. 
IBM had already, in 1991, developed 
a hard drive based on the smaller bulk 
magnetoresistance effect; in 1997, 
thanks largely to the efforts of Stuart 
Parkin and colleagues in the IBM 
laboratories, the first hard drives based 
on GMR were commercialized. 
More recently, a new magnetoresistive 
Image courtesy of P. M. Koenraad, 
Eindhoven University of Technology. 
device has been incorporated into 
hard-disk drives — the magnetic tunnel 
junction. Magnetic tunnel junctions 
(introduced in 1975 by Michel Jullière) 
are similar in structure to the GMR 
trilayers, except that the metallic 
non-magnetic layer is replaced by an 
insulating layer. However, it was only in 
1995, following advances in techniques 
for growing materials, that Jagadeesh 
Moodera and colleagues, and Terunobu 
Miyazaki and Nobuki Tezuka, were 
able to realize tunnel junctions with 
practical magnetoresistance. Further 
work by Parkin et al. and by Shinji Yuasa 
and colleagues, in 2004, proved that 
satisfactory room-temperature operation 
could be achieved when the barrier layer 
was made of magnesium oxide.
The advance made in reducing the size 
of read and write heads has raised the 
hope that such spin-electronic effects 
might also be used for solid-state data 
storage. Indeed, the tunnel-junction 
structure is a useful template for such 
magnetoresistive random-access memory 
(MRAM) devices, as the two possible 
relative orientations of the magnetic 
layers could be interpreted as ‘bits’ in a 
storage device. Moreover, new concepts 
for devices are evolving continually 
(see also Milestone 20) — for example, 
S16 | March 2008 	www.nature.com/milestones/spin
© 2008 Nature Publishing Group 
 
in these systems. The origin of this 
ferromagnetism — and its potential 
use in spintronics devices — is still a 
matter of controversy, yet the search for 
a carrier-mediated room-temperature 
DMS continues.
Fabio Pulizzi, Associate Editor, 
Nature Materials
ORIGINAL RESEARCH PAPERS Von Molnar, S. & 
Methfessel, S. Giant negative magnetoresistance 
in ferromagnetic Eu1–xGdxSe. J. Appl. Phys. 38, 
959–964 (1967) | Gałązka, R. R. in Proceedings of the 
14th International Conference on the Physics of 
Semiconductors (ed. Wilson, B. L. H.) 133 (Institute 
of Physics, Bristol, 1978) | Munekata, H. et al. 
Diluted magnetic III–V semiconductors. 
Phys. Rev. Lett. 63, 1849–1852 (1989) | Ohno, H., 
Munekata, H., Penny, T., von Molnár, S. & 
Chang, L. L. Magnetotransport properties of 
p-type (In,Mn)As diluted magnetic III–IV 
semiconductors. Phys. Rev. Lett. 68, 2664–2667 
(1992) | Ohno, H. et al. (Ga,Mn)As: a new 
ferromagnetic semiconductor based on GaAs 
Appl. Phys. Lett. 69, 363–365 (1996) | Dietl, T., Ohno, 
H., Matsukura, F., Cibert, J. & Ferrand, D. Zener 
model description of ferromagnetism in zinc-
blende magnetic semiconductors. Science 287, 
1019–1022 (2000) | Ohno, H. et al. Electric-field 
control of ferromagnetism. Nature 408, 944–946 
(2000) 
fuRtHER REAdING Furdyna, J. K. & 
Kossut, J. (eds) Diluted Magnetic Semiconductors, 
Semiconductors and Semimetals Vol. 25 
(Academic, London, 1988)
By the late 1980s, magnetic resonance imaging 
(MRI; Milestone 15) had become a standard 
technique in hospitals and laboratories for the 
anatomical imaging of various tissues, from muscle 
to brain. However, MRI was only capable of revealing 
static structure and physiochemical information, 
but not actual function. The options for functional 
imaging were generally cumbersome, with low 
spatial resolution, and could require the injection 
of radioactive tracers into the bloodstream, as 
in positron emission tomography (PET), meaning 
that individual subjects could be scanned only 
infrequently. In 1990, however, Seiji Ogawa and 
colleagues published a series of breakthroughs that 
transformed MRI into a non-invasive and relatively 
inexpensive means of revealing physiological 
activity in the brain, sparking a revolution in the 
study of brain and behaviour. 
Ogawa et al. exploited two physiological 
phenomena that stemmed from observations made 
years earlier. First, in 1890, Charles Roy and Charles 
Sherrington had suggested that metabolic activity 
in the brain could be linked to vascular changes that 
would refresh blood supply. Later work established 
a more refined view: vasculature responds in an 
exquisitely localized fashion to bring oxygenated 
blood to areas of increased neural activity. This 
phenomenon was already being exploited in 
PET and other techniques. Second, Linus Pauling 
and Charles Coryell had reported, in 1936, that 
haemoglobin — the metalloprotein in red blood 
cells that acts as a major transporter of oxygen in 
humans and other species — has different magnetic 
properties in its oxygenated and deoxygenated 
forms. In three papers published in 1990, Ogawa 
and colleagues now showed how these two 
phenomena could be detected using MRI. 
First, they demonstrated that changes in the 
level of deoxygenated haemoglobin in blood 
changed the proton signal from the water molecules 
surrounding the vessels — an effect called blood-
oxygenation-level-dependent (BOLD) contrast. 
Because metabolic activity in the brain involves 
changes in the relative levels of oxyhaemoglobin 
and deoxyhaemoglobin, Ogawa reasoned that it 
should be possible to track changes in brain activity 
by measuring the BOLD contrast — and in his 
third paper of 1990, he demonstrated exactly this. 
Manipulation of the brain metabolism and, hence, 
the blood oxygen of an anaesthetized rat — by 
adjusting anaesthesia or the composition of inhaled 
gas, or by inducing hypoglycaemia — led to changes 
in the BOLD contrast throughout the brain. 
Two years after this ground-breaking proof of 
principle, three independent groups (including 
that of Ogawa) published, almost simultaneously, 
demonstrations of task-related changes in the BOLD 
contrast in the human brain — proving not only that 
this method could be translated from anaesthetized 
animals to awake humans, but also that it could reveal 
localized brain function evoked by specific stimuli, 
such as visual images. 
BOLD-contrast imaging — or functional MRI (fMRI) 
as the technique is now commonly known — quickly 
became a mainstay of cognitive neuroscience. It is an 
accessible option for measuring brain activity with 
relatively high spatial resolution — resolution that 
has improved with advances in MRI hardware and 
techniques, and analysis methods. From the detailed 
characterization of the function of human visual brain 
areas to the discovery of areas that are potentially 
involved in higher cognitive functions, such as 
face recognition, empathy and self-awareness, the 
possibilities revealed by fMRI seem endless.
I-han Chou, Senior Editor, Nature
ORIGINAL RESEARCH PAPERS Roy, C. S. & Sherrington, C. S. 
On the regulation of the blood supply of the brain. J. Physiol. 
11, 85−108 (1890) | Pauling, L. & Coryell, C. D. The magnetic 
properties and structure of hemoglobin, oxyhemoglobin and 
carbonmonoxyhemoglobin. Proc. Natl Acad.