Nature Milestones Photons

8/8/2019 Nature Milestones Photons

1/16

In th ginning or, at last, from arondth sixth ntry bce th Vaishshika

shool of Hind philosophy hld that thworld was asd on th atoms of arth, air,fir and watr. Rays of light wr thoght

to omposd of fast-moving fir atomsor tejas, with th haratristis of th lightdpnding on th spd and arrangmnt

of th tejas. Th natr of light whthr itindd som kind of partil or, instad, awav propagating throgh a mdim was

to om on of th gratst sintifidats of th sding ntris: on thatwas rsolvd arly a ntry ago.

Arond 300 bce, Elid didd that lighttravlld in straight lins, and dsrid thlaws of rfltion. In th sond ntry,

Ptolmy wrot aot rfration. Laws ofrfration wr formlatd y In al-Hay-tham (also known as Alhazn), who wrot hisKitab al-Manazir, or Book of Optics, in 1021.In al-Haytham was a prolifi xprimntal-ist, notaly stdying disprsion too. H alsothoght of light as a stram of mint parti-

ls, travlling at finit spd.

Rn Dsarts, howvr, had othr idas and many of thm, as fittd a Rnaissan

man. In 1637, alongsid his Discours de lamthode (with its mmoral qot, I think,thrfor I am), h plishd thr ssays, on

mtorology, gomtry and optis. This last,La dioptrique, promotd a onpt of light asplss propagating instantanosly throgh

th ontat of alls of som mdim (athr).Similar idas ar fond in Thomas HosTractatus opticus of 1644 and Rort Hooks

Micrographia of 1665. Althogh Igna GastonPardis is thoght to hav takn stps, arond

1670, in dvising a formal wav thory, th

mansript is lost. Howvr, ChristiaanHygns Trait de la lumire of 1690 srvivs.In it, h tratd light as omprssil wavs

in an lasti mdim, analogos to sond;y onsidring th nvlop of sond-ary wavlts, h showd how to onstrt

rfltd, rfratd and srnd wavs; h alsoxplaind dol rfration.

Hygns atifl work did not, howvr,

onqr th ida of light as partils or or-psls. Isaa Bkman, who was a mntor ofDsarts, and Pirr Gassndi ld a rvival

of Grk atomisti thoris, whih inlddth intrprtation of olor as a mixtr oflight and shadow. Bt it was Isaa Nwton

who am th grat hampion of thorpslarists. In his Opticks of 1704, h r-ognizd that olor shold orrspond to th

vloity or mass of th light partils, and thsxplaind why diffrnt olors ar rfratdy diffrnt amonts. H rjtd wav thory,

as light wold al to stray too farinto shadow; diffration h aontd foras th infltion of light partils y mattr.

Althogh Dsarts ndring rptationand Lonhard Elrs 1746 milston work(inlding a disprsion law) nsrd that wav

thory maintaind a following in Fran andGrmany, Nwtonian orpslar thory wasdominant for th rst of th ightnth ntry.

A frsh skirmish gan in th arly 1800s,with what is oftn onsidrd to on ofth most atifl dmonstrations in phys-

is: Thomas Yongs two-slit xprimnt,with whih h introdd th prinipl ofintrfrn for wavs of light. Bt now thorpslarists wr gaining grond in Fran:

polarization, displays of whih wr dlightingParisian salons, was onsidrd to d to

som kind of asymmtry among light orps-ls. Agstin Frsnl tippd th alan, witha pris wav thory of diffration. Having

rvisitd Hygns work and addd intrfr-n twn sondary wavs, h was alto xplain, in wav trms, how shadows form.

Morovr, in 1821, h showd that polarizationold xplaind if light wr a transvrswav, with no longitdinal viration. Now,

wav thory was all; Nwton was spplantd.Bt on prolm rmaind. Althogh

Maxwlls sminal qations of 1865

(Milestone 2) wr gradally and sss-flly adoptd in optis, th athr tospport ltromagnti filds, to yild

Frsnls laws of propagation was miss-ing. Th athr, of ors, wold nvr fond. As th twntith ntry dawnd,

a nw rvoltion in physis ld y MaxPlank (Milestone 3) and Alrt Einstin(Milestone 4) wold again hing on thnatr of light, it wav or partil. Or oth.

Alison Wright,

Chief Editor, Natr Physis

ORIGINAL RESEARCH PAPERS Descartes, R. La dioptrique(1637) | Huygens, C. Trait de la lumire o sont expliques lescauses de ce qui lui arrive dans la rflexion, et dans la

rfraction (1690) | Newton, I. Opticks: or a treatise of thereflections, refractions, inflections and colours of light (1704) |Euler, L. Novia theoria lucis et colorum. Opuscula variiargumenti1, 169244 (1746) | Young, T. Experiments andcalculations relative to physical optics. Phil. Trans. R. Soc.Lond.94, 116 (1804) | Fresnel, A. Mmoire sur la doublerfraction. Mmoires de lAcadmie des Sciences de lInstitutde France7, 45176 (1827)FuRtHER REAdING Frankel, E. Corpuscular optics and thewave theory of light: the science and politics of a revolutionin physics. Social Stud. Sci.6, 141184 (1976)

MILEStONE 1

Let there be light

NATuRE MILESTONES |Photons MAY 2010 |S5

MILESTONES

20 Macmillan Publishers Limited. All rights reserved10
http://www.nature.com/milestones/milephotons/full/milephotons02.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons03.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons04.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons04.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons03.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons02.html


2/16

By th middl of th nintnth ntry,a signifiant ody of xprimntal andthortial knowldg aot ltriity andmagntism had n amlatd. In 1861,

Jams Clrk Maxwll ondnsd it into20 qations. Maxwll plishd variosrdd and simplifid forms, t Olivr

Havisid is frqntly rditd with sim-plifying thm into th modrn st of forpartial diffrntial qations: Faradays law,Amprs law, Gass law for magntism

and Gass law for ltriity.On of th most important ontri-

tions mad y Maxwll was atally a

With the discovery of the principle of energyconservation, the edifice of theoretical physicsis fairly complete. There will be a mote to wipeout in a corner here or there, but somethingfundamentally new you wont find. So spokePhilipp von Jolly when, in 1877, his student MaxPlanck left Munich for Berlin, to spend his lastyear of studies there.

Planck, undeterred, went into theoreticalphysics not hoping to make new discoveries,but driven by his admiration of its elegance. Hismain interest was thermodynamics, but worksby Otto Lummer and Ernst Pringsheim, and

by Heinrich Rubens and Friedrich Kurlbaum,which aimed at constructing a standard forthe measurement of illumination intensities,directed him towards heat radiation. Herevisited Gustav Kirchhoffs theoretical studiesof black-body radiation, which implied thatwhen a substance capable of absorbing andemitting radiation is enclosed in a cavitywith perfectly reflecting walls, the spectraldistribution of the observed radiation atequilibrium is a function only of temperatureand is independent of the substance involved.Intrigued by such an absolute law, Plank

devoted himself, from 1896, to finding anexplanation for it.

Parallel works on black-body radiationproduced confusing results. Lord Rayleigh

MILEStONE 3

te quaum leap

had found a law (which, with James Jeans, helater refined) that well described the emissionspectrum at long wavelengths, but failed at shortones. By contrast, an earlier law by WilhelmWien describing the frequency position of the

radiation maximum which had been observedexperimentally, but was not reproduced by theRayleighJeans theory held for short, butnot for long, wavelengths. By October 1900,Planck had found a formula that interpolatedbetween the curve of Rayleigh and Jeans andthat of Wien. He sent his result, by postcard, toHeinrich Rubens, who immediately compared itto experimental data. It fitted all observationsperfectly. Spurred by the agreement, Planckset about finding the physical character of hisempirical formula.

On 14 December 1900, he presented theoutcome in a lecture given to the German

Physical Society. Planck had indeed found asound derivation to explain the behavioursdescribed by his formula, partially guided by thework of Ludwig Boltzmann on entropy. However,there was one revolutionary assumption that hehad to make: that light was emitted and absorbedin discrete packets of energy quanta. Thesewere not a feature of heat radiation alone, but,as Albert Einstein showed in 1905, also of light.Einstein used the term Lichtquant, or quantumof light. Only in 1926 was the word photonintroduced, by the chemist Gilbert Lewis. Histheory of a hypothetical new atom that is not

orrtion to Amprs law. H had ralizd

that magnti filds an indd yhanging ltri filds an insight thatwas not only nssary for aray t

also ld to a onptal rakthrogh.Maxwll prditd an ltromagntiwav, whih an slf-sstain, vn in a

vam, in th asn of onvntional

rrnts. Morovr, h prditd th spdof this wav to 310,740,000 m s1

within a fw prnt of th xat val ofth spd of light.

Th agrmnt of th rslts sms to

show that light and magntism ar aff-tions of th sam sstan, and light is anltromagnti distran propagatd

throgh th fild aording to ltromag-nti laws, wrot Maxwll in 1865. Th on-pt of light was ths nifid with ltriity

and magntism for th first tim.Maxwlls qations ar as important

today as vr. Thy ld to th dvlopmntof spial rlativity (Milestone 4) and,nowadays, almost vry optis prolmthat an formlatd in trms of dil-

tri prmittivity and magnti prmaility(two ky onstants in Maxwlls qations),ranging from optial-fir wavgids

MILEStONE 2

Claical mume

DavidPile

Image courtesy of Rudolf Dhrkoop

Milestones

S6 | MAY 2010 www.aure.cm/milee/p

http://www.nature.com/milestones/milephotons/full/milephotons04.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons04.html


3/16

light but plays an essential part in every processof radiation did not hold up, but the namephoton stuck.

Without setting out to do so, Planck hadrocked the edifice of physics to its very

foundations. His was, by nature, a conservativemind, wrote Max Born in an obituary of Planck,he had nothing of the revolutionary and wasthoroughly sceptical about speculations. Yethis belief in the compelling force of logicalreasoning from facts was so strong that hedid not flinch from announcing the mostrevolutionary idea which ever has shakenphysics.

Andreas Trabesinger,Senior Editor, Nature Physics

ORIGINAL RESEARCH PAPERS Wien, W. Ueber dieEnergievertheilung im Emissionsspectrum eines schwarzenKrpers.Ann. Phys. 294, 662669 (1896) | Rayleigh. Remarks

upon the law of complete radiation.Phil. Mag.49, 539540(1900) | Planck, M. Entropie und Temperatur strahlenderWrme.Ann. Phys.306, 719737 (1900) | Planck, M. Ueberdas Gesetz der Energieverteilung im Normalspectrum.Ann.Phys.309, 553563 (1901) | Planck, M. Ueber dieElementarquanta der Materie und der Elektricitt.Ann. Phys. 309, 564566 (1901) | Einstein, A. ber einen die Erzeugungund Verwandlung des Lichtes betreffenden heuristischenGesichtspunkt.Ann. Phys. 322, 132148 (1905) | Jeans, J. H.On the partition of energy between matter and aether.Phil.Mag.10, 9198 (1905) | Lewis, G. N. The conservation ofphotons. Nature118, 874875 (1926)FuRtHER REAdING Planck, M. Filmed self-portrayal[online] (1942) |Born, M. Max Planck. Obit. Not. Fellows R. Soc.6, 161188(1948) | Franck, J. Max Planck. Science107, 534537 (1948)

(Milestone 13) to mtamatrials and

transformation optis (Milestone 21),is tratd within th framwork of thsqations or systms of qations drivd

from thm.Thir atal soltion an, howvr,

hallnging for all t th most asi

physial gomtris. Nmrial mthodsfor solving th qations wr pionrdy Kan Y and Alln Taflov, t wntnnotid for many yars owing to th

limitd ompting powr availal at thtim. Now, howvr, ths mthods an asily mployd for solving ltromag-

nti prolms for strtrs as omplxas airraft.

David Pile,Associate Editor, Natr Photonis

ORIGINAL RESEARCH PAPERS Maxwell, J. C. On physicallines of force. Phil. Mag.11, 161175; 281291; 338348(1861); ibid. 12, 1224; 8595 (1862) | Maxwell, J. C. A

dynamical theory of the electromagnetic field.Phil. Trans. R.Soc. Lond.155, 459512 (1865) | Maxwell, J. C.A Treatise onElectricity and Magnetism (Clarendon Press, 1873) | Yee, K. S.Numerical solution of initial boundary value problemsinvolving Maxwells equations in isotropic media. IEEE Trans.Antenn. Propag. 14, 302307 (1966)FuRtHER REAdING Taflove, A. ComputationalElectrodynamics: The Finite-Difference Time-Domain Method (Artech House, 1995)

At the dawn of the twentieth century, lightwas thought to propagate through aether, amedium at absolute rest with respect to thefixed stars and transparent to the motion ofcelestial bodies. There can be no doubt thatthe interplanetary and interstellar spaces arenot empty but are occupied by a materialsubstance or body, which is certainly thelargest, and probably the most uniform,wrote James Clerk Maxwell in 1878. A clearproof of the existence of aether, however,could not be found.

In 1887, Albert Michelson and EdwardMorley published the results of arguably thebest known attempt to detect aether. Theiridea was that if light propagated along thedirection of motion of the Earth, its speedwould change owing to the velocity of ourplanet with respect to the aether. They usedan interferometer purposely designed byMichelson that had sufficient resolution todetect any expected effect. The result,however, was unequivocally null.

Explanations of the negative result

reported by Michelson and Morley wouldintroduce more complications. Thisbothered, not least, Albert Einstein, whotrusted that natural laws obey a universalharmony. From the failure to detect anyvariation in the speed of light in a vacuum,c, he concluded that this ought to be aconstant, regardless of the velocity withwhich the light source moved. He alsoassumed that the laws of physics should bethe same in reference frames moving withuniform translation with respect to oneanother. These two postulates were the

MILEStONE 4

Lig i pecialbasis of the theory he published in June1905, which is now known as specialrelativity.

Einstein derived the transformations ofspace and time coordinates between inertialreference frames, and reproduced equationsthat George FitzGerald and, independently,

Joseph Larmor and Hendrik Lorentz hadfound to make Maxwells equationsconsistent with Newtonian mechanics(which governs the laws of dynamics whenvelocities much lower than c are involved, asin everyday experience). The paper Einsteinpublished in June 1905 was followed by ashorter one in September of the same year,which featured the celebrated equivalencebetween energy and mass, E = mc2. Thespeed of light became the upper limit that nobody having finite mass at rest can reach, asit would need infinite energy.

As far as aether was concerned, specialrelativity made it vanish. As Einstein wrote inthe opening of his original paper, Theintroduction of a luminiferous ether will

prove to be superfluous inasmuch as the viewhere to be developed will not require anabsolutely stationary space provided withspecial properties.

Fabio Pulizzi,Senior Editor, Nature Materials

ORIGINAL RESEARCH PAPERS Maxwell, J. C. inEncyclopaedia Britannica 9th edn Vol. 8 (1878) | Michelson, A. A.& Morley, E. W. On the relative motion of the Earth and theluminiferous ether.Am. J. Sci.34, 333345 (1887) | Einstein, A.On the electrodynamics of moving bodies.Ann. Phys. 17,891921 (1905) | Einstein, A. Does the inertia of a body dependupon its energy content?Ann. Phys. 18, 639641 (1905)

GETTY

Milestones


http://www.nature.com/milestones/milephotons/full/milephotons13.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons21.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons21.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons13.html


4/16

On the basis of the quantumtheory a different hypothesis maybe formed, wrote Arthur HollyCompton in 1922, as he workedon interpreting his data on thescattering of X-rays by weakly-boundelectrons. He performed hisexperiments at a time when theboundary between particle and waverepresentations in physics was startingto blur. This boundary had been firmlyestablished by the mid-nineteenthcentury, after work by Thomas Youngand Augustin Fresnel had put particletheories of light to rest, and atomistictheories of matter consistent with newresults in chemical analysis had foundwidespread acceptance.

But by the beginning of the twentiethcentury, experiments on the interaction

of supposedly wave-like radiationwith particle-like matter had begun tochallenge this dichotomy (Milestone 3),and in 1905 Einstein proposed acompletely quantized picture of light.While Comptons X-ray scatteringexperiments came well afterwards,

they represented an important andindependent confirmation of Einsteinspicture: they delivered the first directevidence that the momentum of light,as well as its energy, was quantized.In the experiments, X-rays that werescattered from electrons increasedtheir wavelength to an extentdepending on the incident angle,but not on the incident wavelength.According to classical physics, theincoming radiation accelerated many

electrons simultaneously and over afinite period of time, and the change inwavelength could result from a Dopplereffect. Such an explanation led, however,to unrealistic electron-recoil velocities,and produced the wrong dependenceon scattering angle and incident

MILEStONE 5

Just a moment

Mh of th grondwork for whatwold om on of th most s-

ssfl thoris vr dvisd waslaid in th 1920s, partilarly in thwork of Pal Dira and th othr,

sal sspts Pasal Jordan,Wrnr Hisnrg and WolfgangPali. As qantm mhanis took

shap, Dira ommntd, in a 1927papr, that hardly anything has ndon p to th prsnt on qantm

ltrodynamis: for instan onhow to dsri th prodtion ofan ltromagnti fild y a moving

ltron and th ration of that fildon th ltron; and what happns

whn for propagats at th spd oflight rathr than instantanosly.

Dira dvisd a workal,non-rlativisti thory sing

a Hamiltonian to dsri thdynamial systm of an atom amidstlight-qanta. Bt th dvil was in th

dtails: it soon am apparnt thatattmpts at arat allation singqantm ltrodynamis wr st

y divrgns. If propr aont wrtakn of all trms in allating thmass or harg of an ltron ndr

th fft of an ltromagnti fild,th answr for ah was infinit.

Not ntil 1947 was th soltionfond. On th opning day of thShltr Island onfrn, in Jn of

that yar, Willis Lam prsntd hisdata showing th splitting of th 2S

1/2

and 2P1/2

ltron nrgy lvls of thhydrogn atom, prditd y Dirato dgnrat. On th train hom

from Shltr Island, Hans Bth ral-izd th signifian of this Lamshift that th ltron mass al-

latd in qantm ltrodynamiswas not th ltron mass masrdin xprimnt, and that a prodr

of rnormalization was nssary tolink th two.

By th tim of th Poono

onfrn in Marh 1948, othrshad workd ot xatly how to

aommodat rnormalization in arlativistially invariant thory. JlianShwingr prsntd, as was his wont,a thorogh mathmatial formla-

tion that was ssqntly fond tohav n mathd, indpndntlyin Japan, y Sin-Itiro Tomonaga.

Rihard Fynman, too, had his ownformlation to offr, althogh hlatr grmld that rnormaliza-

tion thory is simply a way to swpth diffiltis of th divrgns ofltrodynamis ndr th rg.

MILEStONE 6

QEDFrman Dyson soon provd th

qivaln of all thr approahs,and Shwingr, Tomonaga andFynman shard th Nol Priz in

Physis in 1965 for thir fndamntalwork in qantm ltrodynamis,with dp-ploghing onsqns

for th physis of lmntary parti-ls. Arat allation was at lastpossil, and was aidd gratly y th

illstrativ tool that Fynman hadprsntd at Poono: th Fynmandiagram. On of th first vr p-

lishd Fynman diagrams appardin his 1949 papr, showing th

wavelength. The different hypothesisof Compton instead postulated anelastic collision between a single photonof light and a single electron to whichit instantaneously delivered a singlequantum of momentum, leading to areduction in the energy and wavelength

of the photon.These experiments laid the foundationfor the modern quantum theory oflight. In 1929, the Compton effectbecame among the first phenomenato be modelled using quantumelectrodynamics, which would developinto one of the most tested and accurateof all physical theories (Milestone 6).This model, by Oskar Klein and YoshioNishina, applied Paul Diracs relativisticelectron equation, which had beendeveloped only the year before, toreproduce successfully the intensitiesand energies of Compton-scatteredX-rays.

Today, Compton-scattering effectsare found in a variety of pure andapplied disciplines. In medical radiology,

Willis Lamb, Abraham Pais, John Wheeler, Richard Feynman, Herman Feshbach andJulian Schwinger (left to right) at Shelter Island in 1947. Courtesy of AIP Emilio SegrVisual Archives

the most

stringently

tested, most

successful

theory in allof physics

Milestones




5/16

Holograms have become familiar, common even.They appear on credit cards and money, infashion shows, television programmes and worksof art, and beyond. Originally, the inventorDennis Gabor simply wanted to improve theelectron microscope itself a greatimprovement on the resolving power of the lightmicroscope in order to image an atomiclattice.

In 1947, electron microscopy was limited to aresolution of 12 , although the theoretical limitwas 5 . To get around the limiting factor, whichwas the electron lens, Gabor thought about thewave nature of light. Photographs record lightintensity. Suppose, however, that the phases oflight were also recorded? For that there wouldhave to be a reference phase with which tocompare the phase of the wave originating froman object. Interference of the reference andobject waves would create fringes, with maximarecorded on a photographic film where the twowaves are in phase. When this image isilluminated by the same reference wave, it willtransmit light only from the reference wave if it isidentical to the original object wave. Therefore,

the original object appears as a reconstructedimage, as if it were there. Using a mercury arclamp, with the reference source and the objecton the same axis, Gabor was able to reproduce agrainy two-dimensional image.

Unfortunately, Gabor was ahead of his time.His proposed holographic electron microscopesuffered from insufficient coherence of theelectron wave, which led to poorreconstructions. Little wonder that holographydid not become popular until after the inventionof the laser in 1960 (Milestone 9), whichprovided a supply of highly focused

monochromatic light. Within 2 years, hologramsexperienced a step change, literally gaining anextra dimension. Emmett Leith and JurisUpatnieks used a laser and an off-axisconfiguration to produce a three-dimensionalhologram, while Yuri Denisyuk createdthree-dimensional holograms using white lightas a source.

The shimmering futuristic-looking imagessoon spilled into science fiction, most notably inthe 1977 film Star Wars. Some 30 years later,technology has caught up. Companies marketsystems that create three-dimensionalholographic images that walk and talk, withoutthe audience having to wear special glasses. Thistechnique is also used in teleconference systems,where people can beam in from multiplelocations.

One of the most promising technologicalapplications, however, uses three-dimensionalholograms for data storage. Simply by varyingthe reference beam, pages of data can bewritten and then read from the same volume ofmaterial, with storage capacity in the terabyterange. That is equivalent to 100 films on a single

disc. With the ever increasing amount of digitaldata available, such as from the Large HadronCollider (set to produce 15 PB of data per year),we are going to need higher density recordingmedia to store them all.

May Chiao,Senior Editor, Nature Physics

ORIGINAL RESEARCH PAPERSGabor, D. A new microscopicprinciple.Nature161, 777778 (1948) | Leith, E. N. & Upatnieks, J.Reconstructed wavefronts and communication theory.J. Opt. Soc.Am.52, 11231130 (1962) | Denisyuk, Y. N. On the reflection ofoptical properties of an object in a wave field of li ght scattered byit. Doklady Akademii Nauk SSSR144, 12751278 (1962)

now-familiar spatim rprsnta-

tion, as straight and wavy lins, ofltrons xhanging a photon.

Qantm ltrodynamis

is now rognizd as an Aliangag thory with th symmtrygrop u(1). Its allations, sing

rnormalization, hav n shownto math xprimnt to th lvl,so far, of 1 in 1012. What gan

with Maxwlls qations for thltromagnti fild (Milestone 2)has om th most stringntly

tstd, most sssfl thory in allof physis. QED.

Alison Wright,Chief Editor, Natr Physis

ORIGINAL RESEARCH PAPERS Dirac, P. A. M.The quantum theory of the emission andabsorption of radiation. Proc. R. Soc. Lond. A114,243265 (1927) | Tomonaga, S. On a

relativistically invariant formulation of thequantum theory of wave fields. Prog. Theoret.Phys.1, 2742 (1946) | Koba, Z., Tati, S. &Tomonaga, S. Prog. Theoret. Phys.2, 101116(1947); ibid. 2, 198208 (1947) | Lamb, W. E. &Retherford, R. C. Fine structure of the hydrogenatom by a microwave method. Phys. Rev.72, 241243 (1947) | Bethe, H. A. The electromagneticshift of energy levels.Phys. Rev.72, 339341(1947) | Schwinger, J. Quantum electrodynamics I.A covariant formulation. Phys. Rev.74, 14391461(1948); ibid. 75, 651672 (1949); ibid. 76, 790817(1949) | Dyson, F. J. The radiation theories ofTomonaga, Schwinger and Feynman. Phys. Rev.75, 486502 (1949) | Feynman, R. P. Space-timeapproach to quantum electrodynamics. Phys. Rev.76, 769789 (1949)

Compton scattering can describe theinteraction of X-rays both with tissueand with a detector. Cosmic -raydetectors similarly exploit the effect,as do remote probes of extreme statesof matter such as accelerator beamsand high-density plasmas. Inverse

Compton scattering, which increasesthe energy of incident photons, hasbeen used to make bright and fastX-ray sources, and the Comptonformalism has been extended to thescattering of other objects, includingneutrons and subatomic particles.

Michael Segal, Associate Editor,Nature Nanotechnology

ORIGINAL RESEARCH PAPERS Compton,A. H. A quantum theory of the scattering ofX-rays by light elements. Phys. Rev.21, 483502(1923) | Klein, O. & Nishina, Y. ber die Streuungvon Strahlung durch freie Elektronen nach derneuen relativistischen Quantendynamik von

Dirac.Z. Phys. 52, 853868 (1929)FuRtHER REAdING Stuewer, R. H. TheCompton Effect (Science History Publishers,1975) | Messiah, A. Quantum Mechanics (DoverPublications, 1999)

MILEStONE 7

Ghosts of images past

CourtesyofLUCASFILML

td.

Milestones




6/16

Th onvrsion of snlight into ltriity is

on of th most natral pathways to xtratnrgy from th world arond s, andhas n stdid sin at last 1839 whn

Alxandr-Edmond Bqrl osrvdth photoltri fft in rdimntaryltrohmial lls. Th dvlopmnt of

modrn solar lls gan in arnst in 1939,with th aidntal disovry mad y RssllOhl of th pn jntion at Bll Laoratoris.

Whil masring th ltrial proprtis of asilion rystal ontaining a rak, h notid amarkd hang in ltri voltag dpndingon th illmination of th rystal.

Th gam hangr, howvr, was th

dvlopmnt in 1954 y Daryl Chapin,Calvin Fllr and Grald Parson, also at Bll

Laoratoris, of th first pratial solar ll.Having dvlopd a mthod to dop silion,thy wr al to fariat high-qality pn

jntions that, owing to thir prity, wrpartilarly ffiint in sparating th l-trial hargs ratd y th asord light.

Thir first solar lls ahivd an ffiinyof 6%, whih was an improvmnt y mor

than an ordr of magnitd ompard withaltrnativ dsigns.

A fndamntal ndrstanding of solar-ll

prforman was onsqntly rahd in1961 y William Shokly and Hans Qissr,who dtrmind th maximm thortiallight-onvrsion ffiiny of smiondtor

solar lls. Ovr th nxt dads, ffiin-is in light onvrsion improvd slowly asprr matrials old grown. Howvr, an

optimm opling of light into th lls andan ffiint xtration of ltrial arrirsot of th dvi ar also ssntial fators.

Improving th formr, Martin Grn at thunivrsity of Nw Soth Wals in Astraliadvlopd solar-ll dsigns that s invrtd

pyramids on th srfa to dirt light intoth silion mor fftivly.

A way of irmvnting th limitations of

th ShoklyQissr limit is to s mlti-jntion solar lls, in whih svral layrsof smiondtors with diffrnt andgaps

maximiz th asorption of solar light. Eahlayr in sh lls is optimizd for a spifisptral rgion. Ths lls ahiv ffiin-

is of >40%, yt thy rmain xpnsiv andar typially sd only in spa appliations.

MILEStONE 8

Sun power

There are only very few occasions whendiscoveries are made that start an entire newresearch field and at the same time revolutionizeour everyday life. The transistor is one example it led to modern electronics. At least asimportant is the invention of the laser, whichheralded the field of photonics.

The foundations of laser operation werelaid in 1917, when Albert Einstein studied theinteraction of electromagnetic radiation withelectrons that can occupy two energy levels. In

the presence of an incident photon equal to theenergy separating the two states, an electronin the higher state can be stimulated to relax,emitting a photon of the same energy as theincident one. The photons are coherent, that is,they have not only the same wavelength but alsothe same phase.

However, the fact that stimulated emissioncan amplify light fields to generate coherentlight beams was not realized until the 1950s.Then, Nikolay Basov and Alexander Prokhorovdeveloped the principle of the maser which stands for microwave amplification by

MILEStONE 9

All geer w

stimulated emission along with James Gordon,Herbert Zeiger and Charles Townes, whoindependently built the first maser in 1954. Their

maser used a microwave transition betweentwo energetic states of ammonia molecules.They sent a beam of ammonia molecules past anelectric field to focus excited molecules into amicrowave cavity, while defocusing the others.This provides an amplifier and oscillator thatemits coherent radiation.

An extension of the maser concept to opticallight waves was developed in 1958 by ArthurSchawlow and Townes. Gordon Gould, whocoined the term laser, is also credited withindependent contributions to the laser scheme,and after a prolonged court battle was granted asubsidiary patent on the laser. For their work that

led to the concept of masers and lasers, Townes,Basov and Prokhorov were awarded the 1964Nobel Prize in Physics.

After the first demonstration of the maser,a deluge of similar research papers floodedthe office ofPhysical Review, the editors ofwhich consequently decided to stop acceptingany further papers on the topic. So it camethat they also turned down the paper on thefirst working laser, which was demonstratedon 16 May 1960 by the 32-year-old physicistTheodore Maiman from Hughes ResearchLaboratories (pictured). Instead, Maiman

Manwhil, solar-ll thnologis hav

mrgd that do not s pn jntions.In 1991, Mihal Grtzl dvlopd dy-snsitizd solar lls, whih work y sing an

ltrolyt in ontat with a photosnsitizd

Gerald Pearson, Daryl Chapin and Calvin Fuller (from left toright), the inventors of the modern solar cell. Reprinted withpermission of Alcatel-Lucent USA Inc.

Milestones




7/16

sent his manuscript to Nature, where it waspublished in August 1960.

The Maiman laser was based on a ruby crystal

doped with chromium atoms to provide theenergy levels for the laser process. In order toexcite a sufficient number of electrons to passthe laser threshold, Maiman came up with theidea to use a bright flashlight as a pump source.It worked brilliantly.

The importance of lasers cannot be overstated.Among a plethora of applications, lasersare used in nonlinear optics (Milestones 7and 10), telecommunications (Milestone 13),optical disks (Milestone 15) and spectroscopy(Milestones 16 and 22). With the help of thelaser, photons have become a commodity theproperties of which can be designed almost

at will. This makes the laser one of the lastingachievements of modern science.

Joerg Heber, Senior Editor, Nature Materials

ORIGINAL RESEARCH PAPERS Einstein, A.Zur Quantentheorie der Strahlung.Physik. Zeitschr. 18, 121128 (1917) | Basov, N. G. & Prokhorov, A. M. Application ofmolecular clusters to radiospectroscopic study of rotationalspectra of molecules.Zh. Eksperim. i Teor. Fiz.27, 431438(1954) | Gordon, J. P., Zeiger, H. J. & Townes, C. H. Molecularmicrowave oscillator and new hyperfine structure in themicrowave spectrum of NH

3. Phys. Rev.95, 282284 (1954) |

Schawlow, A. L. & Townes, C. H. Infrared and optical masers.Phys. Rev. 112, 19401949 (1958) | Maiman, T. H. Stimulatedoptical radiation in ruby. Nature 187, 493494 (1960)

anod. Thir prforman rmains infrior to

that of silion lls t thy rprsnt a ost-ffiint altrnativ, althogh organi solarlls mad from smiondting polymrs and

fllrns, whih wr pionrd at arond thsam tim y Frd Wdl and ollags, vi forth sam markt.

With an inrasing rlvan of solar

nrgy in rnwal ltriity gnration,th dvlopmnt of novl and mor ffiintsolar-ll dsigns will ontin. For xampl,

som mrging solar lls mak s of nanos-trtrd arhittrs, sh as nanowirs, inwhih photognratd harg arrirs an

mor ffiintly olltd.

Stefano Tonzani,Associate Editor, Natr Commniations

ORIGINAL RESEARCH PAPERS Ohl, R. S. Light-sensitiveelectric device. US patent 2,402,662 (filed 27 May 1941;granted 25 June 1946) | Chapin, D. M., Fuller, C. S. & Pearson,G. L. A new silicon p-n junction photocell for converting solarradiation into electrical power.J. Appl. Phys. 25, 676677(1954) | Shockley, W. & Queisser, H. J. Detailed balance limit ofefficiency of p-n junction solar cells.J. Appl. Phys. 32, 510519(1961) | Blakers, A. W., Wang, A., Milne, A. M., Zhao, J. & Green,M. A. 22.8% efficient silicon solar cell.Appl. Phys. Lett. 55,13631365 (1989) | Grtzel, M. & ORegan, B. A low-cost,high-efficiency solar cell based on dye-sensitized colloidalTiO

2films. Nature353, 737740 (1991) | Sariciftci, N. S.,

Smilowitz, L., Heeger, A. J. & Wudl, F. Photoinduced electrontransfer from a conducting polymer to buckminsterfullerene.Science258, 14741476 (1992)

The development of the laser (Milestone 9)meant that, for the first time, the interaction ofhuge electric fields with matter could bestudied, particularly in the regime where theelectrical polarization created by the laser is nolonger linearly proportional to the light field.Then, higher-order effects occur, similar to theexcitation of higher harmonics in

musical instruments.Indeed, it was only a year after

the first laser was built when, in1961, Peter Franken andcolleagues used a seminalexperiment to demonstrate thefrequency doubling of lightfrom a ruby laser beamfocused into a quartzcrystal. This secondharmonic signal wasimaged as a smallspot on a photographicplate. Unfortunately,however, that tiny spot wasthought by the lithographers at PhysicalReview Letters to be a grain of dust, andwas therefore eliminated from thepublished version of the article.

Nonetheless, the significance of theseresults was widely recognized, and inspiredNicholaas Bloembergen and his group toenter the field; while waiting for a suitablelaser source to conduct their ownexperiments, they developed thetheoretical foundations of the quantummechanical description of nonlinear optics.

This effort was recognized with a share ofthe Nobel Prize in Physics in 1981.

Subsequently, other nonlinear effectswere demonstrated in the early 1960s suchas sum-frequency generation and four-wavemixing. An important nonlinear effect thatforms the basis for continuously tunable lasersources is optical parametric generation.There, two beams of different energy aregenerated from one incoming laser beam.

Another class of nonlinear optical effectsoccurs through the influence ofstrong light fields on material

MILEStONE 10

opic i army

properties themselves. Since the nineteenthcentury, the effects of electric fields on therefractive index of a material the Kerr andPockels effects have been known.High-intensity optical fields can achieve asimilar effect, which is known as self-phasemodulation. In laser pulses this leads to chirp,which is a variation in the frequency spectrum

of the pulse, and istherefore an importantdetrimental effect to

consider in many opticalsystems.

Ever since those earlydiscoveries in the 1960s,

nonlinear optical effectshave been widely used inapplications. Apart fromtelecommunicationapplications in which

nonlinear effects are anideal tool to manipulate the

short, intense laser pulses inoptoelectronic systems, they also form thebasis of imaging and sensing applicationssuch as coherent anti-Stokes Ramanspectroscopy (CARS) and multiphotonfluorescence microscopes.Second-harmonic generation also plays animportant role in the femtosecondfrequency combs used for ultrahigh-resolu-tion laser spectroscopy (Milestone 20). Asin the case of music, the best works arealways those that make perfect use ofhigher harmonics.

Stefano Tonzani,Associate Editor, Nature Communications

ORIGINAL RESEARCH PAPERS Franken, P. A., Hill, A. E.,Peters, C. W. & Weinreich, G. Generation of opticalharmonics. Phys. Rev. Lett.7, 118119 (1961) | Giordmaine,J. A. Mixing of light beams in crystals. Phys. Rev. Lett.8,1920 (1961) | Kroll, N. M. Parametric amplification inspatially extended media and application to the design oftuneable oscillators at optical frequencies.Phys. Rev.127,12071211 (1962) | Armstrong, J. A., Bloembergen, N.,Ducuing, J. & Pershan, P. S. Interactions between light wavesin a nonlinear dielectric.Phys. Rev.127, 19181939 (1962)FuRtHER REAdING Shen, Y. R. Principles of NonlinearOptics (Wiley, 1984) | Boyd, R. W. Nonlinear Optics(Academic Press, 2008)

GETTY

Reprintedfigurew

ithpermissionfromF

ranken,P.A.

e t a l

( 1 9 6 1 ) A m

e r i c a n P h y s i c a l S o c i e t y

Milestones


http://www.nature.com/milestones/milephotons/full/milephotons07.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons10.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons13.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons15.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons16.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons22.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons09.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons20.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons20.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons09.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons22.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons16.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons15.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons13.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons10.htmlhttp://www.nature.com/milestones/milephotons/full/milephotons07.html


8/16

The revolutionary insight by James Clerk Maxwell

that light is an electromagnetic wave, and theequations he set up to describe it formally(Milestone 2), still serve as the basis for thediscussion and analysis of virtually all the opticalinstrumentation we have ever developed, asRoy Glauber put it in his 2005 Nobel lecture.That overwhelming and continuing success mayeventually have led to a certain complacency.

This Nobel Prize came exactly 100 years afterAlbert Einstein had introduced the concept oflight quanta, following Max Plancks work onblack-body radiation (Milestone 3). Still, by themiddle of the past century, the granular natureof light did not seem to play a significant role in

optics. Even today, when we talk about lasers,holography and photonic bandgaps, these arephenomena that do not demand the quantizationof light. Beginning in the 1960s, however, there

has been a growing appreciation of a distinctnon-classical variety of light, which gave rise toan entire new field of research: quantum optics.

It began with experiments aimed atdetermining the angular diameter of radio stars.Robert Hanbury Brown and Richard Twiss hadmeasured the radio-frequency signals of these

celestial objects using two spatially separatedaerials; by correlating the low-frequency outputsof the detectors, the diameter of the star couldbe estimated. In these experiments, it wasexclusively a correlation in signal intensity thatwas measured, making this kind of interferometerfundamentally different from the Michelsoninterferometer, in which the correlation in signalamplitude is crucial. The Hanbury BrownTwissexperiment is understood when considering theradio signals as classical waves, and the methodproved useful for its intended purpose. Thequestion then was whether the same approachcould be used to measure the diameter ofvisible stars whether the concept could beextended to optical wavelengths was far fromobvious. So, Hanbury Brown and Twiss did alaboratory experiment: they passed light from

MILEStONE 11

Quantum light

Hard on th hls of th strggl to ndrstandwavpartil dality was th onfrontationof somthing jst as izarr and forign: non-

loality. Is all of th information that is rlvantto a physial ojt or intration ontaind atth point in spa and tim whr that ojt

or intration is loatd? Th sam qantmformalism that prodd wavpartil dalityanswrd no. Lik dality, non-loality had

a history strthing ak at last to Sir IsaaNwton, whos thory of gravity implidinstantanos ommniation ovr aritrary

distans and drw asations of mystiism.And, lik dality, lar answrs startd tomrg in th arly twntith ntry.

Qantm mhanis dsrid rality asinhrntly non-loal. To som physiists, this

simply mant that qantm mhanis wasinomplt. Th most famos inompltnssargmnt was dvlopd y Alrt Einstin,Boris Podolsky and Nathan Rosn, and thn

rfind y David Bohm. It pointd ot thatth masrmnt of th spin of two widly-sparatd atoms mst orrlatd if thy

originatd from a moll with a known totalspin. A spin masrmnt along on axis ofon atom mant that th spin along th sam

axis was known for th othr atom.Qantm mhanis, howvr, alsostatd

that th spin of an atom old known along

only a singl axis. Thrfor, th atom that wassond to masrd had an indtrminat

(nknowal) spin along x andy if th firstatom was masrd along z, t it had anindtrminat spin alongy and zif th first

atom was masrd along x. How oldthat sond atom instantanosly know toassm a partilar spin along a partilar

axis, (and an nknowal spin along th oth-rs), nlss it arrid with it all of th rlvantinformation for vry axis? Einstin, Podolsky

and Rosn onldd that it old not, givnthat instantanos ommniation twnth atoms violatd Einstins own thory of

rlativity. As qantm mhanis did notaont for sh loal information (indd,it xpliitly dnid it), it mst originat fromhiddn varials, and posd a srios hal-

lng to th mrging qantm pitr.Einstin, Podolsky and Rosn plishd

thir argmnt in th Physical Review in

1935, and a rply was plishd in th samyar and th sam jornal y that famosopponnt to Einstins point of viw, Nils

Bohr. Howvr, it was not ntil John Blltakld th prolm in 1964 that a lar,qantitativ and tstal opposition twn

hiddn varials and qantm mhanis wasstalishd. His argmnt, and ssqntxprimnts, hav falln strongly, if not di-

sivly, on th sid of qantm mhanis.At th or of Blls tratmnt ar Blls

inqalitis. Ths pla an ppr limit on th

orrlations twn masrmnts of rmot

MILEStONE 12

Dia relai

partils in th as that thos orrlations

ar dtrmind y hiddn loal varials. Bllshowd that ths limits ar rokn y thprditions of standard qantm mhanis.Whras Bll onsidrd masrmnts on

ltrons, th strongst tsts of his inqali-tis y John Clasr and Start Frdman,and latr y Alain Aspt hav sd photons

passd throgh optial polarizrs th dir-tions of whih ar st aftr th photons havlft thir sor. This rstrits th fft of any

hiddn varials in th systm to loal to thtravlling photons. Althogh no airtight tsthas n prformd as yt, Blls thorm and

a mercury arc lamp through a beam splitterand looked at correlations between the signalsof the photomultipliers terminating the twobeams. Sure enough, they saw a clear tendencyof the photodetectors to register photonssimultaneously.

Why should the photons arrive in a correlated

manner? The results stirred up controversy,and it was Glauber, in 1963, who presenteda full framework to explain higher-ordercorrelations in multiple-photon coincidencemeasurements. From his quantum theory ofoptical coherence it followed that there arecases for which the classical description oflight is inadequate; only in such non-classicalintensity correlations are the signatures oflight quantization revealed.

In the years that followed, a number ofresearchers demonstrated strictly quantumbehaviour of light. First came the observation byJohn Clauser, in 1974, of non-classical correlationsbetween two photons emitted in cascade bya three-level atom. In 1977, Jeff Kimble, MarioDagenais and Leonard Mandel demonstratedthat photons emitted by a single sodium atom

John Bell with a sketch of Alain Aspects experimental set-up.

CERN

Milestones




9/16

th xprimnts it has inspird hav shown to

a high dgr of onfidn that natr is, atlast to som xtnt, not loal.

This framwork was latr xtndd to

ntanglmnt of mor than two partils,most importantly y Danil Grnrgr,Mihal Horn and Anton Zilingr, whos

GHZ stat am a rial ingrdint toan ntirly nw fild: qantm informationsin (Milestone 17).

Michael Segal,Associate Editor, Natr Nanothnology

ORIGINAL RESEARCH PAPERS Einstein, A., Podolsky, B.& Rosen, N. Can quantum-mechanical description ofphysical reality be considered complete?Phys. Rev.47,777780 (1935) | Bohr, N. Can quantum-mechanicaldescription of physical reality be considered complete?Phys. Rev.48, 696702 (1935) | Bohr, N. Quantum mechanicsand physical reality. Nature136, 65 (1935) | Bohm, D.Quantum TheoryCh. XXII (Prentice-Hall, 1951) | Bohm, D. &Aharonov, Y. Discussion of experimental proof for theparadox of Einstein, Rosen, and Podolsky. Phys. Rev.108,10701076 (1957) | Bell, J. S. On the Einstein Podolsky Rosen

paradox.Physics1, 195200 (1964) | Clauser, J. F., Horne, M. A.,Shimony, A. & Holt, R. A. Proposed experiment to test localhidden-variable theories.Phys. Rev. Lett.23, 880884 (1969) |Freedman, S. J. & Clauser, J. F. Experimental test of localhidden-variable theories.Phys. Rev. Lett.28, 938941 (1972) |Aspect, A., Grangier, P. & Roger, G. Experimental realizationof EinsteinPodolskyRosenBohm Gedankenexperiment: anew violation of Bell inequalities.Phys. Rev. Lett.49, 9194(1982) | Aspect, A., Dalibard, J. & Roger, G. Experimental testof Bell inequalities using time-varying analyzers. Phys. Rev.Lett.49, 18041807 (1982) | Greenberger, D. M., Horne, M.A. & Zeilinger A. inBells Theorem, Quantum Theory, andConceptions of the Universe (ed. Kafatos, M.) 7376 (KluwerAcademics, 1989)FuRtHER REAdING Wick, D. The Infamous Boundary(Birkhuser, 1995) | Ellis, J. & Amati, D. (eds) QuantumReflections (Cambridge University Press, 2000)

are separated in time, which is a phenomenonthat is known as antibunching. In 1986, PhilippeGrangier, Grard Roger and Alain Aspect usedthe same cascade as Clauser to build the firstsource of single photons, and observed theopposite of the Hanbury BrownTwiss effect:anticorrelations in the detection of a single

photon on the two sides of a beam splitter.So light is more than just a wave. There couldbe no further complacency.

Andreas Trabesinger,Senior Editor, Nature Physics

ORIGINAL RESEARCH PAPERSBrown, R. H. & Twiss, R. Q.Correlation between photons in coherent beams of light.Nature177, 2729 (1956) | Glauber, R. J. The quantumtheory of optical coherence.Phys. Rev.130, 25292539(1963) | Glauber, R. J. Coherent and incoherent states ofthe radiation field. Phys. Rev.131, 27662788 (1963) |Clauser, J. F. Experimental distinction between thequantum and classical field-theoretic predictions for thephotoelectric effect.Phys. Rev. D 9, 853860 (1974) |Kimble, H. J., Dagenais, M. & Mandel, L. Photonantibunching in resonance fluorescence.Phys. Rev. Lett.39,

691695 (1977) | Grangier, P., Roger, G. & Aspect, A.Experimental evidence for a photon anticorrelation effecton a beam splitter: a new light on single-photoninterferences.Europhys. Lett.1, 173179 (1986)

Without doubt, our world of high-volumedata communications would not be possiblewithout the advent of optical fibres. Today, wecan send text, images, speech and video filesinstantly from and to anywhere in the worldso conveniently that we have come to takethis accessibility for granted.

Optical fibres have been a prerequisitefor this extremely rapid development,transporting information over distances ofthousands of kilometres. The operation ofoptical fibres is based on Snells Law, whichstates that light can be totally reflected whenit travels from a medium with a higherrefractive index to one with a lowerrefractive index a phenomenon known astotal internal reflection. Based on thisprinciple, optical fibres are composed of ahigh-refractive-index core surrounded bya low-refractive-index cladding layer.

Although the principle of lighttransmission through optical fibres was

known early on, long-distance lighttransmission was hampered by excessiveoptical losses during transmission. Then, in1966, Charles Kao and George Hockham,working for the English company StandardTelephones and Cables, suggested that theattenuation in fibres was caused by impuritiesin the glass, rather than fundamental physicaleffects such as scattering. They proposedthat, for high-purity silica glass, theattenuation of light could be kept at20 dB km1. At the time, optical fibresexhibited losses of 1,000 dB km1. For this

discovery, Kao was awarded the Nobel Prizein Physics in 2009.

However, even at a loss of 20 dB km1, 99%of the light would be lost over a distance ofonly 1 km, which is impractical for long-haultransmission. Work on purifying glass beganto take place. In 1970, Corning scientistsRobert Maurer, Donald Keck and PeterSchultz successfully fabricated a glass fibrewith an attenuation of just over 16 dB km1,exceeding the 20 dB km1 benchmark. It wasmade of a titanium-doped silica core and apure fused silica cladding. Two years later,using a germanium-doped core, Corningproduced multi-mode glass fibres with a lossof ~4 dB km1. Subsequent developmentsreduced the loss to 0.2 dB km1 at awavelength of 1.55 m.

In 1988, the world witnessed the firsttransatlantic optical fibre between theUnited States and Europe, with a length of6,000 km. To date, >1 billion km of optical

fibres has been laid, capable of carrying>10 Gb s1 of data. Moreover, optical fibresfind applications not only in communicationsbut also in imaging, sensing and medicine.

Rachel Won,Associate Editor, Nature Photonics

ORIGINAL RESEARCH PAPERS Kao, K. C. & Hockham, G. A.Dielectric-fibre surface waveguides for optical frequencies.Proc. IEE113, 11511158 (1966) | Kapron, F. P., Keck, D. B. &Maurer, R. D. Radiation losses in glass optical waveguides.Appl. Phys. Lett . 17, 423425 (1970)FuRtHER REAdING Hecht, J. City of Light. The Story of FiberOptics (Oxford Univ. Press, 1999)

MILEStONE 13

The birth of optical communications

GETT

Milestones




10/16

Revolutionary is used too oftento describe advances in science.When applied to the invention ofthe charge-coupled device (CCD)array by Willard Boyle and GeorgeSmith, however, it is not far off. Yet,the CCD was not originally intended

for applications in digital imaging,for which Boyle and Smith receivedthe 2009 Nobel Prize in Physics, butrather as a potential new form of digitalmemory.

In the late summer of 1969, Boyleand Smith, who were working at BellLaboratories, were told to come upwith a semiconductor memory thatcould compete with the so-calledmagnetic bubble memory that wasbeing developed by a rival group oftheir division. Bubble memory workedby injecting magnetic domains into

garnet patterned with an array offerromagnetic bars. Applying analternating magnetic field caused thesedomains, or bubbles, to hop within thegarnet from underneath one bar tothe next, like packages on a conveyorbelt. By taking the presence or absenceof a bubble to represent a 1 or a 0,

MILEStONE 14

Digital photography is born

Today, th s of lasrs is nothingpartilarly xiting. DVD playrs,

lasr pointrs, ar-od sannrs andtlommniations all s lasrsmad from smiondtor matrials.

Th sitation was diffrnt in 1962,whn only xpnsiv lasrs asd onatomi gass xistd (Milestone 9).

Yt, that yar, Rort Hall at GnralEltri ralizd a first ltriallyopratd solid-stat lasr, asd on

th smiondtor gallim arsnid,followd, within 1 month, y similar

disovris y tams lad y MarshallNathan, Bnjamin Lax and NikHolonyak. Howvr, with high lasrthrsholds and poor lasing ffiin-

is vn at ryogni tmpratrs,prospts for th pratial s of thslasrs appard nrtain.

Th following yar, HrrtKromr, as wll as Zhors Alfrovand Rdy Kazarinov from th Ioff

Physio-Thnial Institt of thRssian Aadmy of Sins,indpndntly am p with an

ingnios sggstion: th onpt ofdol-htrostrtr lasrs.

Instad of sing a lk smi-

ondtor, thy sggstd a layrdstrtr mad of a thin smiond-tor film with a smallr and gap sand-

wihd twn smiondtor layrswith a largr and gap. Th larg gapof th nighoring layrs lads to an

ffiint onfinmnt of arrirs inth ntral layr, whih nhans thprforman of th lasrs.

It ing th tim of th old war,

rsarh grops in th Wst as wll asth East gan a ra to fariat th

first room-tmpratr smiond-tor lasr. This important milstonwas vntally ahivd in 1970, whn

grops first from th Ioff Physio-Thnial Institt and thn fromBll Laoratoris ralizd ontinos

room-tmpratr lasing mad fromgallim arsnid sandwihd twnalminim gallim arsnid.

Ths and onsqntahivmnts wold not hav

n possil withot th

paralll driv towards thinfilm-dposition systms dringth lat 1960s. Of partilar

rlvan wr mtallo-organihmial-vapor dpositionoriginating from th work

of Harold Manasvit at thNorth Amrian AviationCompany, and mollar-am

pitaxy pionrd y AlfrdCho and John Arthr at BllLaoratoris. Dspit sh

advans in fariation, it wasnot ntil 1996 that th first lsmiondtor lasr was ral-

izd in gallim nitrid y ShjiNakamra (Milestone 19).

In addition, mor omplxlasr dsigns hav ompossil. An xampl is vrtial-avity srfa-mitting lasrs.

Howvr, th rowning ahiv-mnt of sh fforts is thqantm-asad lasr dvl-

opd y Fdrio Capasso andollags at Bll Laoratorisin 1994. Qantm-asad

lasrs ar dsignd so that dr-ing th asading of ltronsthrogh svral hndrds

respectively, such a device could beused to store a series of digital bits.

Boyle and Smith spent barely anhour at the blackboard devising anelectronic alternative. Instead ofmagnetic bubbles, they proposed touse electronic charges injected into

metal-oxide-semiconductor (MOS)capacitors grown on silicon. By placingtwo capacitors close to each other andapplying electric voltages they couldinduce the charge to move from one tothe next. In this way, packets of chargecould be passed down a linear arrayof charge-coupled MOS capacitors,mimicking the operation of a bubblearray.

Ironically, although the operationof the device was a success, neither itnor bubble memory ever took off as ameans of storing digital information.

But it did not take long for Boyle andSmith to realize that it might haveother uses. At around the same time,Bell Laboratories was working hard todevelop the Picturephone, which was acrude videoconferencing system. Thecommercial cathode-ray-tube camerasused were notoriously unreliable and

The CCD inventors, Willard Boyle (left) and George Smith (right). Image courtesy ofAlcatel-Lucent/Bell Labs

MILEStONE 15

Laer fr e mae

GETTY

Milestones




11/16

of layrs, mor than on photon is

mittd pr ltron. Bing apalof oprating aross a road sptralrang, qantm-asad lasrs ar a

sfl sor of tnal lasr radia-tion with appliations to sptrosopyand hmial snsing.

Noriaki Horiuchi,Associate Editor, Natr Photonis

ORIGINAL RESEARCH PAPERS Hall, R. N. et al.Coherent light emission from GaAs junctions.Phys. Rev. Lett.9, 366369 (1962) | Nathan, M. I.,Dumke, W. P., Burns, G., Dill , F. H. & Lasher, G.Stimulated emission of radiation from GaAs p-njunctions.Appl. Phys. Lett. 1, 6264 (1962) |Holonyak, N. & Bevacqua, S. F. Coherent (visible)light emission from Ga(As

1xP

x) junctions.Appl.

Phys. Lett.1, 8283 (1962) | Quist, T. M. et al.Semiconductor maser of GaAs.Appl. Phys. Lett.1,9192 (1962) | Alferov, Zh. I. & Kazarinov, R. F.

Semiconductor laser with electric pumping.USSR patent 181737 (application 950840; 30March 1963) | Kroemer, H. A proposed class ofheterojunction injection lasers.Proc. IEEE51,17821783 (1963) | Alferov, Zh. I. et al. Investigationof influence of AlAsGaAs heterostructureparameters on laser threshold current andrealization of continuous emission at roomtemperature. Fiz. Tekh. Poluprov.4, 18261829(1970); Sov. Phys. Semicond.4, 15731575 (1971) |Hayashi, I., Panish, M. B., Foy, P. W. & Sumski , S.Junction lasers which operate continuously atroom temperature.Appl. Phys. Lett.17, 109111(1970) | Faist, J. et al. Quantum cascade laser.Science264, 553556 (1994) | Nakamura, S. et al.InGaN-based multi-quantum-well-structure laserdiodes.Jpn J. Appl. Phys. 35, L74L76 (1996)

MILEStONE 16

Absolutelyaccurate

a more dependable alternative waseagerly sought. The CCD provided thesolution.

The simplicity of fabricating largesensor arrays, combined with the linearoptical response to even the mostfaint light sources, has meant that 40

years after their invention they are stillused in large-scale optical telescopes,including the Hubble space telescope.It has also allowed them to becomecheap enough to be integrated intomost modern mobile phones a factthat news agencies increasingly relyon for important events. So, althoughthe revolution might not be televised,thanks to the CCD it will almostcertainly be photographed.

Ed Gerstner,Senior Editor, Nature Physics

ORIGINAL RESEARCH PAPERS Boyle W. S. &

Smith, G. E. Charge coupled semiconductordevices.Bell Syst. Tech. J. 49, 587592 (1970) |Amelio, G. F., Tompsett, M. F. & Smith, G. E.Experimental verification of the chargecoupled device concept.Bell Syst. Tech. J. 49,593600 (1970)FuRtHER REAdING Smith, G. E. Theinvention and early history of the CCD. Nucl.Instr. Meth. Phys. Res. A607, 16 (2009)

The ability to measure optical frequencies withhigh precision and stability has led to a plethoraof applications, including optical atomic clocks,optical metrology, high-resolution spectroscopy,and even the global positioning systems usedin mobile telephones and navigation systemsfor cars.

Traditionally, precision measurements havebeen made by comparing the beat frequencybetween two optical frequencies with amicrowave reference, which is a standard basedon a specific transition between hyperfine levelsof the caesium-133 atom. However, the situationchanged when light pulses became available withdurations on the scale of femtoseconds. Earlyapproaches to generating such ultrashort pulseswere plagued by intrinsic instabilities anduncertainty about the underlying mechanisms. Aremedy came, in 1981, when Charles Shank andco-workers at Bell Laboratories invented thecolliding-pulse mode-locked (CPM) laser, whichgenerated the first coherent photon wave packetsin the sub-100-fs regime.

Crucially, the introduction of titanium-dopedsapphire (Ti:sapphire) as a broadband gainmedium in the near-infrared spectral regionrevolutionized the generation and amplificationof ultrashort pulses. The first broad-bandwidthsolid-state laser was demonstrated by PeterMoulton in 1986, and, together with thesubsequent demonstration of self-mode locking inTi:sapphire lasers by Wilson Sibbett andco-workers in 1991, this paved the way tofemtosecond pulses with high peak powers and

good tunability. Sibbetts group produced pulseswith durations as short as 2.0 ps and, using anintracavity dispersion compensation in amode-locked Ti:sapphire laser, they managed toachieve pulse durations as short as 60 fs and peakpowers of 90 kW. In 1985, Grard Mourou andco-workers introduced a chirped-pulseamplification scheme that allowed them to pushthe intensities of femtosecond lasers to>1021 W cm2. In the 1990s, dispersion control wasdramatically simplified through the use of chirpedmultilayer mirrors, which extended the oscillatorsperformance into the few-cycle frontier.

The development of reliable high-intensity,sub-100-fs laser technology based on thesebreakthroughs has stimulated an explosion ofactivity, leading to fundamental studies into theways photons and matter interact on very shorttimescales. Femtosecond lasers have been used asaccurate stopwatches to observe in real time the

energy transfer and storage process, which is atthe heart of many chemical processes, resulting inthe 1999 Nobel Prize for Chemistry beingawarded to Ahmed Zewail. More recently, thebroadband coherence of femtosecond pulses hasbeen harnessed in the invention of thefemtosecond frequency comb, which is an opticalmeasurement technique that can preciselymeasure different colours or frequencies of light.

John Hall and Theodor Hnsch shared half of theaward for the 2005 Nobel Prize in Physics fortheir contributions to the development oflaser-based precision spectroscopy, including theoptical frequency comb technique.

Their ease of fabrication and simplicity,compared with techniques based on amicrowave standard, have helped to establishfrequency combs as excellent frequencyreference sources and measurement tools. Theyare nowadays commercially available and widelyused for metrological purposes. There should bemore to come: optical atomic clocks usingfrequency combs are expected to haveaccuracies 100 times better than any othertime-keeping systems, making them attractivefor use in global satellite-navigation systems.

Rachel Won,

Associate Editor, Nature PhotonicsORIGINAL RESEARCH PAPERS Fork, R. L., Greene, B. I. & Shank,C. V. Generation of optical pulses shorter than 0.1 psec by collidingpulse mode-locking.Appl. Phys. Lett.38, 671672 (1981) | Strickland,D. & Mourou, G. Compression of amplified chirped optical pulses.Opt. Commun.56, 219221 (1985) | Moulton, P. F. Spectroscopic andlaser characteristics of Ti:Al

2O

3.J. Opt. Soc. Am.B3, 125133 (1986) |

Dantus, M., Rosker, M. J. & Zewail, A. H. Real-time femtosecondprobing of transition states in chemical reactions.J. Chem. Phys.87,23952397 (1987) | Spence, D. E., Kean, P. N. & Sibbett, W. 60-fsecpulse generation from a self-mode-locked Ti:sapphire laser.Opt. Lett.16, 4244 (1991) | Szipocs, R., Ferencz, K., Spielmann, C. & Krausz, F.Chirped multilayer mirrors for broadband dispersion control infemtosecond lasers.Opt. Lett.19, 201203 (1994) | Reichert, J.,Holzwarth, R., Udem, Th. & Hnsch, T. W. Measuring the frequency oflight with mode-locked lasers.Opt. Commun.172, 5968 (1999)

CourtesyofTedHnschandMax-Planck

InstitutfrQuantenop

tik

Milestones




12/16

Th way w pross and ommniat infor-mation has hangd rapidly ovr th past fw

dads, t might ndrgo a frsh rvoltionwith th advnt of qantm information

thnology. Th fild gan in th arly 1980s,whn lasr xprimnts tsting th qantmnatr of light matrd (Milestone 12), andth trnd shiftd to trying to mploy qantm

stats of light for noding information. Inindpndnt dvlopmnts, Rihard Fynmanand David Dtsh introdd th onpt

of a qantm omptr, and Charls Bnnttand Gills Brassard proposd a protool forqantm ryptography. Bnntt and Brassard

wr inspird y idas that wr formlatdy Stphn Wisnr in 1970, and plishdformally in 1983, aot sing qantm-

mhanial ffts to prodanknots that annot ontrfitd.

Ths idas ar asd on th qantm

prinipl of sprposition, whih allows

partils to xist in mltipl stats at on.For xampl, whn a singl photon falls onto

a half-silvrd mirror, thr ar two pos-sil otoms: rfltion and transmission.Instad of going on partilar way, howvr,

th photon is transmittd and rfltd at thsam tim. Only whn dttors ar pt intopla will on of th two paths attritd

to th photon. Withot dttion, th photonfftivly gos oth ways.

A qantm omptr oprats in sh

sprposition mods. Th asi nit is aqantm it (or qit) that, ntil it is rad ot,

In classical physics, whenever a wave encounters achange in density a part of that wave is reflected.In 1887, Lord Rayleigh took this concept furtherby studying what happens if the wave propagatesnot through a homogeneous medium but throughone with a periodic structure. He showed that raysreflected from the multiple interfaces interferewith one another. For a band of wavelengths ofsimilar value to the periodicity of the stack, theinterference is destructive so that this bandgapprevents wave propagation through the structure.

By the 1980s, localization of light in artificial

structures was a hot topic. Combininglocalization with the idea of the Rayleighbandgap, Sajeev John considered, in 1987, howelectromagnetic radiation could be trapped in aperiodic three-dimensional dielectric materialif randomness is introduced. As an illustration ofthis, consider altering the periodicity at just onepoint allows the existence of light at a wavelengthwithin the bandgap; however, this light istrapped in the vicinity of the defect because itis forbidden everywhere else. Applied to chainsof imperfections, light can then be guided withlittle loss. The potential of this approach cannot

be understated: just as semiconductors havemade possible the miniaturization of electricaldevices, so photonic crystals hold the promise ofmicroscale photonic circuitry.

Another landmark was set by Eli Yablonovitchwith his paper published earlier in 1987. Followingthe work of Edward Purcell, scientists had startedto think about controlling spontaneous lightemission by modifying the photonic environment.This is exactly what a photonic crystal does. Aquantum light source surrounded by a photonicbandgap is prevented from decaying because

the photon that it needs to emit cannot exist.Conversely, the spontaneous-emission rate canbe increased if the emitter is placed inside adefect with which it is in resonance.

The next challenge was fabrication. The firstproposed design with a full bandgap compriseddielectric spheres in a configuration similarto atoms in a diamond crystal. However,the eventual structure, which was initiallydemonstrated in 1991 in the microwave regime,used an approach that was better suited to thematerial-processing abilities at the time bydrilling holes in three different directions.

an tak on vals of oth 0 and 1 simltan-

osly. A qit an, for xampl, an atomin its grond or xitd stat, or an ltronwith spin p or down. If w an opl

togthr 10 qits, thy an olltivly inall 1,024 lassial stats of 10 its. Th powrof qantm ompting rlis on having allsprposition stats at disposition in paralll

dring omptation.Howvr, qantm ompting is not

simply a fastr way of information prossing;

nw algorithms ar rqird, tailord to itsstrngths. Intrst in th onpt thrforros sharply in 1994 whn Ptr Shor fond

a task to whih qantm omptrs ar par-tilarly wll sitd: fatoring larg nmrs.using Shors algorithm, qantm ompt-

rs old, in prinipl, fator a 1,000-digitnmr in a fration of a sond a prolmthat lassial omptrs annot solv within

th liftim of th univrs.

Th rot to pratial appliations hasalways n lar for qantm ryptography.

In th protool of Bnntt and Brassard,two partis stalish a ryptographi kyy xhanging polarizd photons that ar

prpard in sprposition stats. Th ssnof th shm is that thy an safly xhangphotons as avsdropping is immdi-

atly dttd. Any attmpt to intrpt pho-tons rats rrors that lgitimat partis anidntify y omparing part of th xhangd

data; th rmaining data ar sd to ild asrt ky.

MILEStONE 17

From paradox to technology

MILEStONE 18

Sparkling traps

Image design by Stephen Eisenmann (University of Illinois atUrbana-Champaign) and Tom Wilson.

GETTY

Milestones




13/16

Of particular relevance to practical applicationsare two-dimensional photonic-crystal designs,which were first realized in 1996. They representa compromise between a full bandgap and asimpler fabrication that makes possible theintegration of both passive and active opticalcomponents on a single photonic chip. Withthe possibility of a revolution on a par withthe development of the semiconductor chip,photonic-crystal research looks set to shine.

David Gevaux, Senior Editor, Nature Physics

ORIGINAL RESEARCH PAPERS Rayleigh. On the

maintenance of vibrations by forces of double frequency, andon the propagation of waves through a medium endowedwith a periodic structure. Phil. Mag. Series 524,145159(1887) | Yablonovitch, E. Inhibited spontaneous emission insolid-state physics and electronics. Phys. Rev. Lett.58,20592062 (1987) | John, S. Strong localization of photons incertain disordered dielectric superlattices.Phys. Rev. Lett.58,24862489 (1987) | Yablonovitch, E., Gmitter, T. J. & Leung,K. M. Photonic band structure: the face-centered-cubiccase employing nonspherical atoms. Phys. Rev. Lett.67,229522298 (1991) | Krauss, T. F., De La Rue, R. M. & Brand, S.Two-dimensional photonic-bandgap structures operatingat near infrared wavelengths.Nature383, 699702 (1996)FuRtHER REAdINGJoannopoulos, J. D., Johnson, S. G.,Winn, J. N. & Meade, R. D. Photonic Crystals: Molding theFlow of Light 2nd edn (Princeton Univ. Press, USA, 2008)

Mor proposals for qantm omm-

niation shms and thir xprimntalvrifiation nsd, notaly thos mployingqantm ntanglmnt and dmonstrating

tlportation. Systms for qantm kydistrition ar now availal ommrially.Bilding a qantm omptr, howvr,rmains a formidal task and as yt only a

fw qits hav n opld to implmntsmall algorithms. In rnt yars, variospromising nw shms hav n xplord,

sing photons as wll as matrial systms shas ions, atoms, molls, qantm dots andsprondting irits for prossing qan-

tm information. Photons ar also mploydas flying qits for transmitting qantminformation, thry nsring a arrying rol

of photons in this mrging thnology.

Liesbeth Venema,Senior Editor, Natr

ORIGINAL RESEARCH PAPERS Feynman, R. Simulatingphysics with computers. Int. J. Theoret. Phys.21, 467488(1982) | Wiesner, S. Conjugate coding.ACM Sigact News15(1), 7888 (1983) | Bennett, C. H. & Brassard, G. in Proc. IEEEInt. Conf. Comp. Syst. Signal Process. 175179 (IEEE, 1984) |Deutsch, D. Quantum theory, the ChurchTuring principleand the universal quantum computer.Proc. R. Soc. Lond. A400, 97117 (1985) | Ekert, A. K. Quantum cryptographybased on Bells theorem.Phys. Rev. Lett.67, 661663 (1991) |Bennett C. H. et al. Teleporting an unknown quantum statevia dual classical and EinsteinPodolskyRosen channels.Phys. Rev. Lett.70, 18951899 (1993) | Shor, P. W. in Proc. 35thAnnu. Symp. Found. Comp. Sci. (ed. Goldwasser, S.) 124134(IEEEComputer Society Press, 1994)FuRtHER REAdING Gisin, N. & Thew, R. Quantumcommunication. Nature Photon.1, 165171 (2007) | Ladd, T. D.et al. Quantum computers. Nature454, 4553 (2010)

Light-emitting diodes (LEDs) are a ubiquitouspart of modern life. Their popularity is evidentfrom their deployment everywhere from carbrake lights and giant display boards to trafficlights and indicator lamps on electronicgoods. Many also predict that LEDs are poisedto play an increasingly important role ininterior lighting thanks to their long lifespanand low power consumption.

The origins of the LED can be traced backto the initial research on electroluminescencefrom semiconductors. In 1907, Henry Roundreported a bright glow from a crystal of siliconcarbide. This was followed in the 1920s byintensive research by the Russian scientistOleg Losev, who studied zinc oxide and siliconcarbide, observing a threshold behaviour ofthe light emission and documenting thespectrum of the light emitted.

However, much credit for the invention ofa practical LED is widely attributed toscientists in the United States in the early1960s. In 1961, scientists at Texas Instrumentsreported that gallium arsenide (GaAs) emittedinfrared light when pumped by an electricalcurrent. The following year saw abreakthrough in LED research with various

papers on GaAs-based red and infrared lightemission, including the report on lasing(Milestone 15). Thanks to his pioneering workon red GaAs LEDs, Nick Holonyak is oftenreported as being the father of the LED.

Although the first versions were dim, LEDsthat were much brighter quickly followed, asdid yellow emitters. However, for many yearsscientists struggled to find a suitable materialsystem for emitting bright blue light. This allchanged in the 1980s with research ongallium nitride (GaN) and the development byShuji Nakamura, a scientist at Nichia, of an

efficient scheme for positive-type doping(p-doping) of GaN LEDs. His research openedthe door to the first commercial high-powerblue LEDs in 1993, completing the colourrange of LEDs across the visible spectrum.It also led to several important spin-offsincluding the white LED (a blue LED chipcoated with a light-converting phosphor).

In many ways, LEDs can be considered asthe first great success of optoelectronics, andimprovements in performance have beencharted by a law akin to Moores law inmicroelectronics. Haitzs law documents thatevery 10 years the amount of light generatedby an LED increases by a factor of 20, whereasthe cost per unit of useful light emitted fallsby a factor of 10. Today, LED research isflourishing around the world, with scientistsattempting to optimize the colour andbrightness of white light, push emission deepinto the ultraviolet, and explore new efficientmaterial systems based on organicsemiconductors as well as quantum dots.

Oliver Graydon,Chief Editor, Nature Photonics

ORIGINAL RESEARCH PAPERS Round, H. J. A note oncarborundum. Electrical World19, 309310 (1907) | Lose v, O.

V. Luminous carborundum [silicon carbide] detector anddetection with crystals. Telegrafiya i Telefoniya bez Provodov(Wireless Telegraphy and Telephony) 44, 485494 (1927) |Nathan, M. I., Dumke, W. P., Burns, G., Dil l, F. H. & Lasher, G.Stimulated emission of radiation from GaAs p-n junctions.Appl. Phys. Lett.1, 6264 (1962) | Holonyak, N. & Bevacqua, S. F.Coherent (visible) light emission from Ga(As

1xP

x) junctions.

Appl. Phys. Lett.1, 8283 (1962) | Quist , T. M. et al.Semiconductor maser of GaAs. Appl. Phys. Lett.1, 9192(1962) | Burroughs, J. H. et al. Light-emitting diodes based onconjugated polymers. Nature347, 539541 (1990) | Nakamura,S., Mukai, T. & Senoh, M. Candela-class high-brightnessInGaN/AlGaN double-heterostructure blue-light-emittingdiodes.Appl. Phys. Lett.64, 16871689 (1994) | Colvin, V. L.,Schlamp, M. C. & Alivisatos, A. P. Light-emitting diodes madefrom cadmium selenide nanocrystals and a semiconductingpolymer.Nature370, 354357 (1994)

MILEStONE 19

Brig ew wrld

GETTY

Milestones




14/16

The materials parameters of any optical apparatus,such as a lens or a prism, determine how theproperties of light can b e controlled. Unfortunately,the range of options available in naturalmaterials is surprisingly limited. Lightmatterinteractions are described by Maxwells equations(Milestone 2), and the values of the parametersthat enter those equations dielectricpermittivity and magnetic permeability meanthat the refractive index of a natural material isalways positive, even though the equations allowthe possibility of a negative refractive index.

The investigation of negative refractive index,

for which permittivity and permeability are bothnegative, began years ago. Notably, in 1968,Victor Veselago worked out that a planar slab ofa material with a negative refractive index wouldfocus light in the same way as would curved lensesmade from conventional materials. In the absenceof any material that had such properties, however,these early studies largely fell into oblivion.

All this changed in 1999, when John Pendry andcolleagues demonstrated a swiss roll structurethat had negative permeability. The key was totailor the structure of the material on a scalesmaller than the wavelength of light passing

through it, so that the optical waves do not resolvethe underlying features. Such devices are nowknown as metamaterials. Importantly, as negativepermittivity occurs in metals close to theirplasmon resonance, the realization of a negativerefractive index seemed within reach.

Metamaterials then received widespreadattention in 2000, when Pendry published hislandmark paper on a perfect lens, which was madeusing a negative-refractive-index material. Thiswork staggered many in the field for its seeminglyunbelievable prediction of perfect imagingcapability, and was the first concrete step towards

metamaterials with functionality that is impossiblefor natural materials. The following year, thepredictions made by Pendry were confirmed whenthe first negative-refractive-index metamaterialwas fabricated by Richard Shelby, Sheldon Schultzand David Smith.

The capabilities of metamaterials are notlimited to homogeneous structures. In 2006, UlfLeonhardt, and independently Pendry, Schurigand Smith, realized that metamaterials withspatially varying properties could be powerfultools for guiding light along almost arbitrary paths.Leonhardt used the ray approximation to make his

mapping. Pendry, Schurig and Smith developeda theoretical description that was based on acoordinate transformation of Maxwells equations,and was hence known as transformation optics.Consequently, transformation optics has broken

down many barriers in the field of optics.In 2006, one of the most widely publicized

metamaterial devices was revealed theoptical cloak, using which, for a specific set ofparameters, an object can be concealed from anobserver. Leonhardt later extended the transfor-mation-optics approach to curved, non-Euclideancoordinate transformations, which allowoperation across a broader range of wavelengths.Equipped with the possibility to manipulatethe optical properties of matter at will, furtherexciting designs will undoubtedly emerge.

Joerg Heber, Senior Editor, Nature Materials

ORIGINAL RESEARCH PAPERS

Veselago, V. G. Theelectrodynamics of substances with simultaneously negativevalues of and. Sov. Phys. USPEKHI10, 509514 (1968) |Pendry, J. B., Holden, A. J., Robbins, D. J. & Stewart, W. J.Magnetism from conductors and enhanced nonlinearphenomena. IEEE Trans. Microw. Theory Tech. 47, 20752084(1999) | Pendry, J. B. Negative refraction makes a perfect lens.Phys. Rev. Lett.85, 39663969 (2000) | Shelby, R. A., Smith, D.R. & Schultz, S. Experimental verification of a negative indexof refraction. Science292, 7779 (2001) | Leonhardt, U.Optical conformal mapping. Science312, 17771780 (2006) |Pendry, J. B., Schurig D. & Smith, D. R. Controllingelectromagnetic fields.Science312, 17801782 (2006) |Schurig, D. et al. Metamaterial electromagnetic cloak atmicrowave frequencies.Science314, 977980 (2006) |Leonhardt, U. & Tyc, T. Broadband invisibility by non-Euclidean cloaking. Science323, 110112 (2009)

Th diffration limit of lassial optis dosnot allow th loalization of light into rgionsthat ar mh smallr than its wavlngth.

As a rslt, th lvl of intgration andminiatrization of photoni irits is not

vn los to that ahival in ltronis.Th thnology that might los this gap isknown as plasmonis. Plasmoni strtrshav atn th diffration limit, and ld to

advans in sptrosopy and snsing, imag-ing, anr thrapy, intgratd nano-optisand solar lls, to nam jst a fw.

Modrn plasmonis startd with apliation in 1998 y Thomas Esn and

ollags, who osrvd a srprisingly ffi-int light transmission throgh a thin mtalfilm with hols tn tims smallr than th

wavlngth of light. Additionally, mor lightwas transmittd throgh th film than wasinidnt onto th ara of th hols. Evntally,

Esn was al to xplain his osrvationswith th proprtis of srfa-plasmonpolaritons (SPPs).

SPPs onsist of photons that intratwith th srfa motions of fr ltrons inmtals. Plasmoni ffts hav inadvrtntly

n xploitd y glass makrs sin at lastth forth ntry, for xampl to gnratth olors sd in staind-glass windows in

mdival athdrals.Th sintifi invstigation of plasmoni

ffts gan as arly as 1899 with thortial

stdis y Arnold Sommrfld and xprimn-

tal osrvations of plasmoni ffts in lightsptra y Rort Wood in 1902. Latr that

dad, J. C. Maxwll Garntt and Gstav Midvlopd thoris xplaining th sattringffts y mtalli nanopartils. Howvr, itwas not ntil a nmr of thortial stdis in

th 1950s that a mor omplt ndrstanding

of SPPs was rahd. Th fondation for th

systmati xprimntal stdy was thn laidin 1968 y Erih Krtshmann and AndrasOtto, who dvisd mthods to xit SPPs with

prisms attahd to mtal srfas.In th lat 1970s, th thnologial xploita-

tion of plasmons gan with th pionringdisovry y Martin Flishmann and Rihard

Van Dyn of signifiant nhanmnts inth Raman sattring of light y mollsattahd to a rogh silvr srfa. This fft is

xplord for dvis that dtt mollar on-ntrations down to th singl-moll lvl.

For appliations in photoni irits, th

Esn disovry has ld to fforts aimdat xploiting th highly loalizd natr ofplasmons to gid light on th nanosal.

Frthrmor, a plasmon-asd analog toth lasr, th spasr, old provid a sor ofohrnt light low th diffration limit.

In addition, SPPs ar xploitd in a

nmr of aras, sh as mtamatrials(Milestone 21). Similarly, plasmoni lightonntration an not only nhan lightasorption in solar lls, t also improv th

ffiiny of light-mitting dvis.David Pile,

Associate Editor, Natr Photonis

IMAGESOURCE

MILEStONE 21

The masters of light

MILEStONE 20

Small and beautiful




15/16

3

Electric

field

1.0

0.5

0.0

0.5

1.0

0

Time (femtoseconds, 1015 s)3

The new millennium has heralded the arrival ofattosecond light pulses, and with it theemergence of a radical new technology that ismoving time-resolved spectroscopy and controltechniques from the molecular (femtosecond) tothe electronic (attosecond) timescale.

In fact, attosecond light pulses were created inthe early 1990s, when physicists ionized rare-gasatoms with intense laser pulses to generateenergetic radiation alongside the original opticalpulse. Theory exploring such high-harmonicgeneration, from Kenneth Kulander andco-workers and from Paul Corkum, resulted in1993 in a simple model for the process: duringeach half-cycle, the oscillating electric field of anintense laser pulse will tear electrons from atoms

in a gas, accelerate them away and then drivethem back to re-collide with their parent ion. Ineach collision, a short burst of extreme ultraviolet(XUV) photons is created.

Theoretical and experimental groundwork notably by Anne LHuillier and colleagues showed that driving high-harmonic generationwith a multi-cycle femtosecond laser shouldproduce attosecond light pulses, which arerepeated at twice the laser frequency. Rigorousproof of attosecond pulse trains arrived only in2001, however, when Pierre Agostini andcolleagues encoded the properties of the pulsesin photo-ionized electrons and then measured thecharacteristics of these so-called photoelectronreplicas.

A few months later, Ferenc Krausz andcolleagues reported the first individual attosecondpulses, filtered out of pulse trains. The team thenperfected the art of steering re-collision electrons,using the electric fields of intense few-cycle laserpulses, with their waveform judiciously adjusted(Milestone 16) so that each pulse generates onlyone reproducible re-collision event and, hence,one reproducible isolated attosecond pulse.Atomic Auger decay and the photo-ionization ofatoms and solids have all been triggered by such

isolated attosecond photon pulses, and theensuing electron dynamics has been probed by thesynchronized oscillating electric field of the laserpulse that generated the attosecond trigger.

The ionization process at the heart ofhigh-harmonic generation itself launche

Documents

Nature Milestones Photons