FEATURES

Didthe great experimentalistdoctor hisown findings to suit hisbeliefs? Perhapsnot,asacloselook at hisbeliefsreveals.

Millikan's struggle with theory

GeraldHolton

n 1916 Millikan published his experi­ by any other experimenter. rectly at the ordinates, he decided"simply I mental determination of the value of Several factors were responsible for the to cut the feet off"where those curves were Planck's constant, with remarkable preci­ success of the experiment, performed at trailing on too far. sion. It was later cited as part ofhis Nobel the Ryerson Laboratory of the University In short, a triumphant work by a su­ Prize award. But as a study of that paper ofChicago. First ofall, Millikan's use of an perbly confident scientist, of highest im­ together with Millikan's subsequent com­ ingenious device he termed "a machine portance in its day, and richly deserving to ments over the years reveal, he initiallyre­ shop in vacuo;' assembled thanks to "the be cited as part of his Nobel Prize award fused to accept the interpretation of the skill and experience of the mechanician, in 1923, "for his work on the elementary theoretical meaning of his work, as seen Mr. Julius Pearson" .Essentially, the device charge of electricityand the photoelectric from the perspective of a present-day allowed a rotating sharp knife, controlled effect". If historically minded scholars physicist.Yet,over time, Millikan adjusted from outside the evacuated glass contain­ look through that 1916 volume of the his view retroactively to accept the mean­ er by electromagnetic means, to clean off Physical Review inwhich Millikan's paper ing ofwhat he had done - namely to pro­ the surface of the metal used (sodium, appeared, they will notice that physics in vide crucial support for Einstein's heuris­ potassium, lithium) before exposing it to America at that time was still a mixed bag. tic point of view on the quantum theory the beam of the fairly monochromatic One finds in the volume a number ofcon­ of light (1905). This case also serves as a light coming from a quartz-mercurylamp. tributions by scientists such as A.H. test whether scientists, as is now often al­ Second, the frequencies of light Millikan Compton, P.W. Bridgman, Irving Lang­ leged, as a rule arrange to obtain their re­ used covered a much larger span,between muir, E.C. Kemble, among others (with sults in the light oftheir own beliefs. 2399 A, and 5461 A, than had been A.A. Michelson listed on the Board ofEd­ achieved previously, permitting the rela­ itors). Their papers show thatthe mainat­ Robert A. Millikan's historic paper of 1916 tion of the kinetic energy of the electron tention was the experimental part of sci­ in Physical Review on the measurement of and the frequency of light to be deter­ ence, in which Americans were regarded Planck's constant (A Direct Photoelectric mined with a "maximum uncertainty" of as mostinterested and competent. But the Determination ofPlanck's h; Physical Re­ 0.5%. In addition, the kinetic energies of volume as a whole, published in monthly view [1]) lends itself to two different yet the photoelectrons were found by mea­ parts of some 200 pages each on average, complementary readings-first through suring the potential energyVe ofthe elec­ indicates that a good deal ofthe work go­ the eyes of a physicist of today, second by tric field needed to stop the electrons­ ing on in physics in the US in the early an historian ofscience, reconstructing the andthere Millikan was inthe excellent po­ years of this century was still narrow and dramatic meaning the paper had in its sition of being able to confidently use the unambitious, even tending to descend to own time. value for the charge e of the electron he lengthy descriptions of improvements in To a physicist Millikan's paper is a pa­ had published in 1913 with unprecedent­ power supplies, pumps, galvanometers, tient and eloquent essay on how he came ed accuracy, obtained with his oil drop ex­ and the like. to the conclusion that has been in our periment. Next, on looking at the early pages in physics textbooks ever since: While it had Last but not least, shining through it all that same volume, and the segment pub­ been known for a long time that light were Millikan's typical characteristics as lishedtwo months earlier mJanuary 1916, falling on metal surfaces may eject elec­ experimenter and person: His penchant we come upon another article byMillikan, trons from them (the photoelectric effect), for experimenting in an areainvolving the on the same subject,inwhich the veryfirst Millikan was the first to determine with hottest question of the day - in this case sentence announces Millikan's opinion great accuracy that the maximum kinetic the value ofa fundamental constant ofna­ that "Einstein's photoelectric equation... energy of the ejected electrons obeys an ture (as he had done in his 1913 paper) cannotin myjudgmentbelooked upon at equation equivalent to the one Einstein and, as we would now put it, the quantum present as resting upon any sort ofa satis­ had proposed in 1905, namely theory oflight; his energetic persistence ­ factory theoretical foundation;' even 1/2mv2= hv - P, where h is Planck's con­ this paperwas the culmination ofwork he though "it actually represents very accu­ stant,v the frequency ofthe incidentlight, had begun in 1905; his passion for obtain­ rately the behavior" ofphotoelectricity (p. and P "the work necessary to get the elec­ ing results "with verygreat precision" (Fig. 18). tron out of the metal" in Millikan's words. 1) - even atthe cost ofsome practices that Indeed, knowing this, and rereading Moreover, Millikan determined h to have are perhaps questionable from our current Millikan's subsequent (March 1916) paper the value 6.57 x 10-27erg-sec, "with a pre­ perspective, for instance, his frank re­ on Planck's constant, we see more clearly cision of about 0.5 per cent"-a value far mark that to get the curves derived from that he is emphatically distancing himself better than had been obtained previously his experimental results to intersect cor- throughout from Einstein's attempt in

12 Article available at http://www.europhysicsnews.org or http://dx.doi.org/10.1051/epn:2000303europhysics news MAY/JUNE 2000 FEATURES

1905 of what Millikan calls a "coupling of he so characterized any of his work, using forms of quantum theory then being dis­ photo effects ... with any form ofquantum the phrase because his was still only a cussed, siding with the "least concentrat­ theory" (p. 355). What we now refer to as "heuristic" proposal, devoid of sufficient ed:' namely that of Planck, as against"the the photonwas, in Millikan's view, a"bold, theoretical grounding. most concentrated;'that ofEinstein,which not to say reckless, hypothesis" - reckless As it happened,Millikan was in Europe he called "radical" because it could not be both because it was contrary to such clas­ in 1912 for six months, primarily to meet reconciled with "the facts of diffraction sical concepts as light being a wave propa­ the foremost physicists, and particularly and interference, so completely in harmo­ gation phenomenon, and because of the to attend in Berlin Planck's lectures on his ny in every particular with the old theory "facts ofinterference" (p. 355). In the back­ theory of heat radiation. He also took the ofetherwaves:' Indeed,he confessed a cor­ ground we glimpse the presence of opportunityin June 1912 to lecture on the puscular theoryoflightwas for him"quite Michelson, the (~tistof Light;' who was results of his oil drop experiment before unthinkable". In short, Millikan's classic Millikan's admired patron and colleague the Deutsche Physikalische Gesellschaft. 1916 paper was purely intended to be the at the Ryerson Laboratory, the 1907 No­ Planck was delighted with this relative verification of Einstein's equation for the belist, famous for his interferometers, the newcomer who was bringing the best photoelectric effect and the determination work carried out with their aid - and for available value for the charge of the elec­ ofh, without accepting anyofthe"radical" his adherence to ether physics until his tron at the time, which fitted well with implications which to us nowseem so nat­ death in 1931. Planck's own deduction of e from his for­ ura!. So Millikan's paper is not at all, as we mula for radiation. Two of the most im­ Recently opened records of the Nobel might now naturally consider it to be, an portant universal constants, hand e, were Prize Foundation showthat from 1916 on, experimental proof ofthe quantum theo­ finding common ground through this per­ Millikan was nominatedconstantly.When ry oflight. Although Millikan admits that sonal encounter. the award at last came to be in 1923, his the facts seem "to demand some modifi­ On returning to the United States, Mil­ Nobel address of May 23,1924, contained cation of classical theory:' (p. 355) that is likan gave a lecture in December 1912 at passages that showed his continuing notwhat he is willing to attempt here. And the Cleveland joint meeting ofthe Ameri­ struggle with the meaning of his own in truth, Einstein himself, from the begin­ can Association for the Advancement of achievement: "After ten years of testing ning, was not comfortable with the quan­ Science and the American Physical Soci­ and changing and learning and sometimes tum theory oflight, calling it, in a letter of ety, in which he clearly regarded himselfas blundering ... this work resulted, contrary May 1905 to his friend Conrad Habicht, the proper presenter of Planck's theory of to my own expectation, in the first direct "very revolutionary:' It was the only time radiation. Millikan compared the various experimental proof ...ofthe exact validity,

°':F~';":;};~~"~'···~·::::~:~;::~':::I~,~i;:"'"~i

=¥.Vo 43.9 X 1013 , :;; .. f ...... ij . . ,[jj:.:". ..._:-.

•• ! I .

:... ! :...•. Fr' :'if .....; , Ih:d::OIIS"}\.lo.;D.:_~ ~ ; . L 1 ·····-,······r' i "-'-' dPD V' h _ .• ,_..

...... mt.X"-;:.cI1:~~ .. t··. ~·il~~·I~ ~'I·~'''~·I:'111'I''~.~"~.~:1~1'~i'~~~~~::~I~:lm~Will~""t'-~:'I"'l"ii:i~···~,_¥!~:·,~~~;/~v~(l~21;'O~Od~~~'2:5)~X13~o~:ioiJ(~:'12~4x;1;eO_l~5 -2 .,'.'," ,""".,,'. '.. ,," h':i;':?~O",'" 6,,."O":1!lI ~ 40xl013 50 60 70 80 90 100 110 120 Frequency

Fig 1 Taken from p. 373 of Millikan's 1916 article, Ref. 1, it is the key figure in that paper. It shows how he calculated h from the slope of the graph plotting the observed stopping potentials versus the light frequencies. As in a similar crucial figure in his 1913 paper on finding the value of e, one notices how small the deviations of the readings are from the straight line - and how casual Millikan was about giving the units of calculated values (then still a practice among many physicists).

europhysics news MAY/JUNE2000 13 FEATURES within narrow limits of experimental er­ case atleast, the carefully established facts Physicsiseverywhere,but ror, of the Einstein equation and the first spoke for themselves through a very scep­ direct photo-electric determination of tical scientist". Millikan is only one of how much do the citizens Planck's h:' In fact it is difficult to find any countless examples of that sort. Planck published basis inMillikan's experimental himself, from 1900 on, was among the of Europereallyknow about papers of that struggle with his own ex­ most conservative and hesitant of the in­ pectations. His internal conflict was of a terpreters struggling with the quantum physics? somewhat different sort, as can be read in theory. His motivation was completely or­ a later passage ofthatsame speech:"[T]he thogonal to his findings [3]. What did ap­ general validity ofEinstein's equation is, I peal to him about his result that nowbears think, now universally conceded, and to his name, and which Millikan was the first that extent the reality of Einstein's light­ to measure with great precision, was Physicson stage quanta may be considered as experimen­ Planck's hope to have found a constant of tally established. But the conception oflo­ nature thatwould be absolute, inthe sense calized light-quanta out ofwhich Einstein that even extraterrestrials, despite all their hree majorEuropean research organ­ got his equation must still be regarded as differences from humans,would agree on T isations, the European Centre for Nu­ far from being established." [Emphases in the experimental value ofh. In the mean­ clear Science (CERN), the European Space original.] These sentiments would not time, we can be quite satisfied with the Agency (ESA) and the European Southern have shocked his audience in Stockholm, fact, so awkward for the "relativists:' that Observatory (ESO) are collaborating to­ for Einstein had received the Nobel Prize h is found to have the same value, when gether to organise a unique European­ in 1922,"especiallyfor his discovery ofthe measured with care or calculated from wide programme to raise the public law of the photoelectric effect:' excluding theory in Millikan's Chicago or anywhere awareness of physics and related sciences both relativity and quanta. Ironically, it else on this globe, regardless of environ­ with support from the European Union. had been Millikan's experiment which mental or social conditions, and even in Other partners in the project are the Eu­ convinced the experimentalist-inclined the face offalse initial presuppositions. ropean Physical Society and the European committee in Stockholm to admit Einstein Association for Astronomy Education. to that select circle. References In 22 European countries National Two final ironies: In 1950, at age 82, Mil­ 1. Physical Review 7 (March 1916), Steering Committees have been formed likan published his Autobiography. Chap­ 355-388.An abstract ofhis paper, andear­ with responsibility for their own national ter 9 is entitled simply"The Experimental ly results, had been published byMillikan programmes to promote national activi­ Proofofthe Existence ofthe Photon - Ein­ in Physical Review 4 (1914), 73 and 6 ties and to survey new and excitingedu­ stein's Photoelectric Equation:' By then, (1915),55.A much abbreviated version of cational approaches to physics. The pro­ Millikan had ofcourse come toterms with partofthis essay was published (electron­ jectwill culminate inNovember 2000 with the photon. Moreover, he had changed his ically only) in Physical Review Focus 3, 400 delegates converging on CERN in mind about what he had done around Story 23 (April 22, 1999). Geneva for the Physics on Stage festival. 1916. For now, he wrote (pp. 101-102) that During this event, science teachers, science as the experimental data became clear in 2. In his lecture "History as Myth and communicators, publishers, top scientists his lab, they "proved simply and ir­ Muse" (given at the University ofAmster­ and high-level representatives ofthe min- refutably, I thought, that the emitted elec­ dam, November 11, 1998), Roger H. . istries and European organisations will tron that escapes with the energy hv gets Stuewer concluded, concerning this brainstorm future solutions to bolster that energy by the direct transfer of hv change ofthe historic record, that"even an physics' popularity. units of energy from the light to the elec­ experimental physicist as great as Millikan The Physics on Stage festival will be tron [emphasis in original] and hence could be dead wrong in his theoretical part of the European Science and Tech­ scarcely permits of any other interpreta­ views - and then also was human enough nologyWeek, which was first launched in tion than that which Einstein had origi­ to try to hide his error much later by giv­ 1993 on the initiative of the European nally suggested, namely that of the semi­ ing a patently false account in his autobi­ Commission to raise public awareness of corpuscular or photon theory of light it­ ography" (p.17). science and technologyand to ensure that self:' In the end, Millikan re-imagined the the public recognises their significance in complex personal history of his splendid 3. For a brief account of that struggle, see all our lives. experiment to fit the simple story told in G. Holton, Thematic Origins of Scientific Luciano Maiani, the Director General so many ofour physics textbooks [2]. Thought (Harvard University Press, 1988), of CERN writes :"Science is a critical re­ The otherironyemerging from this sto­ pp. 156-159. source for mankind and among natural ryconcerns the fact that, as one ofmy per­ sciences, physics will continue to play a sonalcorrespondents has put it so well,"In The author is Gerald Holton at the Jeffer­ crucial role, well into the century. The the current deconstruction of science by son Physical Laboratory Harvard Univer­ young people of Europe deserve the best non-scientists, much is made of the ten­ sity Cambridge, MA possible physics teaching. An enormous dency of researchers to ask the questions resource of first-class teachers, teaching and analyze results in the light of their Phys. perspect. 1 (1999) 231-237 © materials andinnovative thinking exists in own beliefs. Millikan's paper is an excel­ Birkhaiiser Verlag, Basel, 1999 our countries. The Physics on Stage pro­ lent counter-example, being clearly a case 1422-6944:99:030231-07 $ 1.50_0.20:0 ject will bring them together to generate a ofsomeone doing superlative work in es­ new interest in physics education which tablishing a result from a theory in which will be to the long termbenefit ofchildren he himself clearly had little faith. In this all over Europe".

14 europhysics news MAY/JUNE 2000