JULY 1955

LINEAR ELECTRON ACCELERATORS FOR DEEP X-RAY THERAPY

Along with the development of linear electronaccelerators for nuclear physics research, similarmachines have been designed and built for medicalpurposes .. Linear accelerators are capable of pro-ducing an intense narrow beam of electrons havingenergies of several millions of electronvolts 1)2).When impinging on a suitable target, the electronsgenerate X-rays of high penetrating power, soproviding a very efficient source for deep therapy.

621.384.62 :615.849

Ministry of Health specification. The machine wasinstalled in Newcastle General Hospital duringAugust 1953 and has been in use for the treatmentof patients since December 1953. Since the X-raysgenerated by electrons in this energy range areemitted chiefly in a forward direction, the electronbeam itself must be swung round in order to providefor different angles of irradiation of a patient.This facility has been provided with the Newcastle

Fig. 1. Medi.callinear accelerator generating 4 MeV X-rays, .installed at Newcastle GeneraHospital (England). This was the first of a series of 4 MeV-accelerators to be used fordeep X-ray therapy.

The photograph shows a patient being set up for treatment. The double-ended gantryhouses the accelerator with accessories and X-ray head. The assistant on the left adjuststhe height of the couch and the angle of the gantry to provide the correct point andangle of incidence of the radiation.

The photograph fig. 1 shows a 4 MeV medicallinear accelerator designed and built at the MullardResearch Laboratories, Salfords (England), to

1) D. W. Fry, The linear electron accelerator, Philips tech.Rev. 14, 1-12, 1952/53.

machine by mounting the accelerator on a largedouble-ended gantry, the X-ray beam being directed .towards the axis of rotation where the patient is

2) C. F. Bareford and M. G. Kelliher, The 15 million electron-volt linear electron accelerator for Harwell, Philips tech.Rev. 15, 1-26, 1953/54.

31

32 PHILlPS TECHNICAL REVIEW VOL. 17, No. 1

located. Fig. 2 may serve to illustrate the generallay-out. A more detailed description of this equip-ment 3) will appear in a future issue of this Review.

Fig. 3 gives an impression of an even larger medi-cal linear accelerator producing 15 MeV X-rays.This machine has recently been installed in St.Bartholomew's Hospital Medical School, London,hy a team of scientists of the Mullard Laboratoriesheaded by T. R. Chippendale under the directionof P. E. Trier 4). The accelerator is similar to thatalready described in this Review 2). Swinging thewhole accelerator round for changing the angle ofirradiation is not a practical proposition for a ma-chine of this size. In this case, therefore, the accele-rating waveguide is mounted in a fixed horizontalposition and the electron beam emerging from it ishent through 90° by a powerful electromagnet be-fore hitting the target. The X-ray head housing the

3) A preliminary account was given by T. R. Chippendale andM. G. Kelliher, A linear accelerator for X-ray therapy,Discover 15, 397-404., 1954 (No. 10).

4) A 15 MeV linear accelerator for medical use, ElectronicEngineering 26, 527-528, Dec. 1954 (No. 12).

target and the electromagnet can he rotated aboutthe horizontal axis of the machine to enahle theangle of irradiation to be varied.

M

G1\I \I \

. ._._.L__ ..L._._._._._Cl

p

~HIT

Fig. 2. Schematic representation of the machine shown infig. LA is the accelerator proper (corrugated waveguide 1metre long), mounted in gantry G rotatable about the axis aby means of an oil motor D. E electron gun, X X-ray head,M magnetron, F feedback bridge, V diffusion pump, J rota-ting vacnum joint, W counterweights, P patient's couch,L water load.

Fig. 3. 15 MeV linear accelerator installed in its temporary building at St. Bartholomew'sHospital Medical School, London.

. JULY 1955 LINEAR ELECTRON ACCELERATORS 33

In fig. 3 it is seen that the target end and X-rayhead of the machine are supported from ahove to ,permit the positioning of patients underneath. Theangular position of the X-ray head is indicated bythe séale at top centre. The Xvray beam field size,which may he adjusted using the control huttonson the head, is indicated on the two lower scales.

Four o'f the six metres length of the acceleratorare housed in a separate room which is dividedfrom the treatment room by a concrete wall I metrethick. The gap in the wall visible in the phtographand permitting a complete view of the acceleratoris normally closed with concrete blocks.

T. R. CHIPPENDALE.

ABSTRACTS OF RECENT SCIENTIFIC PUBLICATIONS OFN.V. PHILIPS' GLOEILAMPENFABRIEKEN

Reprints of these papers not marked with an asterisk • can be obtained free of chargeupon application to Philips Electrical Ltd., Century House, Shaftesbury Avenue, LondonW.C.2.

2161: R. Vermeulen: Sound recording. Generalreview, presented at the first I.C:A. Congresson Electro-accoustics, Delft, 1953 (Acustica4, 17-21, 1954.)

General considerations on sound recording. Thefollowing topics are dealt with; improvements ingramophone records, gramophone recording andreproduction, tape recorders; information, theoryin relation to reproduetion of music, fidelity of thereproduction, stereophony, artificial reverberation.

2162: J. Rodrigues de Miranda: The radio set asan instrument for the reproduetion of music(Acustica 4,38-41, }954).

In connection with the reproduetion of music byradio the following desiderata are put forward:1. Distortion should be decreased. as the high fre-quency range increases. 2. Means are desirahle foradjusting bass response, for cutting óff and for gra-dual attenuation of treble, 3. The electro-acousticalengineer and the architect should cooperate closely.The cabinet must he rigid, the loudspeaker placedforward; its cloth chosen carefully. 4. The speakershould be chosen in accordance with the acousticalproperties of the .cabinet, The audio-frequencycharacteristic of the set should' he carefullydetermined.

2163: W. K. Westmijze: Application of the recipro-city theorem to magnetic reproducing heads(Acustica 4,50-52, 1954). 'ill!) ,

It is shown that the output of a magnetic repro-ducing head can be calculated if the field distribu-tion in front of the head is known for the case whenthe latter 'is energised, Application of this method'enables us to predict the response curve of a wide-g'ap head and to explain the existing differencebetween wide-gap and narrow-gap measurements.

2164: A. G. Th. Becking and A. Rademakers.Noise in condenser microphones (Acustica'4,96-98, 1954).

It is pointed out that the mecha'nical resistance,used for damping a simple mechanical oscillator,acts as the source of á fluctuating force withspectral intensity 4RkT. In the case of a eondensermicrophone this results in a noise pressure on thediaphragm, which lies in the region of audible soundpressures. Experimental -results, obta~ed with twomicrophones, agree well with th~ theory. . '

.,

2165: A. G. Th. Beckint Methods of measurementfor hearing aids (Acustica 4, ~43-145, 19~4).

The different ways of defining the input soillidpressure on a hearing aid are discussed. In addition,a comparison is given of thespecifications for 'accept-able hearing ai~s in s?me Europeancountries.

2166: J. Volger, J. M. Stevels and C. van Ame-rongen: The dielectri~ l'elaxati0!1 of glass andthe pseudo-capacity of metal-to-glass inter-faces, measured at extremely low frequen~cies. (Philips Res. Rep. 8, 452-470, 1953,No.6).

See R 232.

. '

. 2167*: H. ~reminer: On.a phase-contrast theory ofelectron-optical image formation. (ElectronPhysics, Circular 527; Nat. Bur. Stand.,March 1954, 145-158).

The scattering of electrons by an object, as appliedin electron microscopy, can he dealt with by meansof an integral equation, which follows from theSchrödinger equation corresponding to the electro-static potential field inside the object. This integralequation can be solved by a series, the consecutive

PHILIPS TECHNICAL REVIEW VOL. 17, No. 1

terms of which may be interpreted as the contri-bution due to repeated scatterings of the incidentelectron waves by the object. Summation of themain contrib~tion (for very small electron wave-lengths) of these terms leads to a wave function ,inthe object plane (this is a plane immediatelybehindthe object), which in every point depends only onthe potential distribution along tbe straight linejoining this point with the electron source. Fromthe values of the wave function in the object plane-it is.possible, by well-known mathematical methods,to derive the behaviour of the wave function in thespace behind that plane. In this way it is possible,in principle, to investigate also the image formation.

2168: J. Haantjes: Die Fernsehübertragung derenglischen Kröningsfeier nach dem Fest-land (Fernmeldetechn. Z.7, 129-133, 1954,No. 3). (Television transmission of theEnglish. coronation festivities to the con-tinent; in German.)

See Philips techno Rev:. 15, 297-306, 1953.

2169: A. M. Kruithof and A. L. Zijlstra : Lespropriétés élastiques d'un certain type deverre (Verres et Réfractaires 8, 1-13, 1954,No. I). (The elastic properties of a certaintype of glass; in French.)

The disappearance of internal stresses in glass atconstant temperature, turns out to be dependent onthree factors: the viscosity, the instantaneous

/' elasticity and the elastic after-effect. The first part. of the article deals with these factors for a certaintype of glass. Special attention is paid to the elasticafter-effect. This can be described. by 'a so-calleddelayed elongation which, after the sudden appli-cation of a constant stress, can be represented as afunction of time by the sum of twó exponentialfactors.

In the second part,. a formula is derived for thedisappearance of internal stresses in a rod ofstabilized glass at constant temperature, which issuddenly stretched to a specified exent by the appli-cation of a constant tensile force. As a result of themutual effects of viscosity, elasticity and elasticafter-effect the resultant stresses disappear. Theirdisappearance can be described by the sum of threeexponential functions. This formula is testedexperimentally. The results agree very well withthe theory.

2170: E. J. W. Verwey: Das Kräftespiel zwischenTeilchen in lyophoben Kolloidsystemen(Kolloid Z. 136, 46-52, 1954). (The effect

of forces between particles in lyophobiccolloids; in Ger~an.)

It appears from the theory that the behavior andparticularly the stability of lyophobic colloids is inaccordance with the balance between the Van del'Waals-London forces of attraction and the repul-sion forces of the electrolytic double layer. The natureof these forces is examined more closely. The resultsof the theory point to an elegant explanation of thewell known Schulze-Hardy rule, and thereby afurther extension of our knowledge of van del'Waals-London forces. .

2171: H. J. G. Meyer: Theory of radiationlesstransitions of F centres (Physica 20,181-182,1954.).

Calculations are reported, based on the Huang-Rhys model, on the' probality of radiationless transi-tions in F centres. By using a very accurate approxi-mation formula- for modified Bessel functions oflarge index and large argument, an expression isderived which allows a simple physical interpreta-tion for the cases of high and low temperature andwhich is easily calculated numerically. It is shownthat in principle with this model large probahilitiesfor radiationless transitions can be obtained.

2172: J. Volger: Electrical properties of ceramics,Part I.Semiconductors (Research 7,196-2031954).

A review is given of some ceramic products ofvalue in the electrical industry, most of the materialsconsidered being oxides of the transition metals.Some introductory remarks on their crystal struc-tures and on the occurrence of certain suhatitutionsand imperfections in the lattice are followed by adiscussion on' their electronic conductivity and inparticular on the relation between the conductionmechanism and the occurrence of certain transitionmetal ions in different valency states. The influenceof the polycrystalline nature of these materials onconduction and on a number of second-order con-duction effects (e.g. Hall effect and frequencydependence of resistivity and magneto-resistance)is briefly discussed. Some applications of ceramicoxidic semiconductors are mentioned.

2173*: C. J. Bouwkamp: Diffraction theory (Rep ..Progr. Phys.17, 35-100, 1954).

A critical review is presented of recent progress inclassical diffraction theory. Both scalar and electro-magnetic problems are discussed. The report mayserve as an introduetion to general .diffraction

JULY 1955 ABSTRACTS OF RECENT SCIENTIFIC PUBLICATIONS 35

theory although the main emphasis is on diffractionby plane obstacles. Various modifications of theKirchhoff and Kottler theories are presented.Diffraction by obstacles small compared with thewavelength is discussed in some detail. Other topicsincluded are: variational formulation of diffractionproblems, the Wiener-Hop( technique of solvingintegral equations of diffraction theory; the rigorousformulation of Babinet's principle, the nature of:field singularities at sharp edges, the application ofMathieu functions and spheroidal wave functions to ,,diffraction theory. Reference is made to more than500 papers published since 1940.

2174: J. Volger: Electrical properties of ceramics,Part 11. Dielectries and ferromagnetics(Research 7, 230-235, 1954).

After some remarks on ceramics. as insulatingmaterials, the use of ceramic dielectrics for use incondensors is discussed. The factors favouring ahigh value of the dielectric constant are considered 'and in particular the occurrence of such high valueswith BaTi03. A number of interesting properties ofsoft ferromagnetic ceramics than are given, all ofthe ferrite or - ferroxcube type., The saturationmagnetization values are discussed in connectionwith the theory of Nëel, Attention i,Sgiven to theh.f.properties of ferroxcube, i.e. the various resonanceand relaxation phenomena which occur. Finally,a hard ferromagnetic ceramic called magnadur (also,known as ferroxdure) is dealt with. This is ahexagonal oxidic compound of iron and bariumwith an extremely large coercive force. It may beused for permanent magnets which must withstandlargè demagnetizing fields.

2175: R. Vermeulen: Stereophonic reproduction(Audio Engineering 38,21,1954, No. 5)~

Recapitulation of the work of K. de Boer on themechanisms of binaur~l and stereophonic soundphenomena. The perception of direction is discussedas being due to differences both in the time ofarrival of the sound and in the intensity. Thedifferences between binaural and stereophoniclistening are outlined; in the latter case the soundimage is less well defined. It is pointed out that theaim is not to try to reconstruct the original soundfield, hut to deliver the correct sound to each ear tosimulate a sound source from a certain direction.·Some empirical rules for the placing of microphonesand loudspeakers are given. Finally mention is madeof the part played by head movements in the locationof the sound image by the listener.

2176: J. L. H: Jonker: Ten-volt effect with oxide-coated cathode (Electronic Engineering 26;282, 1954). .

According to a theory given earlier, the anomalyin the diode characteristic that occurs at an anodepotenrial. of 10 V should be attributed to the factthat the space-charge contribution of electronsreflected from the anode then shows a minimum(see Philips Res. Rep. 2, 331-339, 1947). Thistheory is confirmed by measurements on the reflec-tion and secondary emission of a surface that hadbeen maintained for some time facing an emittingoxide cathode.

2177: B. Combée and A. Engström: A new devicefor micro-radiography and a simplified tech-,nique for the determination of the mass ofcytological structures (Biochim. et Bioph,Acta 14,432-434,1954).

A sealed-off X-ray tube with a very thin Bewindow is used' as a source of soft Xvrays for quanti-tative historadiology by means of contact micro-radiography. A procedure is described whereby themass of the specimens may be determined withoutthe necessity of reference exposures. 'The resolutionofthe image obtained with this tube is about 0.5 IL.

2178: H. de Lange: Relationship between criticalflicker-frequency and a set of low-frequencycharacteristics of the eye (J. Opt. Soc.America 44,380-389,1954)

Measurements of the critical flicker frequency ofthe eye as, a function of both the average luminanceof the test field and the time variation of thisluminance were recorded by plotting the "rippleratio;' r against the critical frequency; r is defined as(amplitude of first Fourier component)/(averageluminance) of the stimulus. It is shown that withconstant average luminance, the points observedfor various time functions :fitinto one smooth curve,which for low luminance is monotonic. At highluminances the curve shows a minimum in r at acritical frequency of about 9 cis. This means thatthe eye has a maximum sensitivity to flicker atthis frequency.

2179: L. A. }E. Sluyterman and B. Labruyère:Side reactions in the polymerization of a-amino acid N-carbonic anhydrides. Titra-tion data of polyglycine and polyala~ine(Rec. trav. chim. Pays-Bas 73, 347-354-1954).

-Polyglycine and polyalanine obtained by poly-merization of their N-carhonic anhydrides often

36 , PHILIPS TECHNICAL REVIEW VOL. 17, No. 1

appeared to certain more acid groups than ~nimo'groups. This is ascribed, at least partly, to the for-, mation of hydantoin groups. A small amount of2.5-dik~topiperazme was isolated from polyglycine.A discussion of the results and their bearing uponthe calculation of molecular weights is given,

2180: F. A. Kröger, H. J. Vink and J. van denBoomgaard: Controlled conductivity inCdS single crystals (Z. phys. Chem. 203, 1-721954, No. 1/2). '

The electrical properties of single crystals of pureCdS or of CdS containing gallium, indium, antimony,chlorine or silver depend markedly on the atmo-sphere of preparatien. Crystals subjected to anoxidizing atmosphere (i.e. sulphur vapour) areinsulators or semi-conductors, showing photocon-ductivity; crystals subjected to a reducing a~mos-phere show quasi-metallic, electronic conductivity,For crystals doped 'with Ga or C;1, the number ofcarriers in. the reduced crystals is constant over. a,Wide range of temperatu~es" a~d equal to the eon-centrations of foreign ions' oyer a' wide range ofatmospheres {controlled valency}, The optical pro-.perties (absorption, fluorescence] vary also with theatmosphere, absorption bands in the yellow part ofthe spectrum' appearing' in oxidized, but not inreduced crystals. A general theory is developed byextending Schottky and Wagners theory of latticeimperfections along the lines indicated by Schottky ,(1935), By means of this theory it is possible tocalculate the' dependence of the concentratien ofthe various kinds of lattice imperfections, donors,traps, and acceptors, on the concentratien andnature of the impurities and the reducing power ofthe atmosphere. Applying this theory to CdS, asatisfactory agreement with. the experiments isobtained.

2181:" J. te Winkel: Lijnversterkers voor draag-'golfsystemen op coaxiale kabels (De Inge-nieur 66, E.61-65, 1954, No. 25). (Lineamplifiers for carrier telephone systems oncoaxial cable; in Dutch.)

A brief survey of -the principles governingthe design of these amplifiers.

2182: E. Havinga and J. P. L. Bots: Studies onVitamin D, 1. The synthesis of vitamin n,3 C14(Rec. trav. chim. Pays-Bas73, 393-400,1954).

In connection with the study of the metabolism ofvitamin P, vitamin Da has been synthesised with aC14=v= in place of the third carbon atom. The pre-paration of this vitamin n;: 3 C14 is described.Special attention is paid to the technique of thephotochemical conversion of 7-dehydrocholesterol3 C14(small quantities) and to the isolation of thelabelled vitamin D, 'from the irradiated mixture.Some biological ~xpèriments are briefly described: '

2183: G. W. Rathenau and ç.. Baas: Croissance desgrains, observée par microscopie electroniqueà emission (Métaux, Corros. Ind, 29, 139-150,1954 (No. 344). (Grain growth observedby means of the emission' electron micros-cope:' in' French.)

Changes in structure.in metals and alloys atelevated temperatures can directly, be observedapplying e~ssion' electron microscopy. With theinstrument' ~sed for these investigations a resolvingpower ~f about 1000 Á is obtained.

Rolled face-centred cubic NiFe alloys have beenused to study grain growth.in a texture. This,'investigation includes growth of one large grain ofdeviating orientation in a well-pronounced textureas well as grain growth in an imperfect texture. Inboth cases only high energy boundaries move atmeasurable speed. The results are discussed interms of interfacial tension effects.

An example of grain growth hindered by inclu-sions is given for a Cr Ni 'steel specimen, containinginclusions of a-phase.

Grain growth accompanying the transformationa-+y in a SiFe alloy has been described. In this casethe' rate of diffusion determines the rate of growthof the y grains.The austenite-pearlite transformation in eutectoid

carbon steel, has been directly observed. The exis-tence of an' orientation relationship betweenaustenite and pearlite is improbable.as shown by the'observations. Grain-boundary displacements in, the.austenite have been observed near a growingcolony of pearlite.