Philips Tech. Rev. 42, No. 3, 93-96, Oct. 1985 93

Continuous skull-melting of glass .

In the manufacture of pure and homogeneous softglass, e.g. as used for glass fibres in data transmission,the conventional method of batch production in acrucible or in a tank furnace - for large quantities -is unsuitable. This is because glass produced in thisway always contains impurities originating from thewall material, and these effects increase with the re-activity of the molten glass and the wall material. Noris the batch production of glass compatible with con-tinuous production processes such as the double-crucible method used in the manufacture of opticalfibres [11.Philips Research Laboratories in Eindhovennow have an experimental arrangement, based on the'skull-melting' method, which can produce pure andhomogeneous glass with no inclusions or gas bubblesin a continuous process.

'Skull melting' [21 is an established method ofmanufacturing pure refractory materials. The melt iscontained in a shell or 'skull' of the cooled unmolten 109.Q cmmaterial. Since the melt is not in contact with a cruc-ible wall there is no risk of it becoming contaminated.Another advantage, compared with the production ofglass in a tank furnace, is that the type of glass beingmade can be changed quickly. The material is heatedby r.f. heating, a method in which eddy currents areinduced in the material by a rapidly alternating mag-netic field, so that energy is transferred to it.The principle used in the skull-melting method is

that the electrical conductivity of molten oxides ismuch greater than that of oxides in the solid state,partly because of higher ion mobility. The resistivityof pure crystalline oxides as a function of temperaturehas a step near their melting point. Glass, whichbehaves as a supercooled liquid during cooling, doesnot have a melting point, but a melting range. Thecurve of the resistivity of glass as a function of tem-perature is therefore more gradual; this is illustratedin fig. 1 for a silicate glass with 26.56,10 by weight ofNa20 [31. The discontinuity in the curve is near thetransition temperature, the temperature at which adiscontinuity also appears in the curves for o~herphysical quantities. The viscosity changes from about1019Pa s at 20°C to about 102 Pa s at 1200 °C; at thetransition temperature the viscosity of all types ofglass is 1012Pa s [41.

In the r. f. heating of molten oxides in a skull of un-melted material the skull is 'transparent' to the energyin the magnetic field. The field can however deliverenergy to the molten material. Sufficient energy canonly be transferred if the resistivity, for, particular

skull dimensions, is lower than a threshold valueproportional to the frequency of the r.f. field. At2.6 MHz, the frequency that we use, this thresholdvalue. is about 100 ohm cm-Starting the melting pro-cess is thus a problem, and we shall return to thispresently.Until now the skull-melting method has been used

only in batch production. This is done in a 'cold cruc-ible' formed by free-standing water-cooled axial cop-per fingers inside the r .f. heating coil [21. The coil hasalways taken the form of a copper tube with coolingwater flowing through it. The distance between coiland crucible must be relatively large, otherwise ioniza-tion of the air could cause voltage breakdown (flash-over). In such an arrangement flashovers can also

1

1000 600 400

10'

_T-1

Fig.!. The electrical resistivity {l of glass with 26.5% by weight ofNa20, the rest Si02, as a function of the reciprocal of the absolutetemperature [3J. (A scale in °C is given at the top.) The transitiontemperature of the glass is near the discontinuity in the curve at430°C. In the low-temperature range the curve corresponds to thefunction {l = {lo exp (Q/RT), where {lo is a constant, Q is the activa-tion energy and R the gas constant, so that with the scale divisionsused a straight line is found in this range.

[IJ H. M. J. M. van Ass, P. Geittner, R. G. Gossink, D. Küppersand P. J. W. Severin, The manufacture of glass fibres foroptical communication, Philips Tech. Rev. 36, 182-189, 1976.

[2J V. I. Aleksandrov, V. V. Osiko, A. M. Prokhorov and V. M.Tatarintsev, Synthesis and crystal growth of refractory mat-erials by RF melting in a cold container, in: E. Kaldis (ed.),Current topics in materials science, Vol. 1, North-Holland,Amsterdam 1978, pp. 421-480.

[3J R. H. Doremus, Glass science, Wiley, New York 1973.[4J H. Scholze, Glas, 2nd edition, Springer, Berlin 1977.

94 w. C. P. M. MEERMAN et al. Philips Tech. Rev. 42, No. 3

occur between the turns of the coil, since the voltagebetween turns can rise to 1000 V. The risk of flash-overs also increases with the frequency. The energytransfer is proportional to the fourth power of theratio of the outside diameter of the molten glass to theinside diameter of the coil [51. For efficient energytransfer this ratio should be as large as possible, butthis conflicts with the need to have large air gaps toavoid voltage breakdown. At Philips Research Labor-atories a continuous skull-melting process was there-

There are three distinct zones in the glass mass inthe inner tube during the process:• The outer, ring-shaped 'skull' of cooled, unmeltedmaterial, which absorbs no energy from the r. f. fieldbecause of the high resistivity; the radial temperaturegradient in this zone can be as high as 500°C/mm .• A hot outer liquid zone in which the resistivity islower than 100 ohm cm; the eddy currents are concen-trated here, because of the skin effect; the maximumtemperature in this zone is 1650 °C.

G

.---f,~-+- CJ------I~:;:t__ C

.r------H~4_ PO

HF

~.~~~--_~l''--'

-" -

Fig. 2. a) The experimental arrangement for producing pure glass rods. PI inner tube of silicaglass. PO outer tube of silica glass. C r.f. coil. Cl cooling jacket with deionized water. RM feedof raw mix. L liquid glass. S solidified glass. (In reality the transition from Sto L is much moregradual than shown here.) HF connections to r.f. generator. Wi, Wo inlet and outlet for coolingwater. OR the glass rod produced. AM mechanism for adjusting the inner tube. b) Interior on alarger scale. (For clarity the distance between the two tubes is exaggerated.) 0 earthed screen ofvertical copper wires, extending over the whole circumference between the inner and outer tube .

fore developed from completely different principles,with the risk of dangerous flashovers now completelyeliminated.

The new arrangement is shown in jig. 2. The coil issurrounded by cooling water, deionized to ensure highresistivity. The 'mix' of raw materials for the glass isfed into a tapering silica-glass inner tube. This tube,in which the actual process takes place, is surroundedby a silica-glass outer tube. Between these concentrictubes there is a strong flow of deionized cooling water.

• A colder inner liquid zone in which, although theresistivity is sufficiently low, there are no eddy cur-rents, because of the skin effect; the temperature inthis zone is about 1550 "C.

The heat transfer from the outer liquid zone to theinner one takes place by convection and radiation.Fig. 2b shows the flow pattern produced in the moltenglass. This flow is maintained by density differencesresulting from the temperature gradient in the melt.Because of the intensive circulation and the low vis-

Philips Tech. Rev. 42, No. 3 CONTINUOUS SKULL-MELTING OF GLASS

cosity of the hot glass, large gas bubbles can easilyescape at the outside. The small gas bubbles concen-trate in the upper liquid region, so that the lowerregion is free of impurities. (This is comparable withthe situation in zone melting.)

95

where material is fed in. The heat transfer from themolten zone to the cooling water takes place mainlyby radiation. The silica of the inner tube is almostcompletely transparent to radiation at a wavelengthof less than 4 urn, so that this heat radiation is directly

I1~

Ic

Fig. 3. a) The arrangement with the control equipment. b) The process in operation; the white-hotglass rod is clearly visible. c) The product. (The changes in diameter are due to changes in processparameters.)

Overheating of the molten glass cannot cause vol-atile components (e.g. B203 or ZnO) to escape, be-cause these components condense in the colder zone

[SJ J. F. Wenckus and W. P. Menashi, Growth of high purityoxygen-free silicon by cold crucible techniques, Report RADC-TR-82-17 I (NTIS Order No. AD-Al 18992), 1982.

absorbed by the cooling water without causing any in-crease in the temperature of the silica. (In the conven-tional crucible with fingers, the heat has to be trans-ferred to the cooling water through the copper.)

At the start of the process the passage through thetapering part of the inner tube is closed off by a glass

96 CONTINUOUS SKULL-MELTING OF GLASS Philips Tech. Rev. 42, No. 3

rod. The mix is then poured into the inner tube. Theprocess is started by inserting a silica tube containinga carbon rod surrounded by a shielding gas into themix. Because of its low resistivity the carbon rod ab-sorbs energy from the r.f. field, so that it heats up andmelts the surrounding mix. When sufficient liquidglass has formed, the silica tube with the carbon rod isremoved. The mass is then heated further, and a bathof molten glass forms inside the mix. Once the righttemperature distribution has been reached inside thetapered constriction, a rod of clear glass can be drawnfrom it with very little force.The viscosity of the molten glass at the constriction

determines the mass flow. This viscosity depends onthe temperature and hence on the transferred power.The mass flow can therefore be varied by changing theposition of the constriction in relation to the r.f. coil,using a mechanism for adjusting the inner tube asillustrated in fig.2a. The rate at which the rod is with-drawn from the inner tube determines the diameter.The present arrangement can produce 75 to 100gramsof glass per hour with 10 to 15 kW of continuous-wave power from the r.f. generator and a rated powerof 25 kW.The capacitive coupling of the r.f. coil to the elec-

trostatic field could cause plasma discharges in thegases escaping from the mix. (The gas has a lower di-electric constant than the surrounding material.) Anearthed screen is therefore mounted in the coolingwater between the coil and the inner tube. The screenconsists of vertical copper wires, see fig. 2b; it passesthe magnetic field and screens off the electrostaticfield.

The transferred power is controlled by keeping theanode current of the output valve of the r. f. generatorconstant by means of a PID controller. Since almostthe entire transferred power enters the cooling water,the product of the volume flow and the increase intemperature of the cooling water is a measure of thispower. Both quantities are therefore continuouslymeasured and monitored. Control of the anode volt-age of the generator is not practical, since a very smallincrease in the voltage causes an almost explosive in-

crease in the transferred power. The reason for this isthat changes in the transferred power are accompaniedby changes in other process parameters, such as theresistivity and the thickness of the skull, and hence theratio of diameters mentioned above, so that the trans-ferred power rises even more.

As we mentioned earlier, the glass production de-pends on the temperature of the liquid glass at thelocation of the constriction in the inner tube, and isconsequently determined by the heat economy in theinner tube. The level of the glass bath is therefore con-tinuously monitored, and adjusted if necessary bychanging the amount of mix fed in per unit time. Thelevel of the mix is measured by radiation detectorsaround the circumference of the inner tube.

In the conventional method of glass production theglass remains in the tank furnace for many hours,sometimes longer than a week. In the method des-cribed here the glass remains in the hot zone for nolonger than two hours. Assuming a 'classical' mix con-sisting of a dry mixture of crystalline raw materials,the composition of the initial charge will not usuallybe sufficiently homogeneous. Since there is insufficienttime for diffusion of the constituents in the glass bath,it is better to prepare the constituents so that thecharge will have the correct composition right fromthe start. This is often done by wet granulation orpelletization of the mix.

With the arrangement described here we have madeglass rods with a diameter of about 4 cm that are freefrom gas inclusions and are of pure composition.Fig. 3 shows photographs of the experimental arrange-ment and a rod produced in it.

W. C. P. M. MeermanT. L. van RooyM. C. M. Voss

w. c. P. M. Meerman and Ing. T. L. van Rooy are with PhilipsResearch Laboratories, Eindhoven; Ing. M. C. M. Voss, formerlywith Philips Research Laboratories, is now with the Philips Light-ing Division, Maarheeze. .

Philips Tech. Rev. 42, No. 3, 97-100, Oct. 1985

Scientific publicationsThese publications are contributed by staff of laboratories and plants that form part of orcooperate with enterprises of the Philips group of companies, particularly by staff of the re-search laboratories mentioned below. The publications are listed alphabetically by journal title.

1.Donjon

P. J. 't Hoen (Philips Ultrasound,Santa Ana, CA)

C. Méquio, R. H. Coursant & P. Pes-qué L

P. Pesqué, R. H. Coursant & C.Méquio L

M. NaiIlon, R. H. Coursant & F. Bes-nier (CISI, Paris) L

F. Hottier & J.-L. Bernatets

M. Fink, J.-F. Cardoso & P. Laugier(UER Cochin Port-Royal, Paris) L

F. Towfiq*, C. W. Barnes (Univ. Cali-fornia, Irvine, CA) & E. J. Pisa*(* Philips Ultrasound, Santa Ana,CA)

H.Schomberg, W.Beil, G.C. McKin-non, R. Pro ksa & O. Tschendel H

B. Jacobs, J. Braat & M. HartmannE,H

G. B. McMiIlan (INMOS, Newport),D. J. Smith*, J. P. Gowers & H.Ahmed* (* Univ. Cambridge) R

Philips GmbH Forschungslaboratorium Aachen, AWeiJ3hausstraJ3e,5100 Aachen, Germany

Philips Research Laberatory Brussels, B2 avenue Van Becelaere, 1170 Brussels, Belgium

Philips Natuurkundig Laboratorium, EPostbus 80000,5600 JA Eindhoven, The Netherlands

Philips GmbH Forschungslaboratorium Hamburg, HVogt-Kölln-StraJ3e 30, 2000 Hamburg ~4, Germany'

Laboratoires d'Electronique et de Physique Appliquée, L3 avenue Descartes, 94450 Limeil-Brévannes, France

Philips Laboratories, N.A.P.C., N345 Scarborough Road, Briarcliff Manor, N.Y. 10510, U.S.A.

Philips Research Laboratories, RCross Oak Lane, RedhiIl, Surrey RHI 5HA, England

Philips Research Laboratories Sunnyvale SP.O. Box 9052, Sunnyvale, CA 94086, U.S.A.

L The place of ultrasonics in medical imaging Acta Electronica 25

Design of ultrasonographic linear arrays Acta Electronica 25

Simulation of the acousto-electric response of piezo- Acta Electronica 25electric structures by means of a Fast Fourier-Trans-.form algorithm

Methodology for the characterization and design of Acta Electronica 25linear arrays of ultrasonic transducers

Analysis of piezoelectric structures by a finite ele- Acta Electronica 25ment method

L Estimation of ultrasonic attenuation in biological Acta Electronica 26tissues

Diffraction effect analysis in medical echography Acta Electronica 26

Tissue classification based on autoregressive modelsfor ultrasound pulse echo data

Acta Electronica 26

Ultrasound computerized tomography Acta Electronica 26

Aging characteristics of amorphous magnetoopticrecording media

Transmission electron microscopy and high resolu-tion electron microscopy studies of shallow (Rp - 20nm) As and B implanted and electron beam annealedsilicon

Appl. Opt. 23 .

Appl. Phys, Lett. 44

J. M. Woodcock & J. M. Shannon R Thermionic emission in bulk unipolar camel diodes Appl. Phys. Lett. 45

G. J. van Gurp, P. J. de Waard & Thermomigration in indium filmsF. J. du Chatenier E

P. Dawson & K. Woodbridge

J. S. Nadan

J. M. Goldstein*, Y. Z. Efe* (* Coo-per Union School of Eng., NewYork) & J. J. Bernitt N

Appl. Phys. Lett. 45

R Effects of prelayers on minority-carrier lifetime in Appl, Phys. Lett. 45GaAs! AlGaAs double heterostructures grown bymolecular beam epitaxy

N A glimpse into future television Byte 10

97

A fast compositional modeling system for computergraphics

Control & Comput. 12

293-300 1983

301-310 1983

311-323 1983

325-340 1983

341-362 1983

33-58 1984

59-80 1984

95-110 1984

121-128 1984

3979-3982 1984

1081-1083 1984

876-878 1984

1054-1056 1984

1227-1229 1984

135-150 1985

34-38 1984

98 SCIENTIFIC PUBLICATIONS Philips Tech. Rev. 42, No. 3. .

G. B. McMilIan*, J. M. Shannon & Processing of unipolar diodes with electron beams Electronics Lett. 20H. Ahmed* (* Univ. Cambridge) R

863-865 1984 .

A. Thayse B Synthesis and asynchronous implementation of algo- IEEE Trans. C-33rithms using a generalized P-function concept

861-868 1984

J. L. W. Kessels

S. B. Luit jens, J. W. Smits & V.Zieren E

J. J. M. Ruigrok

J. J. M. Ruigrok

J. W. D. Martens & A. B. VoermansE

A.' M. van der Kraan (Interuniv.Reactor Inst., Delft) & K. H. J.Buschow E

G. A. C. M. Spierings (Philips GlassDiv., Eindhoven), C. M. G. Jochem,T.P.M.Meeuwsen&G.E.Thomas E

A. H. van Ommen

H. A. van Sprang & J. L. M. van deVenne E

C. P. Janse & A. J. M. Kaizer

P. J. Schoenmakers

T. F. McGee Ill, C. Werkhoven & J.Jansen N, E

W. H. de Roode (Philips LightingDiv., Eindhoven) & C. A. P. W. vande Pavert E

W. J. M. J. Josquin & M. J. E. Ule-naers E

J. Boersma*, J. J. E. Indenkleef*(* Univ. of Technoi., Eindhoven) &H. K. Kuiken E

D. B. M. Klaassen, D. M. de Leeuw& T. Welker .....E, A

P. Schobinger-Papamantellos (ETH-Zentrum, Zürich) &K.H. J. Buschow

E

S. Sinnema*, R. J. Radwanski*,J. J . M. Franse" (* Univ. Amster-dam), D. B. de Mooij & K. H. J.Buschow E

E. Dormann*, U. Dressel* (* Univ.Bayreuth), H. Kropp (TH Darmstadt)& K. H. J. Buschow E

V. Weissenberger*, B. Elschner*(* TH Darmstadt) &K.H. J. Buschow

E

E Two designs of a fault-tolerant clocking system

The write field of a ring head for a double layer per-pendicular medium

E Analysis of metal-in-gap heads

E Cross measurements in magnetic recording: a newway of determining head performance

Cobalt ferrite thin films for magneto-optical record-ing

Magnetism and 57FeMössbauer effect in amorphousFe-base alloys: significanee of charge transfer andshort-range atomic ordering

A new class of glasses for double-crucible opticalfibers with high numerical apertures

E Diffusion of ion-implanted As in Si02

Influence of the surface interaction on thresholdvalues in the cholesteric-nematic phase transition

E The Wigner distribution: a valuable tool for investi-gating transient distortion

E Thermodynamic model for supercritical fluidchromatography

Segregation coefficients of selected impurities inZnSe grown by LPE

Incorporation of manganese in LPE-grown Mn,YIGfilms in the presence of Ca2+ and Ge4+ ions

Oxidation-induced defects at the poly/mono siliconinterface

A diffusion problem in semiconductor technology

Systematic analysis of phosphor degradation undercathode-ray excitation

Magnetic structure and incommensurate phase tran-sition in HoGe

Magnetic properties of ternary rare-earth compoundsof the type R2Fe14B

Quadrupolar interaction of gadolinium nuclei at thecubic sites of ferromagnetic GdAI2

Ferromagnetic resonance in amorphous Y1-xCOx

IEEE Trans. C-33 912-919 1984·

IEEE Trans. MAG-20 724-726 1984

IEEE Trans. MAG-20 872-874 1984

IEEE Trans. MAG-20 875-877 1984

IEEE Trans. MAG-20 1007-1012 1984

IEEE Trans. MAG-20 1284-1289 1984

I

J. Am. Ceram. Soc. 67 657-663 1984

J. Appl. Phys. 56 2708-2715 1984

J. Appl. Phys. 57 175-179 1985

J. Audio Eng. Soc. 32 868-882 1984

J. Chromatogr. 315 1-18 1984

J. Cryst. Growth 59 649-650 1982

J. Cryst. Growth 69 173-181 1984

J. Electrochem. Soc. 2380-2386 1984131

J. Eng. Math. 18 315-333 1984

J. Lumin. 31 & 32 687-689 1984

J. Magn. &Magn. 149-157 1984Mater.44

J. Magn. &Magn. 333-341 1984Mater.44

J. Magn. &Magn. 207-218 1984Mater.45

J.Magn.&Magn. 19-28 1984Mater. 46

G. deWith E Note on the use of the diametral compression test J. Mater. Sci. Lett. 3for the strength measurement of ceramics

1000-1002 1984

B.Nienhuis, H.J.Hilhorst* & H.W.J.Blöte* (* Univ. of Technoi., Delft) E

J. C. M. Henning, J. P. M. Ansems&A. G. M. de Nijs E

'Triangular SOS models and cubic-crystal shapes

Photoluminescence excitation of Saxena's deepdonor in AlG aAs

J. Phys. A 17 3559-3581 1984

J. Phys. C 17 L915-L921 1984

Philips Tech. Rev. 42, No. 3

P. Friedel, P. K. Larsen, S. Gourrier,J. P. Cabanie & W. M. Gerits L, E

. J. M. B. Terken (Inst. for PerceptionRes., Eindhoven)

P. Delsarte & Y. Genin

P. van Dooren

F. L. van Nes (Inst. for PerceptionRes., Eindhoven)

C. M. J. van Uijen & J. H. den BoefE

M. J. Powell

P. Branquart

K. H. J. Buschow

SCIENTIFIC PUBLICATIONS

Photoemission studies of the interaction of hydro-gen plasrnas with GaAs (OOI)

The distribution of pitch accents in instructions as afunction of discourse structure

B Spectral properties of finite Toeplitz matrices

B A unitary method for deadbeat control

J. Vac. Sci. &Technol.B2

Lang. & Speech 27

99

675-680 1984

269-289 1984

Lect. Notes Control & 194-213 1984Inf. Sci., Vol. 58,P. A. Fuhrmann (ed.),Springer, Berlin

Lect. Notes Control & 864-881 1984Inf. Sci., Vol. 58,P. A. Fuhrmann (ed.),Springer, Berlin

Perceptuallimits in man-machine communication Limits in perception,A. J. van Doorn, W. A.van de Grind & J. J.Koenderink(eds), VNUScience Press, Utrecht

Driven-equilibrium radiofrequency pulses in NMR Magn. Resonanceimaging Med. 1

R Material properties controlling the performance ofamorphous silicon thin film transistors

B A high-level intermediate code

E Formation, thermal stability and physical propertiesof amorphous 3d-based alloys

J. L. C. Daams & J. H. N. van Vucht Contribution to the system Mg-Au-HgE

W. J. van Gils

Mater. Res. Soc.Symp. Proc. 33

Methods and tools forcompiler construction,B. Lorho (ed.), Cam-bridge Univ. Press,Cambridge

Philips J. Res. 39

Philips J. Res. 39

E Some constructions of optimal binary linear unequal Philips J. Res. 39error proteetion codes

C. Ronse B Networks for sorting with fusion Philips J. Res. 39

H. W. A. M. Rompa, M. F. H.Schuurmans & F. Williams (Univ.Delaware, Newark, DE) E

H. M. van Noort, D. B. de Mooij &K. H. J. Buschow E

A. Huijser

Predicted modifications in the direct and indirect Phys. Rev. Lett. 52gaps of tetrahedral semiconductors

Crystal structure, magnetic properties, and 67Fe Phys. Stat. Sol. a 86Mössbauer effect of PtFeSn

502-507 1984

259-273 1984

317-344 1984

255-274 1984

275-292 1984

293-304 1984

305-316 1984

675-678 1984

655-662 1984

G. de With & J. E. D. Parren

E Optical recording Physica 1278 90-94 1984

E Surface stresses in modified BaTi03 ceramics Proc. Br. Ceram. Soc. 99-108 1984NO.34

J. T. Klomp E Ceramic and metal surfaces in ceramic-to-metal Proc. Br. Ceram. Soc. 249-259 1984bonding No. 34

R. Cuppens, C. D. Hartgring, J. F.Verwey & H. L. Peek E

H. J. M. Veendrick, L. C. Pfennings,M. J. J. C. Annegarn, H. A. Harwig,M. J. M. Pelgrom, H. J. F. Peuscher,J. G. Raven, A. Slob & J. W. Slot-boom E

F. P. J. M. Welten, J. Lohstroh &A. J. Linssen (La Radiotechnique-Compelec, Caen) E

G. de With & J. E. D. Parren

An EEPROM for microprocessors and custom logic

A 40MHz 308Kb CCD video memory

A 25/50MHz dual-mode parallel multiplier/accu-mulator

Proc. ISSCC 84, SanFrancisco 1984

Proc. ISSCC 84, SanFrancisco 1984

Proc. ISSCC 84, SanFrancisco 1984

G. deWith

E Fracture of modified BaTi03 ceramics Silicates Ind. 49

E Anisotropy in mechanical properties of, sintered Silicates Ind. 49Sr-hexaferrite

F. Berz, J. Pritchard & A. B. Crow-ley (R. Mil. College Sci., Shriven-ham) R

3 pp. 1984

3 pp. 1984

3 pp. 1984

179-183 1984

185-189 1984

Modelling pin diode switch off with the enthalpy Solid-State Electron. 769-774 1984method 27

100

P. W. J. M. Boumans & J. J. A. M.Vrakking E

P. W. J. M. Boumans & J. J. A. M.Vrakking E

P. W. J. M. Boumans & J. J. A. M.Vrakking E

P. R. Boudewijn, H. W. P. Aker-boom & M. N. C. Kempeners E

SCIENTIFIC PUBLICATIONS

High-resolution spectroscopy using an echelle spectro-meter with predisperser - 1. Characteristics of theinstrument and approach for measuring physicallinewidths in an inductively coupled plasma

High-resolution spectroscopy using an echelle spectro-meter with predisperser - II. Analytical optimiza-tion for inductively coupled plasma atomic emissionspeetrometry

High-resolution spectroscopy using an echelle spectro-meter with predisperser - Ill. A study of line wingsas a major contribution to the background in line-rich spectra emitted by an inductively coupled plasma

Profile distortion in SIMS

Philips Tech. Rev. 42, No. 3

Spectrochim. Acta 398 1239-1260 1984

Spectrochim. Acta398 126-1290 1984

Spectrochim. Acta398 1291-1305 1984

Spectrochim. Acta 398 1567-1571 1984

A. Thayse BP-functions and Boolean matrix factorization (Lect. Springer, BerlinNotes Comput. Sci., Vol. 175)

248 pp 1984

E An lETS study of the reaction of 5-bromo-2-hydroxy- Surf. Sci. 146benzaldehyde withmonomolecular layers of 3-amino-propyltriethoxysilane and 3-(2-aminoethyl)amino-propyltrimethoxysilane immobilized on aluminiumoxide

P. R. Boudewijn, H. W. P. Aker-boom, C. W. T. Bulle-Lieuwma & J.Haisma E

P. N. T. van Velzen

Ion-bombardment-induced changes in the surfacetopography of MBE-grown silicon on gallium phos-phide

P. van Dooren B Reduced order observers: a new algorithm and proof

P. J. Severin & E. A. Dirven E Optische fiber sensor werking, een doel en een mid-del

J. S. Nadan & R. N. Jackson N, R Signal processing for wide screen television: thesmart receiver

P. E. Wierenga, A. G. Dirks & J. J. Ultramicrohardness experiments on vapour-depos-van den Broek E ited films of pure metals and alloys

A. van Oostrom

A. G. Tangena, H. van Wijngaarden& J. G. Fijnvandraat E

G. M. Baudet (Brown Univ., Provi-dence, RI), M. Cutler (AerospaceCorp., LosAngeles), M. Davio, A. M.Peskin (Brookhaven Natl. Lab.,Upton, NY) & F. J. Rammig (Univ.Dortmund) B

E Characterization of semiconductor materials anddevices by surface analysis techniques

A comment on "An energy-based model of frictionand its application to coated systems"

The relationship between HDLs and programminglanguages

Contents of Philips Telecommunication Review 43, No. 2, 1985

Surf. & InterfaceAnal. 7

49-52 1985

319-328 1984

Syst. & Control Lett. 4 243-251 1984

T. Ned. Elektron. & 169-183 1984Radiogenoot. 49

Television image qual- 301-314 1984ity, J. B. Friedman(ed.), SMPTE, Scars-dale, NY

Thin Solid Films 119 375-382 1984

Vacuum 34 881-892 1984

Wear 97 303-306 1984

Workshop Rep. VLSI 64-69 1984& Software Eng. Work-shop, Port Chester,NY,1983

H. van Kampen & R. T. van der Schaaf: Office communication (pp. 75-82)

R. T. van der Schaaf: The SOPHO S family (pp. 83-~1)

M. B. Geelhoed & M. J. Jordaan: SOPHO S25OO,the high-range communication switch (pp. 92-113)

R. J. Mulder & A. C. Zwinkels: SOPHO S250/1OOO,a proven concept (pp. 115-128)

R. J. Mulder: SOPHO Sl Oû, a small but powerful junior (pp. 129-136)

W. J. A. Pasman: SOPHO S data facilities (pp. 137-149)

P. B. Hesdahl: Digital voice/data terminals and data-terminal adapters (pp. 150-163)

H. van Kampen: SOPHO-TEXT, the text server for SOPHO S (pp. 164-171)

C. M. Smits: Telephone management functions (pp. 172-181)

G. G. Maks & R. J. Parsons: Analogue voice terminals (pp. 182-188)

Volume 42,No. 3 Published 29th October 1985pages 69-100

Supplement to Philips Technical ReviewBijlage van Philips Technisch TijdschriftNO.28

September 1985

Recent United States PatentsAbstracts from patents that describe inventions from the following research laboratories,which form part of or cooperate with the Philips group of companies:

Philips GmbH Forschungslaboratorium Aachen, WeiBhausstraJ3e,5100 Aachen, Germany A

Philips Research Laboratory Brussels, 2 avenue Van Becelaere, 1170Brussels,Belgium B

Philips Natuurkundig Laboratorium, Postbus 80000, 5600 JA Eindhoven,The Netherlands E

Philips GmbH Forschungslaboratorium Hamburg, Vogt-Kölln-StraJ3e 30,2000 Hamburg 54, Germany H

Laboratoires d'Electronique et de Physique Appliquée, 3 avenue Descartes,94450 Limeil-Brévannes, France L

Philips Laboratories, N.A.P.C., 345 Scarborough Road, Briarcliff Manor,N.Y. 10510, U.S.A. N

Philips Research Laboratories, Cross Oak Lane, Redhill, Surrey RH1 5HA,England R

Philips Research Laboratories Sunnyvale, P.O. Box 9052, Sunnyvale;CA 94086, U.S.A. S

4504930Charge-coupled deviceH.A. HarwigJ. W. SlotboomM. J. M. PelgromThe invention relates to a charge-coupled SPS memory comprisinga series input register, a parallel section and a series output register.In order to increase the retention time leakage current drain regionsare provided beside the memory. Since the charge collected as aresult of leakage current is largest during the transport through theoutermost registers of the parallel section, only the sides of theparallel section are screened by the said draining regions which pre-ferably consist of dummy registers.

4504936Modular data storage systemJ. W. FaberJ.A. de VosA modular data storage system for a plurality of identical data car-riers includes at least one recorder apparatus and at least one stor-age module. The storage module includes an internal transport de-vice. A'n external transport device is associated with the outside ofthe storage module, which external transport device can transportdata carriers between the recorder apparatus and the storagemodule. The recorder apparatus includes a recorder module com-prising a module frame having outer dimensions which are substan-tially equal to those of a module frame of the storage module sothat all of these modules can be arranged successively adjacent toform optional configurations adapted to suit specific requirements.The external transport device cooperates with the recorder and stor-age modules at their backs and is located inside a cabinet of themodular data storage system.

4507619Amplifier with signal-dependent voltage supply sourceJ. Dijkstra EE.RozaIn a combination of an amplifier and a signal-dependent voltagesupply source coupled thereto, a portion of the signal from a signalsource is applied by an amplifier to a comparator, the output signal

E

of which controls a limiter. A portion of the output signal thereofis applied as a feedback signal to a second input of the comparatorby means of a low-pass filter and a feedback network in series. Thefeedback loop thus formed is arranged so that it oscillates at a com-paratively high frequency. A signal-dependent modulation of thepulse width and/or the pulse density then occurs at the limiter out-put. A supply voltage for the amplifier is derived from the output ofthe filter. Owing to the high feedback factor of the feedback loop atthe signal frequencies the supply voltage can adequately follow theoutput signalof the amplifier independently of component toleran-ces and any load variations.

E

4508811Recording element having a pyrylium or thiopyryllum-squarylium dye layer and new pyrylium or thiopyry-lium-squarylium compoundsD. J. Gravesteijn EC. SteenbergenJ. van der VeenW. P. M. NijssenAn optical recording element having a recording layer which com-prises an alkylpyrylium-squarylium dye, as well as novel alkylpyry-lium-squaryliurn compounds.

4509020Push-pull amplifierR. J. van de PlasscheE. C. DijkmansFor a satisfactory cross-over behaviour of the transistors Tl and T2of a push-pull amplifier, it is necessary that the sum of the base-emitter voltages of the transistors Tl and T2 .remains substantiallyconstant. For this purpose a first voltage-current converter iscoupled between the base and the emitter of transistor Tl, the inver-ting input of this converter being coupled to the base of transistorTl via a first reference-voltage source and the non-inverting input tothe emitter of transistor Tl. Similarly, a second voltage-currentconverter and a second reference-voltage source are arranged be-tween the base and the emitter oftransistor T2. The output currentsof the first and the second voltage-current converters are comparedwith each other in a combining circuit which drives a control ampli-

E

PHILIPS

fier, which in its turn controls the base-emitter voltage of transistorT2 in such a way that the sum of the base-emitter voltages of tran-sistors Tl and T2 remains constant. The push-pull amplifier exhibitsa minimal amount of second-harmonic distortion, because only thedifference of the errors introduced by the first and the second volt-age-current converters is of importance in this respect and theseerrors are substantially equal to each other due to the method offabrication of the voltage-current converters.

4509021Circuit for amplifying and/or attenuating a signalA. J. P. M. van Uden EIn a circuit for amplifying and/or attenuating a signal, which cir-cuits comprises an amplifier stage having an inverting input and anon-inverting input and an output, a first voltage divider having ntaps is arranged between an input terminal and ultimately a point ofconstant potential. The taps are connected to a first controllableswitching unit for switching individualones of the taps to the non-inverting input of the amplifier stage. Further, a second voltagedivider having m taps is connected between an output terminal,which is the output of the amplifier stage, and (ultimately) the pointof constant potential. The taps are connected to a second con-trollable switching unit for switching individualones of the taps tothe inverting input of the amplifier stage.

4509205Radio receiver comprising a frequency-locked loopwith audio frequency feedback, and a muting circuitW. G. Kasperkovitz ERadio-receiver having a frequency-locked loop which comprises,arranged one after the other, a tunable voltage-controlled oscilla-tor, a mixer stage, a filtering element, as well as a frequency-voltageconverter which is connected to the tunable voltage-controlled oscil-lator, several stable tunings being possible for each transmitter. Anunambiguous selection ismade from these stable tunings of a tuningto a desired station by muting the radio receiver in the other stabletuning frequency ranges. To this end the radio receiver according tothe invention comprises a muting circuit as well as a control circuitfor the muting circuit in which use is made of a further frequency-voltage converter comprising an allpass frequency-dependent 1800

phase shifting network, a phase detector and a limiter. By adjustingthe muting circuit of the radio receiver to the quiescient conditiononly in the desired tuning range and by activating it outside thisrange, a suppression of interstation noise is obtained at the sametime.

4510471Acoustic surface wave device including a reflectivemultistrip coupler1.FUn RR. MurrayAn acoustic surface wave device comprising a piezoelectric sub-strate provided with an input transducer an output transducer and acompact reflective multistrip coupler in which the strips of each pairof comparatively widely spaced strips in one array of mutuallyparallel strips are connected to respective strips of a pair of com-paratively narrowly spaced strips in the other array. The wide andnarrow spacings alternate along each array. The improvement pro-vides further electrodes located in the relatively wide spaces in eacharray and electrically interconnected so as to effect a transfer of sig-nal currents along each array and from one array to the other. Thedevice improves efficiency, reduces insertion loss and improves theshape of the pass band.

I

4510520Television transmission systemD. W. ParkerL. J. 'van de PolderA high definition colour television transmission system in whichwide band luminance information is divided to produce a first lumi-nance information portion (below 3.8 MHz) and a second lumi-

nance information portion (between 3.8 and 8.8 MHz). The firstluminance information portion is transmitted by way of a first trans-mission path together with a colour subcarrier modulated by colourinformation. The bandwidth of the colour information is such thatthe modulation components lie outside the bandwidth of the firstluminance information portion. The second and higher frequencyluminance information portion is frequency shifted to lie within 0and 5 MHz and is transmitted by way of a second transmissionpath. The two transmissions mayalso carry sound signals. The twotransmissions may be jointly received for the provision of a high de-finition display but the invention has the distinct advantage that thesignalof the first transmission path may be received alone by cur-rently manufactured television receivers for reproduetion of a dis-play of a quality which is currently acceptable.

4510528Smear reduction for a television pick-upF. H. M. Bergen EA television pick-up arrangement comprises a solid state pick-updevice which is read by field transfer. The pick-up produces a pic-ture signal which on display shows information smear. It is knownto perform a correction by obtaining signal smear information fromunder an opaque, masking strip on the picture pick-up portion ofthe pick-up device. The smear picture signal is stored once in everyfield period in a memory device which is repeatedly readable at linefrequency. The smear picture signal is subtracted from the picturesignal to be corrected. If there is movement in the scene in the hori-zontal direction, an unacceptable smear is, however, still observ-able. According to the invention, a correction may be effected,using a second memory device. The second memory device may beused for performing a signal averaging operation over several fieldperiods or for storing a smear picture signal so that different smearcorrections can be effected in the upper and lower picture halves. Inboth cases the smear is reduced to an acceptable level.

4510612System comprising mutually synchronizing first andsecond active functional unitsC. S. Scholten EA. SlobP. G. JansenA synchronization system for two active functional units which areinterconnected by means of a synchronization connection. In orderto obtain reliable synchronization without using an untoward num-ber of wires in the connection, at least one of the two functionalunits has a generator which is capable of generating three successivediscrete signallevels on a single connection wire of the synchroniza-tion connection. This function unit also has a detector for detectinga signal transition which is produced by a generator in the otheractive functional unit and for generating an activation signal inreaction thereto. In response to this activation signal, the generatorof the same functional unit produces a succession of two signaltransitions from one' prevailing extreme signal level to the otherextreme signal level of the three successive signal levels. The otherfunctional unit has a detector which produces, in response to thesecond signal transition of the succession, an activation signal forthe generator of the other functional unit in order to produce a syn-chronized signal transition or transitions. Such successions of sig-nal transitions can be transmitted in one or both directions; the syn-chronization connection may even consist of a single connectionwire on which a total of five different signals levels may then occur.

RE

4511 601Copper metallization for dielectric materialsJ.R.Akse NS.A.LongA method of providing a copper metallization on a dielectric orsemiconductive body, and a dielectric or semiconductive bodyhaving a metallization consisting essentially of copper. Accordingto the method, a mixture of copper oxide powder and 0 to 15weightpercent reduction-resistant glass frit is dispersed in an organicvehicle and a solvent to produce a paste. The paste is applied to thebody to provide a coating thereon. The coating is dried to remove

the solvent, and then the coated body is fired in an oxidizing atmos-phere at a temperature below the melting temperature of the glassfrit to remove the organic vehicle. Finally, the coated body is fired asecond time in an atmosphere which is reducing to the copper oxidebut substantially nonreducing to the glass frit. The second firing isat a temperature from 700 to 1050°C for from 120 to 15 minutes toconvert the copper oxide to copper metal. The metaIIized body maycomprise a reduction-resistant dielectric body and a metaIIizationconsisting essentially of copper, with no glass frit.

4511 855Compensation for differences in gain among ampli-fiersK. R. Wittig NA method and apparatus for measuring the small signal gain of anamplifier. Each amplifier responds, in operation, to a small signalinput. The small signal input varies substantially linearly over afield time interval. A small bias signal is added to the bias level in-put of the amplifier during a portion of the field time interval. Theoutput of the amplifier is measured at three or more different timesduring the field time interval, at least one time when the small biassignal has not been added to the bias level input and at least onetime being when the small bias signal has been added to the biaslevel input. By processing the measured outputs, a signal which isdirectly proportional to the product of the small bias signal with thesmall signal gain of the amplifier is obtained. A method and appa-ratus for actively compensating for differences in the small signalgain among two or more amplifiers is obtained by dividing the smallsignal output of each amplifier by the signal proportional to thesmall signal gain of the amplifier.

4511933Method and apparatus for recording a digital infor-mation signalM. H. H. Höfelt EE. de NietA. M. A. RijckaertFor a controlled positioning of a read element relative to a track tobe read during the read-out of such a record carrier, Iow-frequencytracking signals may be recorded in the tracks. Instead of addingseparate tracking signals to the information signal, these trackingsignals are represented by the d.c, content of the information signal.During conversion of the information words of the informationsignal into channel words steps are therefore taken to provide twochannel words whose d.c. content is equal but opposite for each in-formation word. The choice between these two channel words ismade depending on a control signal which depends on the desiredtracking signal.

4512020Data processing device for processing multiple-symboldata-words based on a symbol-correcting code andhaving multiple operating modesT.Krol EB. J. VonkA computer system based on a symbol-correcting code. The codewords consist of a number of code symbols. In the normal operatingmode of the error correction members, correction is possible of allerrors which are either limited to one code symbol or which concernonly two arbitrarily situated code bits. During operation in theerasure mode, a predetermined code symbol within the code word isnot taken into account; therefore, it may contain an arbitrary, un-known error. An error can be corrected which concerns only onearbitrarily situated code bit. In the selection mode, two predeter-mined code symbols within the code word are not taken into ac-count. The data words can be reconstructed from the others. Themode is controlled by the content of the mode register. The moderegister is controlled by the output signals of syndrome generators.

4513262Acoustic surface wave deviceJ. SchofieldR. F. Mi/som

R

An acoustic surface wave device using an interdigital electrodearray to launch and receive surface waves overcomes problems ofdiffraction by making the arrays approximately 3Àc wide betweenthe outer boundaries of the bus bars. As a result, the arrays caneach only propagate and transduce a single acoustic surface wave-guide mode which is symmetrical about the axis of the array.

4513 263Bandpass filtersB. J.Minnis RThe specification describes four classes of microwave bandpassfilters formed in triple plate stripline with portions of line having acommensurate length equal to a quarter-wavelength at the center ofthe stopband, enabling the widths of the pass and stop bands to bespecified independently; lumped capacitors are also used to assist inproviding elements with high series capacitance. The four classestogether cover a wide range of electrical specifications, and enablewide pass and stop bands and high selectively to be obtained. Eachclass corresponds to a bandpass S-plane prototype network con-figuration derived using exact synthesis procedures from a specifica-tion of transmission zero' locations. The filters can be manufacturedusing photolithographic technology to have consistently accurateperformance.

4513 272Devices for manufacturing a magnetic quadrupolepost-focusing maskJ. Verweel EH. ZijlstraIn a first method, a meander-shaped coil is provided on each side ofa mask having apertures. The coils are oriented perpendicularlyrelative to each other. If a current flows through the coils in the cor-rect direction, a magnetic quadrupole is formed along the circum-ference of each aperture. In a second method, a holder with per-manent magnetic strips is provided on each side of the mask. Thestrips in the one holder are oriented perpendicularly relative to thestrips in the other holder. A coil is provided around the holderswith which a decaying magnetic alternating field is generated whichinitially drives the material' of the mask on both sides of the hys-teresis curve into saturation. After the decay of said alternatingfield a magnetic quadrupole is present along the circumference ofeach aperture.

4513 315Community antenna television arrangement for thereception and distribution of TV signals and digitalaudio signalsC. B. Dekker EL. B. VriesCommunity antenna television arrangement for the reception anddistribution of TV signals and digital audio signals, in particularsignals which are transmitted per satellite, including a head-endconnected to a receiving antenna and a signal distribution network,a time-division multiplex signal which comprises the digital audiosignals in a time-multiplex distribution, being applied to the head-end, which time-division multiplex signal is modulated on a soundcarrier, the bit rate of the digital audio signals to be distributedbeing reduced in the head-end of the community antenna televisionarrangement by a TDM/FDM conversion in order to reduce signalechoes.

4513388Electronic device for the execution of a mathematicaloperation on sets of three digital variablesH. J. M. Veendrick EL. C. M. G. PfenningsJ. G. RavenA. H. H. J. NillesenA device is described for electronically executing a mathematicaloperation, being Z = KA + (1 - K)B. It is also described how thisdevice or how several of such devices can be used for the design of anumber of realizations, such as a recursive filter, a digital mixer etc.The basic idea is the electronic implementation of a mathematicalfunction for binary variables.

4513 433Fluoroscopy apparatus for forming layer images of athree-dimensional objectH. Weiss HR. LindeU. TiemensE. KlotzApparatus for the layer-wise imaging of an object by means of alarge number of radiation sources which are situated in a radiationsource plane and whose radiation beams interseet one anöther sothat the approximately equally large radiation beam cross-sectionswhich are situated in one irradiation plane almost completely coverone another. A record carrier is arranged on the other side of theobject in order to record perspective images produced by the radia-tion beams. An imaging matrix which comprises imaging elementsfor the superposition of the perspective images on an image displaydevice is disposed behind the record carrier. Displacement meansare provided in order to realize a relative movement between theirradiation plane and the object or an object table carrying theobject.

4514251Method of manufacturing a semiconductor device, inwhich patterns are formed in a layer of silicon nitrideby means of ion implantationA. H. van OmmenH. G. R.MaasJ.A.AppelsW. J. M. J. JosquinIn a method of manufacturing a semiconductor device, ions areimplanted into a layer of silicon nitride over a part of its surface,and the layer is then subjected to an etching treatment. Accordingto the present invention, before the etching treatment takes place,but after the ion implantation, the layer is subjected to a heat treat-ment in which the implanted part of the layer obtains a higher resis-tance to etching than the non-implanted part. The heat treatmentoccurs at temperatures above 750 oe. Thus, a negative image of apatterned ion irradiation can be formed in the silicon nitride layer.As a result, the number of cases in which an etching or oxidationmask can be formed in a silicon nitride layer without using addi-tional mask is considerably increased.

4514754Digital colour television signal processing circuitA. H. H. J. Nillesen EP. W. G. WellesIn a phase control loop for the A/O converter of a digital colourtelevision signal processing circuit an input and an output of a delaycircuit are connected to a comparator to derive a control signalfrom the inequality of the signals at said input and output.

4514837Apparatus for optically reading and/or recordinginformationG. E. van Rosmalen EAn apparatus for optically reading and/or recording informationon a rotary optical disc, such as a videodisc player or an opticalrecorder, comprises an objective carrier for an objective which isradially moved to and fro over a frame by a drive unit. Driving iseffected by means of two electric motors which rotate in oppositedirections and which move the objective carrier to and fro and,simultaneously but in an opposite direction, move two balancingmasses in paths on each side of the path of movement of the objec-tive carrier, so that during accelerations and decelerations of the to-and-fro movements of the objective carrier no reactive forces orreaction moments are transmitted to the frame.

4516146Electron sources and equipment having electron sourcesJ. M. Shannon EA. M. E. HoeberechtsG. G. P. van GorkomAn electron source having a rapid response time comprises at leastone n-p-n structure (and possibly an array of said n-p-n structure)formed in a silicon or other semiconductor body by a p-type firstregion between n-type second and third regions. Electrons are gener-ated in the n-p-n structure for emission into free space from a surfacearea of the body affer flowing from the second region through thefirst and third regions. The n-p-n structure has electrode connectionsonly to the n-type second and third regions. The first region providesa barrier region restricting the flow of electrons from the secondregion to the third region until a potential difference is applied be-tween the electrode connections to bias the third region positive withrespect to the second region and to establish a supply of hot electronsinjected into the third region with sufficient energy to overcome thepotential barrier present between the surface area and free space.The barrier region forms depletion layers with both the n-type secondand third regions and is depleted of holes by the merging together ofthese depletion layers at least when the potential difference is appliedto establish said supply of hot electrons. The n-p-n structure can beprovided in a mesa portion of the body at a window in an insulatinglayer so as to form a compact arrangement having very low asso-ciated capacitances. The electron sources may be used in cathode-raytubes, display devices and even electron lithography equipment.

E

4516252Device for imaging layers of a bodyR. LindeE. Klotz

H

In a short-time tomosynthesis apparatus for imaging layers of abody to be examined, the body is irradiated by radiation beamsfrom a large number of different directions in order to form pers-pective images. The radiation beams are detected by a detectorwhich is arranged underneath the body, The perspective imagesappearing on the exit of the detector will overlap when the radiationbeams have a comparatively large angle of aperture. For separatingthe perspective images, an image separating system is provided atthe exit of the detector. The image separating system may comprisea fiber-optical system or an electron-optical deflection unit.

4516261Device for reducing faults in layer images of a th ree-dimensional object formed by means of penetratingradiationG. Harding HE. KlotzA device for reducing artefacts in layer images. Severallayer imagesare formed of an object layer by irradiation from different directions.The same points of these layer images are compared in order to ob-tain a corrected layer image: when corresponding image informationis present in all images points compared, the image information istransferred to corresponding layer image points in a corrected layerimage; when the image information in the compared image points isnot the same, that information is at least partly suppressed.