DTIC ADA031355: Voltage Locking in Two Coupled Microbridge Josephson Junctions.

AD-A031 355

UNCLASSIFIED

STATE UNI V OF NEW YORK AT STONY BROOK DEPT OF PHYSICS F/G 20/3

voltage locking in two coupled MICROBRIDGE JOSEPhSON junctions. <u>
AU6 76 D w JILLIE. J E LUKENS* Y H KAO N00014-75-C-0769

NL

AD A 081 355

SECURITY CLASSIFICATION OF THIS p*at. (Whmn Df Bnltttd) f /f TMgS- S / /~ x -- -

REPORT DOCUMENTATION PAGE before%£cSq N form

I. REPORT NUMbtR" |2. GOVT ACCESSION NO. > RECIPIENT'S CATALOG NUMBER

Mf^TITLE (tnd Submit)

^OLTAGE LOCKING IN TWO COUPLED MICROBRIDGE
JOSEPHSON ^JUNCTIONS* J j ^

authors;

DGE |

3. TYPE OP REPORT A PERIOD COVERED

Technical Report
02/01/7 6 - 08/31/76

6 PERFORMING ORG. REPORT NUMBER

CONTRACT OR GRANT NUMBERf A)

/Cy D. V.jjtllle, VE.^Lukens M* Y. H./Kao N0jOpll*-75-C-,0769j

19 PERFORMING ORGANIZATION NAME AND ADDRESS

10. PROGRAM ELEMENT. PROJECT. TASK
N AREA A WORK UNIT NUMBERS

NR-319-062

12. REPORT DATE

Dr. James E. Lukens Dept, of Physics / / ✓ j

State University of N.Y. at Stony Brook NR-319-062

Stony Brook. New York 1179**

II. CONTROLLING OFFICE NAME AND ADDRESS "/ \ 12. REPORT D ATE

Office of Naval Research Code ^27 ( / / y Au gvs& &l6 |

800 N. Quincy Street — _TT; HU " ot " --urf

Arlington. Va. 22217 \

14. MONITORING AGENCY NAME A AODRESSCO dllltrtnt In m Contro/Hn< Olllct) IS. SECURITY CLASS, (ot tbit report;

V %^lcd rift. ~

I 4 - Feb — if.i Ai'ijj

~~~~ is. DISTRIBUTION STATEMENT rot MU Report;

©

Unclassified

ISa. DECL ASSIFICATlOtT DOWNORADINO
SCHEDULE 0/}VCV\

— _

Approved for Public Release: Distribution Unlimited

I 17. DISTRIBUTION STATEMENT f©/ //»• abstract entered In Block 20, II dltlaront from Report)

xy

(W

MS. SUPPLEMENTARY NOTES

Presented at the Applied Superconductivity Conference, Stanford, California,
August, 1976.

To be published in the IEEE.

19. KEY WORDS (Continue on rovoroo old* II necoooary and Identity by block n umber)

Josephson Effect.

CSPY AVAILABLE TO DOG DOES

PEWT FULL y 'MW r! < ^"CJ

20. ABSTRACT (Continue on rarer to aide II neceaaary and Identify by block number)

X

Voltage locking, defined as the production of an identical non-zero DC
voltage in the absence of external microwave radiation across each
Josephson Junction of an array, hasjxeen observed in two microbridge
Josephson Junctions separated by aT^^wide strip of superconductor. The
voltage and temperature dependence oiTthe locking is described.

^ rr^ercHw e-fey X

DO ,52",, 1473 EDITION OF I NOV SS IS OBSOLETE

S/N 0102-014-6601 I

6X2U

SECURITY CLASSIFICATION OF THIS PAGE (Wh*t 1>M« Bnttttd)

C r j avcn.lablo to DDC does ro*

pjisut fully legiblo leptoducUoo

VOLTAGE '.OCT;:'’ m -jvo COUPLED Ml CkOS RIDGE JOSEPHSON JUNCTIONS*

ABSTRACT

Yoliagc locking* defined as the prod'** tier of an
identical non-zee > DC voltage in the 1 1- : o of

external microwave reflation across e^ch Joscp
junction of an array, has t?> :i observed in two ■\icro-
brldge JosepltBon junctions separated hv a tp.n wide
strip of superconductor. Voltage locking occuv s when
the bridges are biased with the cm tent f'owin. in
opposing directions thru On* bridges a; , out <i scrlj
connecting them. The vo taj in

be pulled over lpV until each bridge dT i lays *
identical non- zero voltage, with the totel voli .ge
across both bridges equal to zero. lull lockli , as
defined above, is observed in c::ccss of 40uV. The
voltage and tempera ture dependence of the lotui is
described.

A typical mc.isuitunenl circuit is shown
schematically in Fig. lb. Light superconducting leads
01 e provided, and iuc usure.. e nts a;e trade using two four-
luminal circuit - allowin' the midges to be
independently current biased and the vol across
either or both bridges to bu monitored. Ir i« also
possible to apply current betwee* any twi# of the four
current lends mid monitor the voltage acio»s any two
of the four voltage leads Ibis flexibility has
pi oven very nseJul in ’indctstanding the on-;crvcd
interactions. Measurements are made in a fully
shielded devar. Cryogenic lew pass fillet b
e f f ui 1 1 vely proic. L the bridge*- item external noise.

The temperature is measured by a germanium thermo-
mc-.'r, with electronic feedback used to regulate the
temperature to within a few pU.

I. INTRODUCTION

Josephson junctions h?ve excited g cat deal of
interest in recent years lor use ns mictowc.ve t utter,
detectors, ulxers and parametric amplifiers 1 in
particular, scries arrays of thin- film mlcrob ridge
Josephson junctions are being investigated due to
their roggedness and ease of fabrication, thni broad
band capabilities and their superior impedance
matching characteristics. However, lor many applica-
tions it is necessary to use coherent arrays. 2 » *» i, » ®

A coherent array is one in which the frequency mid
phase of the AC Josephson oscillation is the same la
each bridge of the array. Our previous woik with
series arrays of microbridges indicates that such an
array is not normally coherent, however with the

possible to synchronize the bridges to the applied
frequency - 5

la order to understand why series ariayu sre no';
normally coherent we have been investigating the
Interactions between two mlc^obr tdges in detail, k’e
have fabricate 4 two sub micron indium bridges separated
by as little as 1.0pm of superconducting film,
capable of being independently current biased, and
their individual voltages monitored. This arrangement
•has allowed us to study the interactions between two
mi crob ridges both with and without external rad iatioti
and determine if the possibility exists for fabrica-
tion of a coherent array.

II. EXPERIMENTAL TECHNIQUES

figure la 6how3 an indium sample similar to those
used for these experiments. The individual micro-
bridges are always < 1pm square and typically
1000 X thick. The fabrication and gener.il c.harac- •
teristics of our mlcrobrldges has been described in
detail elsewhere^; we would like to reiterate the
fact that electron beam lithography permits us to
fabricate two submicron bridges with various
separations and a reasonably well defined geometry.

The experimental results described here are for
bridges with a Am sepatatlon.

Manuscript received August 17, 19/6

Fig. 1. a) SF.M micrograph of a sample, tilted 45
to the beam, showing typical geometry. The light
areas are 1000 X In film. The bridge separation
here Is 1.2pm. b) A schematic diagram of the
measurement circuit. For AC measurements » O.lpA
AC current ir. added to the DC. current and the
voltage Is monitoied with a lock-in amplifier.

'•'Work supported by the National Science Foundation

and the Office of Naval Research.

T Department of Physics, State University of New York
at Stony Brook, Stony Brook, New York 11794.

Office <f Natal r., \icarc! C ;i:m - .; N0‘. :;4'7SC-P7f ?
A^'ycic ■ [ ; r Hi-: i' asc : D '.ributior Unlimited 1

, I

i'-cprqductun I.; ole or In /’art Is Permitted l : or

Copy available to DDC does not

TV

r III. EXPERT MENTAL RESULTS

These coupled nlcrobridges exhibit * sreat
madber of interactions, aa manifest'd by the dletor-
dm of Chair I-V curvaa. Boma of our Initial
observations have bean deacrlbad elsewhere*, hare v<
wlgh CO foeue upon our observation of voltage locking,
alum Chat ralacae directly to the fabrication of
coha rant arraya. An examla of voltage locking la
ahovn la Fig. 2a. The current chru bridge 2 la find
at a value slightly above lea critical currant and
ewrrant la avept In the oppoelte direction thru
bridge 1 and out the auperconductlng pad aaparatlng
the two brldgea. V; and Vj are eonltored aa I 2 varlea.
Nate that the two voltagea are equal from I) • S3 to
■TpA, with Vj being pulled in lock wi'h Vj over a
range of 1«V, even though the current thru bridge 1
lo held eonetent.

In a practical array both currents would be
awept together. True the luV pulling observed with
I] fixed lap! l»e that the bridges would stsy locked
over a wide voltage range. We have, In fact, observed
continuous locking from 0 to 40uV when both currents
are awapt togathar. Since the opposing currant bias
It essentially a parallel DC connection ■ the
brldgee it would be simple to maintain nearly equal
voltagaa across all bridges with the external bias
circuit, thue Improving the locking range even
further.

To check the oomplateneea of the locking with
greater precision a O.LuA AC signal is added to I 2
only and the voltag s monitored with a lock-ln
amp 11 Tier. The reault la dV./dlj and dVj/dl,, the
•lopaa of the tvo curves In Fig. 2s. In addition,
the total AC voltage (Vt) may be monitored. Thle Is
shown In Pig. 2b as dV^/dlj. During locking
dVj/dlj “ dV^/dtji since trie voltages are In
opposite directions this Implies dV^/dJ^ “ 0. The
daciuaae In dVp/dli In the locking region can be
used aa one measure of the strength of the Inter-
action. Thus dV T /dli ■ 0 indicates complete locking,
and the measurement of A(dVf/dli) provides a gauga
of the strength of tha interaction, even though
complete locking may not occur. Aa s further check
upon tha locking strength we Increased the amplitude
and frequency of the AC modulation current and
monitored dV^/dt, while sweeping through the lock-
ing region. Wc found that dV-p/dT^ » 0 for a modula-
tion corresponding to a voltage rate of change of
3mV/*ac (or a frequency modulation of » 1.5 x 10^
Cllc/aec). Higher modulation rates were not possible
due to circuit limitations.

Voltage locking Is observed both when the current
flows thru the two bridges and out opposite sides of
the Interconnecting pad, or out only one side of the
pad. In the latter case there Is usually s voltage
In the pad, and at ouf flclently high current lavala
a pad volr.ge *pp«*nrs In the former case. The
appearance of a resistance In the pad doe* not pre-
clude observation of voltage locking between tha two
bridges. However, the pad voltage does add
considerably to the complexity of rhe Interactions
observed, and in order to understand the locking
Interaction we have confined most of our recent
observations to situations in which the pad rpmslns
superconducting. Thus the data in Pig*. 2 and 3 wars
taken with no pad voltage.

We have found tha voltage locking to be both
voltage and temperature dependent. Typically full
locking (dVy/dlp * 0) is observed within sow* voltage
range for temperatures ranging from T c to about

>,<**>

rig. 2. An example of voltage lock'ng bstvssn two
microbridgat . a) Ij la fixed and both Vp end Vj are
shown as Ip varies, b) The total differential
resistance of both bridge*. Kota that dVj/dlp goes to
taro in tha lucking rtgicu. *.( JV-./dlp) la a caesura
of tha locking strength.

0.96 T c . Less complete locking interactions ars
observed down to • 0.90 T c , however at this point tha
strength of the Interaction has decreased considerably.
Tha variation of locking strength with voltage at two
dlffArent temperatures la shown In Fig. 3. At
T/T c ■ 0.992 full locking Is observed from 13 to 20wV,
tha peak of tha Interaction strength. At the lower
temperature the peak at 301/V doe* not correspond to
full locking. Note that tha main peak he* moved to a
higher voltage at the lower temperature . Also not*
the oscillation In the locking strength. This
oscillation was particularly pronounced In this sample,
buc has been uhwviveu tu s ieeewt daglee lu ell ,'L..e,
sample* tested. However, with more strongly coupled
samples at an appropriately thooen temperature, con-
tinuous full locking has been observed from 0 to 40yV.
We believe that thl* behavior may be ralatad to a
frequency dependent phase shift In tha locking 1 n ter-
se t Ion . Investigation of this point la continuing.

Tha locking Interaction does persist to higher
voltages than shown hare, but a pad voltage develop*
at tha same time.

A mlcrobrldga at a finite voltage supports a
supercurrent and a normal current, both oscillating at
the Josephaon frequency. The observed behavior of tha
voltage locking Interaction Is coos latent with tha
assumption that pulaaa of the normal currant dlffim
across tha pad aaparatlng tha bridges and Interact with
tha other bridge. In tha oppnmmd current situation tha
normal currant would tend to increase tha instantaneous
eupercurrent In the effected bridge, and could trigger
an Impending phase slip. Cons latent with this Idee la
the fact that locking la never observed with currents
(loving through tha bridges in earlas. This 1* because.
In tha series cate, tha normal current diffusing from
the other bridge delays tba onset of a phase slip by
reducing the eupercurrent, tAich moves to compensate
for tha diffusion of the qusalpartlclaa and maintain'
steady current flow. Thl* model 1* supported by the
feet that stronger voltag* locking la observed when
the pad realstanca la higher, since this result* In a
larger fraction of tha normal ewrrant flowing thru tha
other bridge. W* are presently working on a dynamic
aodal of the Interaction betwmem tha bridges due to
tha normal curraota.

2

»«• 3. Variation of the locking strength with

at fixed ten>er«ture. This figure Illustrates
the oeclllatory nature of the locking with voltage,
and the Movement of the maximum locking strength peak
to higher voltages aa the temperature Is reduced.

IV. CONCLUSIONS

We have observed an Interaction between two
mlcrobrldges that acts to lock the voltages of the
bridges to a single cocoon value. Locking Is
observed only when the DC current flows In opposite
directions tl.ru the two bridges and out the connecting
superconducting pad. Continuous full locking has been
observed In excess of 40uV under approprl.- 'e
conditions . No locking Is observed when current flows
In thn sane direction thru the two bridges (series
biasing); in fact the Interaction tends to force the

wlra«M sneer In *hls cane nccnv« nf nlnrn-

brldgaa can be forcibly synchronized with the
application of sufficient silcrowave power, but. for
close separations (< Sun) the locking Interaction works
against this synchronization; however when current Is
Witt thru the two bridges In opposing directions
(parallel biasing) perfect voltage locking can be
maintained without any external microwave radiation.

It reaalns to be demonstrated that the Josephson
oscillations arc in phase before full coherence may
bn claimed. We are presently working on ascertaining
the exact nature of the locking Interaction.

1* See, for example, P.L. Richards, Semiconducto rs &
Semi metals . Vll , Wlllardsen & Beer, Editors (1974)

2. T.D. Clark, Phys. Rev. B. 8 ( 1973 ) 137

3. T.P. Finnegan & S. Wahlaten, Low Temp. Phy sics
LT 13. Vol. 3(1974) 272.

4. D.W. Palmer & J.E. Mercereau, Appl. Phys
(1974) 467.

3. D.W. Jlllle, J. Lukena 6 Y.H. Kao, IEEE Trans, on
Magnetics, MAC- 11 (1975) 671.

6. D.W. Jlllle, J.E. Lukens, T.H. 1
Physics Letters, 55* (1976) 381