Research Article Fabrication of a 77GHz Rotman Lens on a High … · 2019. 7. 31. · F : A...

Research ArticleFabrication of a 77 GHz Rotman Lens on a High ResistivitySilicon Wafer Using Lift-Off Process

Ali Attaran and Sazzadur Chowdhury

Department of Electrical and Computer Engineering University of Windsor CEI 401 Sunset Avenue WindsorON Canada N9B 3P4

Correspondence should be addressed to Ali Attaran ali1111uwindsorca

Received 4 December 2013 Accepted 4 April 2014 Published 4 May 2014

Academic Editor Young Joong Yoon

Copyright copy 2014 A Attaran and S ChowdhuryThis is an open access article distributed under theCreativeCommonsAttributionLicense which permits unrestricted use distribution and reproduction in anymedium provided the originalwork is properly cited

Fabrication of a high resistivity silicon based microstrip Rotman lens using a lift-off process has been presented The lens features3 beam ports 5 array ports 16 dummy ports and beam steering angles of plusmn10 degrees The lens was fabricated on a 200120583m thickhigh resistivity silicon wafer and has a footprint area of 197mmtimes 156mmThe lens was tested as an integral part of a 77GHz radarwhere a tunable X band source along with an 8 times multiplier was used as the RF source and the resulting millimeter wave signalcentered at 77GHz was radiated through a lens-antenna combination A horn antenna with a downconverter harmonic mixer wasused to receive the radiated signal and display the received signal in an Advantest R3271A spectrum analyzer The superimposedtransmit and receive signal in the spectrum analyzer showed the proper radar operation confirming the Rotman lens design

1 Introduction

Frequency-Modulated Continuous-Wave (FMCW) radarsensors based on analog or digital beamforming technologyhave been proven to be effective for avoidance of automotivecollision or mitigation of collision damage Investigationshows that instead of using microelectronic based analog ordigital beamforming engines if Rotman lens type passivebeamformers can be used the signal processing time as wellas power consumption size integration complexity and sys-tem building cost will be reduced significantly [1ndash5] Severaldesign approaches are available that propose using Rotmanlens for automotive radar applications However almost allof them depend on conventional fabrication technique [2]with expensive mass production Recent investigation showsthat high resistivity silicon is an attractive choice to realizehigh performance millimeter wave MMICs due to lowerattenuation and lower cost [4ndash6] Additionally the siliconprocessing technology is highly matured and the lenses canbe batch fabricated like conventional ICs to lower the costof automotive radars In [4] a 60GHz silicon based Rotmanlens has been presented In this context this paper presentsthe design procedure of the Rotman lens at mm-frequency

range followed by 3D full wave simulation results Proposedfabricated and tested silicon based Rotman lens is a potentialto use in a low cost 77GHz automotive radar

2 Rotman Lens Design

Rotman lens is a passive beamforming device and the beamcan be steered in a specified direction by selective excitationof any one of the input ports (beamports) of the lens as shownin Figure 1 Detailed design principles of Rotman lens are wellestablished [4 7 8] and can be summarized in 3 major steps

Step 1 The lens geometrical specifications such as positionsof the beam ports on the focal arc and the array ports onthe inner contour are determined from the theory of rayoptics Detailed procedure to determine the initial geometryfollowing the ray optics theory is available in [7]

Step 2 Once the initial geometry is determined the two-dimensional aperture theory of classical antenna analysis isapplied to determine electrical characteristics of the lens suchas the return loss insertion loss port isolation and powerefficiency

Hindawi Publishing CorporationInternational Journal of Antennas and PropagationVolume 2014 Article ID 471935 9 pageshttpdxdoiorg1011552014471935

2 International Journal of Antennas and Propagation

Beam port

Array contourRadiating element

Beam contourTransmission line Outer contour

Wave front

Figure 1 A conceptual diagram of a Rotman lens

Step 3 Finally a 3D field solver like HFSS or XFDTDis used to carry out a full-wave analysis that extends thetwo-dimensional analysis to three-dimensional analysis toincorporate the effects of reflections and scattering fromthe sidewalls (spillover power) and the lens cavity surfacewave propagation conductivity of the metal layer conduc-tive attenuation coefficient and the dielectric attenuationcoefficient [6] The initial specifications determined in Step 1are more or less modified at this step to obtain optimumperformance

Following this procedure for the target automotive appli-cation the lens center frequency has been selected as 77GHzfor use in an automotive radar For use in a trimode (short-range mid-range and long-range) FMCW radar the lensbandwidth has been selected to be 8GHz to match themaximum chirp bandwidth of the FMCW signal [5]

Following [9] a scan angle of 10∘ has been selected tocover an azimuthal angle of 30∘ at minus6 dB of antenna radiationpattern The number of array port has been chosen to be5 to minimize the lens width Following [7] the distancebetween array elements is calculated to be 065120582This elementspacing plays an important role in determining beam widthof the microstrip patch antenna arrays This gives a length of52mm for119873max which is the farthest output port in the outercontour Instead of normalizing the parameters with respectto the focal length following themethod presented in [8] theparameters are normalized with respect to the total length ofthe array aperture

119865 = 2119891119873maxsin (120573)

sin (120572)

(1)

where 120573 is the angular direction of the wave front associatedwith the perfect focal point at 120572 with respect to the lens axisTo keep the geometry compact the scan angle 120573 has beenchosen to be the same as the focal angle 120572 Following [8]the ratio of the on-axis and off-axis focal length 119892 has beendetermined from 119892 = 1 cos(120572) as 10154 In [8] it has beenshown that for any value of 119891 as defined in (1) if the focalangle 120572 decreases the value of 119892 approaches towards unityFor a value of 119892 = 1014 and a scan angle 120572 = 10∘ 119891 hasbeen determined to be 05 In [8] it has also been shown that

though the path length error will be at its minimum at 119891 = 1for 15∘ lt 120572 lt 60∘ however for small angles such as 10∘ndash15∘the path length errors can be very small by using a value of 119891very close to 05 By keeping 119891 asymp 05 a very compact lens canbe realized with a little compromise on negligible path lengtherror To fabricate the lens usingmicrofabrication techniquesthe lens geometry needs to be compact to optimize the yieldConsequently a ratio of 053 has been chosen for 119891 thatyielded a value for the off-axis focal length 119865 as 551mmAccordingly the on-axis focal length 119866 has been determinedto be 556mm

A high resistivity silicon substrate has been chosen tofabricate the lens as it has a very low loss tangent and absorp-tion coefficient [7] After initial design specifications were setthe lens geometry was simulated in Rotman Lens Designer(RLD) software from Remcom Inc RLD is a 2D solver thatuses the antenna aperture theory to determine electricallosses lens geometry microstrip tapers and transmissionline lengths The design parameters (geometric dimensionsand losses) were then optimized by using sidelobe levelsphase errors and sidewall curvature as the optimizing costfunctions Following (2) below

119891119905=

150

120587ℎ

radic2

120576119903minus 1

tanminus1120576119903 (2)

where 119891119905is the cutoff frequency and ℎ is the substrate thick-

ness in millimeters it was determined that a 200 micrometerthick silicon substrate with a loss tangent of 00005 anddielectric constant of 118 would yield the optimum resultsconsidering port coupling and sidelobe levels A lower thick-ness substrate also prevents generation of the spurious surfacemodes [10]

Considering the microstrip as a quasi-TEM line theattenuation 120572

119889due to dielectric losses can be determined

from [6]

120572119889=

1198960120576119903(120576119890minus 1) tan 120575

119889

2radic120576119890(120576119903minus 1)

Npm (3)

where tan 120575119889is the loss tangent Therefore a 15 micrometer

thick gold layer with conductivity 452 times 107 Sm was chosen

International Journal of Antennas and Propagation 3

Air (120576r = 1)

Metal (gold)15 120583m

Silicon (120576r = 118)200120583m

(a)

Dummy ports

Dummy ports

4

5

6

7

8

1

2

3

Beam

por

ts

Arr

ay p

orts

Transmission lines

156

mm

197mm

(b)

Figure 2 Rotman lens geometry (a) Cross section (b) Top view

Table 1 Rotman Lens Design Specifications

Parameter ValueGold conductor thickness (120583m) 15Silicon substrate thickness (120583m) 200Silicon permittivity 118Silicon loss tangent 00005Length (mm) times width (mm) 197 times 156

Center frequency 77GHzBandwidth 8GHzBeamwidth 20∘

Focal angle 10∘

Scan angle 10∘

Beam ports 3Array ports 5Dummy ports 16Element spacing 26mmZ system 50 ohmsPort width 016mm

minus50 0 50minus30

minus20

minus10

0

Scan angle (deg)

Arr

ay fa

ctor

(dB)

Active beam 1Active beam 2

Active beam 3

Figure 3 Array factor for three input ports excited separately

to prevent degradation in the performance The attenuation120572119888due to the conductor loss can be determined approximately

from [6]

120572119888=

119877119904

(1198850119882 )

Npmm (4)


4 5 6 7 8Array port number

minus105

minus10

minus95

minus9

Am

plitu

de (d

B)

Beam port 1Beam port 2

Beam port 3

(a)

0

50

100

minus50

minus100

Phas

e (de

g)



Beam port 3

(b)

Figure 4 (a) Beam to array coupling amplitude (b) beam to array coupling phase

2 4 6 8 10 12 14 16minus45

minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus5

Sidewall sample number


Beam port 3

Am

plitu

de (d

B)

Figure 5 Beam to sidewall coupling amplitude

where 119877119904= radic120596120583

02120590 is the surface resistivity for conductor

The loss associated with the connecting transmission lineswas calculated following 119890

minus120572119871 where 120572 = 120572119888+ 120572119889[6] The use

of high resistivity silicon to reduce substrate loss and parasiticcomponents led to conduction between the semiconductorwafer and the metal layers in a microstrip configuration Inorder to prevent current leakage to the substrate a 100 nmthick SiO

2layer was used between the high resistivity silicon

wafer and the gold layers at the top and the bottom A taper-matching technique with a taper width of 1205822 was used toavoid second-order mode Through a simulation study inRLD a total of 16 dummy ports were determined to generatelow level sidelobes [8 10 11] while generating an optimumarray factor

Figure 2(a) shows a cross-sectional view of the designedRotman lens and Figure 2(b) shows the top view of the lensin XFDTD 3D simulation environment The 3D full wave

Array port number

Am

plitu

de (d

B)minus40

minus45

minus50

minus55

minus604 5 6 7 8


Beam port 3

Figure 6 Beam to array spillover coupling amplitude

simulation capabilities of XFDTD software were used tooptimize the final design specifications as listed in Table 1

3 Simulation Results

The array factors of the designed lens for three input portsexcited separately are shown in Figure 3 From Figure 3 it isclear that for each beam port excitation the sidelobe levelsare less than minus12 dB Increasing the contour curvature as wellas number of the dummy ports can reduce the sidelobe levelsLow curvature lens sidewalls have much higher sidelobesA Gaussian optimization method was used to determinethe number of dummy ports to match the determined lenscurvature Phase error for individual beam port through allthe array ports is calculated as low as 05 degreesThis amountof error is very low as compared to 144∘ [4] and 24∘ [12] for77GHz lens

Figure 4(a) shows beam to array coupling amplitude forindividual excitation of different beam ports In worst case


minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B1SA5B1SA6B1

SA7B1SA8B1

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

minus200

minus100

0

100

200

Frequency (Hz)

Phas

e (de

g)

Array port 4Array port 5Array port 6

Array port 7Array port 8

(b)

Figure 7 119878-parameter for beam port number one active (a) magnitude (b) phase

minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B2SA5B2SA6B2

SA7B2SA8B2

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

Frequency (Hz)

minus200

minus100

0

100

200Ph

ase (

deg)



(b)

Figure 8 119878-parameter for beam port number two active (a) magnitude (b) phase

array port number six has the most coupling amplitude withamount of minus945 dB for beam port number one and three andalso minus93 dB for beam port number two Figure 4(b) showsthe amount of beam to array coupling phase

The amount of beam to sidewall coupling amplitude(coupling of energy from the beam ports into the sidewalls)among the beam ports and the 16 dummy ports (8 on each ofthe side walls) is shown in Figure 5 About minus10 dB maximumcoupling occurs for dummy ports that are close to arrayports as they will absorb most of the back scattered radiationFurthermore beam ports to array ports spillover couplingamplitude (coupling of energy from the beam ports to arrayports due to sidewall reflection) is between minus58 dB to minus43 dBas shown in Figure 6

Figure 7(a) shows the 119878-parameter amplitude andFigure 7(b) shows the 119878-parameter phase for beam port 1active for a frequency range of 60GHz to 90 The figuresshow that 119878-parameter amplitudes for beam port numberone along the array ports number 4 5 6 7 and 8 are minus111 dB

minus117 dB minus116 dB minus127 dB and minus109 dB respectively andthe associated phases are 376∘ 4744∘ 8791∘ 13040∘ and16932∘ respectively Beam port number three has almost thesame results

Figures 8(a) and 8(b) show the 119878-parameter valuesamplitude and phase for beam port number two activeAt 77GHz all the array port phases are around 63∘ with avariation of 2∘

Figure 9 shows the 119878-parameters (magnitude and phase)for all the beam ports and array ports superimposed

Figure 10 presents the return losses for individual excita-tion of the beamports 1 to 3 From the figure it is clear that thereturn losses associated with individual beam port excitationare in the vicinity of minus10 dB

Crosstalk is one of the consequential factors when asignal is transmitted on transmission line and producesan undesired effect in adjacent channels or transmissionlines Crosstalk is usually created by unwanted capacitiveinductive or conductive coupling between transmission lines



Active port 1Active port 2

Active port 3

0

minus5

minus10

S-p

aram

eter

(dB)

(a)

0

100

200

minus200

minus100

Phas

e (de

g)



Active port 3

(b)

Figure 9 119878-parameters for beam ports number one two and threeactive (a) magnitude (b) phase

since frequency is very high and size of the wafer is shrunkFigure 11 shows that incidence of crosstalk for all input andoutput lines is less than minus12 dB

Figure 12 shows that the insertion losses of beam ports1 and 3 respectively are 416 dB and for the beam port2 the insertion loss is 329 dB at the center frequency of77GHz It is to be noted here that due to themaximummeshsize limitations associated with XFDTD the 100 nm thicksilicon dioxide layer was not included in XFDTD simulationto determine to insertion loss If the 100 nm thick silicondioxide layer could be included in simulation the insertionloss would be much lower

The lens efficiency can be defined as the ratio of the sumof the output power at all the array ports to the power fed atone of the beam ports [4] and can be determined from the 3Dsimulation results of the following transmission coefficients

Efficiency =119875out119875in

= sum

119899

10038161003816100381610038161198781198991

1003816100381610038161003816

2

= (11987841

2

+ 11987851

2

+ 11987861

2

+ 11987871

2

+ 11987881

2

)

(5)

60 65 70 75 80 85 90Frequency (GHz)

minus20

minus10

minus15

minus5

0

S-p

aram

eter

(dB)

S11

S22

S33

Figure 10 Return loss (dB) over frequency

minus20

minus30

minus10

0

S-p

aram

eter

(dB)

SB2B1SB3B1SA5A4

SA6A5SA7A6

60 65 70 75 80 85 90Frequency (GHz)

Figure 11 Crosstalk (dB) over frequency

Figure 13 shows the lens efficiency as a function offrequency for different beam port excitation The efficiencyof beam ports 1 and 3 is 3457 and for beam port 2 and theefficiency is 468 at the center frequency of 77GHz Thisproves that beam port number 2 which is in the center of thebeam ports can transmit higher power to the array ports

Figure 14 shows the E-field pattern of current propagatingthrough the lens with beam port 2 active The wave front willreach at all the array ports and only two closest of the dummyports with similar amplitude levels therefore most of thecurrent will be collected by the array ports Table 2 comparesthe key performance parameters of the designed Rotman lenswith two other similar microwave lenses

4 Lens Fabrication Process

A 200120583m thick high resistivity (gt10000Ω-cm) silicon waferwas RCA cleaned and a 100 nm thick layer of low temperatureoxide (LTO) was grown on both sides of the wafer A 15120583mthick layer of gold was deposited by e-beam evaporation ontothe backside of wafer Figure 15(a) LOR 30B lift-off resistwas then spin deposited at 2000 rpm and baked at 170∘Cfor 10min Figure 15(b) Shipley 1827 photoresist was then


Table 2 Comparison of key features

Parameter Reference [13] Reference [14] This paperFrequency (GHz) 24 60 77Material RO3003 (120576

119903

= 3) Silicon and SiO2 (120576119903 = 117 and 39) 120576119903

= 118

Thickness (120583m) 508 300 and 15 200Length timesWidth (mm) 7580 196 times 20 197 times 156

No of beamports 5 5 3No of array ports 7 7 5Phase error (deg) lowast lowast lowast 144 05Insertion loss (dB) 185 2 329Return loss (dB) Almost 10 Almost 10 Almost 10Max Efficiency () 3230 50 468

60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)

Beam port 1 activeBeam port 2 activeBeam port 3 active

Inse

rtio

n lo

ss (d

B)

Figure 12 Insertion loss (dB) over frequency

60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)

Beam port number 1Beam port number 2Beam port number 3

Effici

ency

()

Figure 13 Power efficiency of the proposed Rotman lens for threebeam ports

spin deposited at 2500 rpm and baked 3 minutes at 113∘CFigure 15(c) The LOR and Shipley 1827 layers were thenpatterned with exposing the resist in the Karl Suss MA6aligner with 540mJcm2 and developed for 3 minutes in aMF-319 dish Figure 15(d)

Figure 14 E-Field simulation and current propagating for beamport number two active

Following DI water washing in the spin rinse dryer a15 micrometer thick layer of gold was deposited by e-beamevaporation Figure 15(e) on the topside of the processedwaferThe resist was then dissolved in Remover PG during anovernight soak to complete themetal lift-off step Figure 15(f)The wafer was then bonded to a carrier wafer with a layer ofcrystal bond to provide mechanical support during dicing tosingulate the devices Figure 16 shows the final fabricated lens

5 Testing

The lens was tested in a 77GHz radar setup To enabletesting the lens wasmounted on a special fixture along with a77GHzmicrostrip antenna array fabricated on aDuroid 5880substrate as shown in Figure 17

The complete test setup is shown in Figure 18 A tunableX band source was used as the RF source which was thenconnected to an 8 timesmultiplier unit to output amillimeterwave signal in the vicinity of 77GHz The output of the 8times multiplier was then fed to one of the beam ports of theRotman lens A 77GHz horn antenna was used to pick up theradiation from the antennaThe horn antenna was connectedto an Advantest R3271A Spectrum analyzer (not in the figure)using a downconverter harmonic mixer A screen shot of thespectrum analyzer is shown in Figure 19 where the outputof the harmonic mixer was superimposed on the original


Silicon

15 120583m of gold layer

100nm silicon dioxide

200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)

Figure 15 Fabrication steps of the proposed Rotman lens

Figure 16 Fabricated Rotman lens

Figure 17 Fabricated Rotman lens combined with a microstriparray antenna

transmitted signal at 77GHz The frequency of the receivedsignal clearly indicates that the Rotman lens is functioningproperly as an integral part of the radar unit

Figure 18 Complete test setup

6 Conclusions

The design fabrication and testing of a high resistivitysiliconwafer based 77GHzmicrostripRotman lens have beenpresented Full wave simulation of the lens was carried out


Figure 19 A screen shot of the spectrum analyzer showing both thetransmitted and the received signal to verify the functionality of theRotman lens

using XFDTD software and the simulation results show thatthe lens exhibits very low phase error compact size andexcellent RF performanceThe lens has been fabricated usinga lift-off process using a silicon dioxide layer between the topmetal layers the silicon substrate to reduce substrate loss

A functional test of the lens has been carried out usinga 77GHz radar setup and the test results show that thelens is working fine as designed Further testing of the lensis in progress The relatively less expensive microfabricatedRotman lens can be used to realize a high performance lowcost MEMS based radar sensor for automotive applicationsby replacing expensive analog andor digital beamformers ascurrently being used in commercially available automotiveradar sensors The developed Rotman lens can easily beintegrated with the Radar front end circuitry and siliconbased microstrip antenna array to pave the way to realize lowcost high performance radar on a chip

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgments

This research has been supported by the Natural Sciences andEngineering Research Council of Canada (NSERC)rsquos Discov-ery Grant no RGPIN 293218 Ontario Centres of Excellence(OCE)rsquos Grant no IC50659 and Auto21 NCE Canadarsquos Grantno F503-FSS The authors would like to greatly acknowledgethe packaging and testing support provided by Advotech Incof Tempe Arizona The authors also would like to acknowl-edge the supports provided by Canadian MicroelectronicsCorporation (CMCMicrosystems)

References

[1] M Schneider ldquoAutomotive radarmdashstatus and trendsrdquo in Pro-ceedings of the German Microwave Conference (GeMiC rsquo05) pp144ndash147 2005

[2] J Schoebel and P Herrero ldquoPlanar antenna technology formm-wave automotive radar sensing and communicationsrdquo inRadar Technology pp 297ndash318 Intech Rijeka Croatia 2009

[3] R Schneider H-L Blocher and K M Strohm ldquoKOKONmdashautomotive high frequency technology at 7779 GHzrdquo in Pro-ceedings of the 4th European Radar Conference pp 247ndash250Munich Germany October 2007

[4] W Lee J Kim C S Cho and Y J Yoon ldquoBeamforming lensantenna on a high resistivity silicon wafer for 60 GHz WPANrdquoIEEE Transactions on Antennas and Propagation vol 58 no 3pp 706ndash713 2010

[5] S Lal R Muscedere and S Chowdhury ldquoAn FPGA-basedsignal processing system for a 77 GHz MEMS tri-mode auto-motive radarrdquo in Proceedings of the 22nd IEEE InternationalSymposium on Rapid System Prototyping Shortening the Pathfrom Specification to Prototype (RSP rsquo11) pp 2ndash8 May 2011

[6] D M Pozar Microwave Engineering John Wiley amp Sons NewYork NY USA 2nd edition 1998

[7] W Rotman and R F Turner ldquoWide-angle microwave lens forline source applicationsrdquo IEEE Transactions on Antennas andPropagation vol 11 no 6 pp 623ndash632 1963

[8] L T Hall Broadband monolithic constrained lens design [PhDthesis] Department of Electrical EngineeringTheUniversity ofAdelaide Adelaide Australia 2000

[9] D Freundt and B Lucas ldquoLong range radar sensor for high-volume driver assistance systems marketrdquo in Proceedings of theSAE World Congress amp Exhibition pp 117ndash124 2008

[10] K C Gupta R Garg I Bahl and P Bhartia Microstrip Linesand Slotlines Artech House Norwood Mass USA 1996

[11] J Kim and F Barnes ldquoScaling and focusing of the Rotman lensrdquoin Proceedings of the IEEE Antennas and Propagation SocietyInternational Symposium pp 773ndash776 Boston Mass USA July2001

[12] I S Song J Kim D Y Jung et al ldquo60GHz rotman lens andnew compact low loss delay line using LTCC technologyrdquo inProceedings of the IEEE Radio and Wireless Symposium (RWSrsquo09) pp 663ndash666 January 2009

[13] W Lee J Kim and Y J Yoon ldquoCompact two-layer rotman lens-fed microstrip antenna array at 24 GHzrdquo IEEE Transactions onAntennas and Propagation vol 59 no 2 pp 460ndash466 2011

[14] W Lee J Kim C S Cho and Y J Yoon ldquoA 60 GHz rotman lenson a silicon wafer for system-on-a-chip and system-in-packageapplicationsrdquo in Proceedings of the IEEE MTT-S InternationalMicrowave Symposium (IMS rsquo09) pp 1189ndash1192 June 2009

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



Beam port

Array contourRadiating element

Beam contourTransmission line Outer contour

Wave front

Figure 1 A conceptual diagram of a Rotman lens

Step 3 Finally a 3D field solver like HFSS or XFDTDis used to carry out a full-wave analysis that extends thetwo-dimensional analysis to three-dimensional analysis toincorporate the effects of reflections and scattering fromthe sidewalls (spillover power) and the lens cavity surfacewave propagation conductivity of the metal layer conduc-tive attenuation coefficient and the dielectric attenuationcoefficient [6] The initial specifications determined in Step 1are more or less modified at this step to obtain optimumperformance

Following this procedure for the target automotive appli-cation the lens center frequency has been selected as 77GHzfor use in an automotive radar For use in a trimode (short-range mid-range and long-range) FMCW radar the lensbandwidth has been selected to be 8GHz to match themaximum chirp bandwidth of the FMCW signal [5]

Following [9] a scan angle of 10∘ has been selected tocover an azimuthal angle of 30∘ at minus6 dB of antenna radiationpattern The number of array port has been chosen to be5 to minimize the lens width Following [7] the distancebetween array elements is calculated to be 065120582This elementspacing plays an important role in determining beam widthof the microstrip patch antenna arrays This gives a length of52mm for119873max which is the farthest output port in the outercontour Instead of normalizing the parameters with respectto the focal length following themethod presented in [8] theparameters are normalized with respect to the total length ofthe array aperture

119865 = 2119891119873maxsin (120573)

sin (120572)

(1)

where 120573 is the angular direction of the wave front associatedwith the perfect focal point at 120572 with respect to the lens axisTo keep the geometry compact the scan angle 120573 has beenchosen to be the same as the focal angle 120572 Following [8]the ratio of the on-axis and off-axis focal length 119892 has beendetermined from 119892 = 1 cos(120572) as 10154 In [8] it has beenshown that for any value of 119891 as defined in (1) if the focalangle 120572 decreases the value of 119892 approaches towards unityFor a value of 119892 = 1014 and a scan angle 120572 = 10∘ 119891 hasbeen determined to be 05 In [8] it has also been shown that

though the path length error will be at its minimum at 119891 = 1for 15∘ lt 120572 lt 60∘ however for small angles such as 10∘ndash15∘the path length errors can be very small by using a value of 119891very close to 05 By keeping 119891 asymp 05 a very compact lens canbe realized with a little compromise on negligible path lengtherror To fabricate the lens usingmicrofabrication techniquesthe lens geometry needs to be compact to optimize the yieldConsequently a ratio of 053 has been chosen for 119891 thatyielded a value for the off-axis focal length 119865 as 551mmAccordingly the on-axis focal length 119866 has been determinedto be 556mm

A high resistivity silicon substrate has been chosen tofabricate the lens as it has a very low loss tangent and absorp-tion coefficient [7] After initial design specifications were setthe lens geometry was simulated in Rotman Lens Designer(RLD) software from Remcom Inc RLD is a 2D solver thatuses the antenna aperture theory to determine electricallosses lens geometry microstrip tapers and transmissionline lengths The design parameters (geometric dimensionsand losses) were then optimized by using sidelobe levelsphase errors and sidewall curvature as the optimizing costfunctions Following (2) below

119891119905=

150

120587ℎ

radic2

120576119903minus 1

tanminus1120576119903 (2)

where 119891119905is the cutoff frequency and ℎ is the substrate thick-

ness in millimeters it was determined that a 200 micrometerthick silicon substrate with a loss tangent of 00005 anddielectric constant of 118 would yield the optimum resultsconsidering port coupling and sidelobe levels A lower thick-ness substrate also prevents generation of the spurious surfacemodes [10]

Considering the microstrip as a quasi-TEM line theattenuation 120572

119889due to dielectric losses can be determined

from [6]

120572119889=

1198960120576119903(120576119890minus 1) tan 120575

119889

2radic120576119890(120576119903minus 1)

Npm (3)

where tan 120575119889is the loss tangent Therefore a 15 micrometer

thick gold layer with conductivity 452 times 107 Sm was chosen


Air (120576r = 1)


Silicon (120576r = 118)200120583m

(a)

Dummy ports

Dummy ports

4

5

6

7

8

1

2

3

Beam

por

ts

Arr

ay p

orts

Transmission lines

156

mm

197mm

(b)





Focal angle 10∘

Scan angle 10∘


minus50 0 50minus30

minus20

minus10

0

Scan angle (deg)

Arr

ay fa

ctor

(dB)


Active beam 3



from [6]

120572119888=

119877119904

(1198850119882 )

Npmm (4)



minus105

minus10

minus95

minus9

Am

plitu

de (d

B)


Beam port 3

(a)

0

50

100

minus50

minus100

Phas

e (de

g)



Beam port 3

(b)


2 4 6 8 10 12 14 16minus45

minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus5



Beam port 3

Am

plitu

de (d

B)


where 119877119904= radic120596120583








Array port number

Am

plitu

de (d

B)minus40

minus45

minus50

minus55

minus604 5 6 7 8


Beam port 3







minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B1SA5B1SA6B1

SA7B1SA8B1

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

minus200

minus100

0

100

200

Frequency (Hz)

Phas

e (de

g)



(b)


minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B2SA5B2SA6B2

SA7B2SA8B2

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

Frequency (Hz)

minus200

minus100

0

100

200Ph

ase (

deg)



(b)













Active port 3

0

minus5

minus10

S-p

aram

eter

(dB)

(a)

0

100

200

minus200

minus100

Phas

e (de

g)



Active port 3

(b)






= sum

119899

10038161003816100381610038161198781198991

1003816100381610038161003816

2

= (11987841

2

+ 11987851

2

+ 11987861

2

+ 11987871

2

+ 11987881

2

)

(5)

60 65 70 75 80 85 90Frequency (GHz)

minus20

minus10

minus15

minus5

0

S-p

aram

eter

(dB)

S11

S22

S33


minus20

minus30

minus10

0

S-p

aram

eter

(dB)

SB2B1SB3B1SA5A4

SA6A5SA7A6

60 65 70 75 80 85 90Frequency (GHz)









119903


= 118



60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)


Inse

rtio

n lo

ss (d

B)


60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)


Effici

ency

()





5 Testing




Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Air (120576r = 1)


Silicon (120576r = 118)200120583m

(a)

Dummy ports

Dummy ports

4

5

6

7

8

1

2

3

Beam

por

ts

Arr

ay p

orts

Transmission lines

156

mm

197mm

(b)





Focal angle 10∘

Scan angle 10∘


minus50 0 50minus30

minus20

minus10

0

Scan angle (deg)

Arr

ay fa

ctor

(dB)


Active beam 3



from [6]

120572119888=

119877119904

(1198850119882 )

Npmm (4)



minus105

minus10

minus95

minus9

Am

plitu

de (d

B)


Beam port 3

(a)

0

50

100

minus50

minus100

Phas

e (de

g)



Beam port 3

(b)


2 4 6 8 10 12 14 16minus45

minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus5



Beam port 3

Am

plitu

de (d

B)


where 119877119904= radic120596120583








Array port number

Am

plitu

de (d

B)minus40

minus45

minus50

minus55

minus604 5 6 7 8


Beam port 3







minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B1SA5B1SA6B1

SA7B1SA8B1

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

minus200

minus100

0

100

200

Frequency (Hz)

Phas

e (de

g)



(b)


minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B2SA5B2SA6B2

SA7B2SA8B2

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

Frequency (Hz)

minus200

minus100

0

100

200Ph

ase (

deg)



(b)













Active port 3

0

minus5

minus10

S-p

aram

eter

(dB)

(a)

0

100

200

minus200

minus100

Phas

e (de

g)



Active port 3

(b)






= sum

119899

10038161003816100381610038161198781198991

1003816100381610038161003816

2

= (11987841

2

+ 11987851

2

+ 11987861

2

+ 11987871

2

+ 11987881

2

)

(5)

60 65 70 75 80 85 90Frequency (GHz)

minus20

minus10

minus15

minus5

0

S-p

aram

eter

(dB)

S11

S22

S33


minus20

minus30

minus10

0

S-p

aram

eter

(dB)

SB2B1SB3B1SA5A4

SA6A5SA7A6

60 65 70 75 80 85 90Frequency (GHz)









119903


= 118



60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)


Inse

rtio

n lo

ss (d

B)


60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)


Effici

ency

()





5 Testing




Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











minus105

minus10

minus95

minus9

Am

plitu

de (d

B)


Beam port 3

(a)

0

50

100

minus50

minus100

Phas

e (de

g)



Beam port 3

(b)


2 4 6 8 10 12 14 16minus45

minus40

minus35

minus30

minus25

minus20

minus15

minus10

minus5



Beam port 3

Am

plitu

de (d

B)


where 119877119904= radic120596120583








Array port number

Am

plitu

de (d

B)minus40

minus45

minus50

minus55

minus604 5 6 7 8


Beam port 3







minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B1SA5B1SA6B1

SA7B1SA8B1

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

minus200

minus100

0

100

200

Frequency (Hz)

Phas

e (de

g)



(b)


minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B2SA5B2SA6B2

SA7B2SA8B2

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

Frequency (Hz)

minus200

minus100

0

100

200Ph

ase (

deg)



(b)













Active port 3

0

minus5

minus10

S-p

aram

eter

(dB)

(a)

0

100

200

minus200

minus100

Phas

e (de

g)



Active port 3

(b)






= sum

119899

10038161003816100381610038161198781198991

1003816100381610038161003816

2

= (11987841

2

+ 11987851

2

+ 11987861

2

+ 11987871

2

+ 11987881

2

)

(5)

60 65 70 75 80 85 90Frequency (GHz)

minus20

minus10

minus15

minus5

0

S-p

aram

eter

(dB)

S11

S22

S33


minus20

minus30

minus10

0

S-p

aram

eter

(dB)

SB2B1SB3B1SA5A4

SA6A5SA7A6

60 65 70 75 80 85 90Frequency (GHz)









119903


= 118



60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)


Inse

rtio

n lo

ss (d

B)


60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)


Effici

ency

()





5 Testing




Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B1SA5B1SA6B1

SA7B1SA8B1

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

minus200

minus100

0

100

200

Frequency (Hz)

Phas

e (de

g)



(b)


minus15

minus10

minus5

0

S-p

aram

eter

(dB)

SA4B2SA5B2SA6B2

SA7B2SA8B2

60 65 70 75 80 85 90

Frequency (GHz)

(a)

6 65 7 75 8 85 9

times1010

Frequency (Hz)

minus200

minus100

0

100

200Ph

ase (

deg)



(b)













Active port 3

0

minus5

minus10

S-p

aram

eter

(dB)

(a)

0

100

200

minus200

minus100

Phas

e (de

g)



Active port 3

(b)






= sum

119899

10038161003816100381610038161198781198991

1003816100381610038161003816

2

= (11987841

2

+ 11987851

2

+ 11987861

2

+ 11987871

2

+ 11987881

2

)

(5)

60 65 70 75 80 85 90Frequency (GHz)

minus20

minus10

minus15

minus5

0

S-p

aram

eter

(dB)

S11

S22

S33


minus20

minus30

minus10

0

S-p

aram

eter

(dB)

SB2B1SB3B1SA5A4

SA6A5SA7A6

60 65 70 75 80 85 90Frequency (GHz)









119903


= 118



60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)


Inse

rtio

n lo

ss (d

B)


60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)


Effici

ency

()





5 Testing




Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












Active port 3

0

minus5

minus10

S-p

aram

eter

(dB)

(a)

0

100

200

minus200

minus100

Phas

e (de

g)



Active port 3

(b)






= sum

119899

10038161003816100381610038161198781198991

1003816100381610038161003816

2

= (11987841

2

+ 11987851

2

+ 11987861

2

+ 11987871

2

+ 11987881

2

)

(5)

60 65 70 75 80 85 90Frequency (GHz)

minus20

minus10

minus15

minus5

0

S-p

aram

eter

(dB)

S11

S22

S33


minus20

minus30

minus10

0

S-p

aram

eter

(dB)

SB2B1SB3B1SA5A4

SA6A5SA7A6

60 65 70 75 80 85 90Frequency (GHz)









119903


= 118



60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)


Inse

rtio

n lo

ss (d

B)


60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)


Effici

ency

()





5 Testing




Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












119903


= 118



60 65 70 75 80 85 900

2

4

6

8

10

Frequency (GHz)


Inse

rtio

n lo

ss (d

B)


60 65 70 75 80 85 900

20

40

60

80

100

Frequency (GHz)


Effici

ency

()





5 Testing




Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Silicon



200120583m

(a)

LOR resist

(b)

Shipley 1827

(c)

LOR patterning

(d)


(e)

Resist removal

(f)






6 Conclusions








Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation















Acknowledgments


References

















RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Research Article Fabrication of a 77GHz Rotman Lens on a High … · 2019. 7. 31. · F : A...

Documents

Transcript of Research Article Fabrication of a 77GHz Rotman Lens on a High … · 2019. 7. 31. · F : A...