The design of microstrip six-pole quasi-elliptic filter with linear phase response

合集下载

IEEEStdC57.124-1991

IEEEStdC57.124-1991

IEEE Std C57.124-1991Reconized as an American National Standard (ANSI)IEEE Recommended Practice for theDetection of Partial Discharge and the Measurement of Apparent Charge inDry-Type TransformersSponsorTransformers Committeeof theIEEE Power Engineering SocietyApproved June 27, 1991Reaffirmed February 6, 1997Institute of Electrical and Electronics EngineersApproved October 11, 1991Reaffirmed September 19, 1996American National Standards InstituteAbstract: IEEE Std C57.124-1991 covers the detection of partial discharges occurring in the insulation of dry-type transformers of their components and the measurement of the associated apparent charge at the terminals when alternating test voltage is applied. The wideband method is used. The detection system and calibrator characteristics are described, and the test procedure is established.Keywords: Apparent charge, corona, cost coil transformers, dry-type transformers, partial discharge, ventilated dry-type transformersThe Institute of Electrical and Electronics Engineers, Inc.345 East 47th Street, New York, NY 10017-2394, USACopyright © 1992 byThe Institute of Electrical and Electronics Engineers, Inc.All rights reserved. Published 1992Printed in the United States of AmericaISBN 1-55937-159-5No part of this publication may be reproduced in any form, in an electronic retrieval system or otherwise, without the prior written permission of the publisher.IEEE Standards documents are developed within the Technical Committees of the IEEE Societies and the Standards Coordinating Committees of the IEEE Standards Board. Members of the committees serve voluntarily and without compensation. They are not necessarily members of the Institute. The standards developed within IEEE represent a consensus of the broad expertise on the subject within the Institute as well as those activities outside of IEEE that have expressed an interest in participating in the development of the standard.Use of an IEEE Standard is wholly voluntary. The existence of an IEEE Standard does not imply that there are no other ways to produce, test, measure, purchase, market, or provide other goods and services related to the scope of the IEEE Standard. Furthermore, the viewpoint expressed at the time a standard is approved and issued is subject to change brought about through developments in the state of the art and comments received from users of the standard. Every IEEE Standard is subjected to review at least every five years for revision or reaffirmation. When a document is more than five years old and has not been reaffirmed, it is reasonable to conclude that its contents, although still of some value, do not wholly reflect the present state of the art. Users are cautioned to check to determine that they have the latest edition of any IEEE Standard.Comments for revision of IEEE Standards are welcome from any interested party, regardless of membership affiliation with IEEE. Suggestions for changes in documents should be in the form of a proposed change of text, together with appropriate supporting comments.Interpretations: Occasionally questions may arise regarding the meaning of portions of standards as they relate to specific applications. When the need for interpretations is brought to the attention of IEEE, the Institute will initiate action to prepare appropriate responses. Since IEEE Standards represent a consensus of all concerned interests, it is important to ensure that any interpretation has also received the concurrence of a balance of interests. For this reason IEEE and the members of its technical committees are not able to provide an instant response to interpretation requests except in those cases where the matter has previously received formal consideration.Comments on standards and requests for interpretations should be addressed to:Secretary, IEEE Standards Board445 Hoes LaneP.O. Box 1331Piscataway, NJ 08855-1331USAIEEE Standards documents are adopted by the Institute of Electrical and Electronics Engineers without regard to whether their adoption may involve patents on articles, materials, or processes. Such adoption does not assume any liability to any patent owner, nor does it assume any obligation whatever to parties adopting the standards documents.Foreword(This foreword is not a part of IEEE Std C57.124-1991, IEEE Recommended Practice for the Detection of Partial Discharge and the Measurement of Apparent Charge in Dry-Type Transformers.)This recommended practice for measuring partial discharge of dry-type transformers was conceived for the purpose of establishing a standardized method for conducting partial discharge tests of dry-type transformers. The results of the tests may be compared with various transformer designs and manufacturers to establish a partial discharge limit for dry-type transformers.This recommended practice follows the format of IEEE Std 454, IEEE Recommended Practice for the Detection and Measurement of Partial Discharge (Corona) During Dielectric Tests, and IEEE Std C57.113, IEEE Guide for Partial Discharge Measurement in Liquid-Filled Power Transformers and Shunt Reactors. Sections on current detection were purposely omitted, as this technique is normally not used for dry-type transformers.There is no recognized definition of “partial discharge-free"”when referring to partial discharge inception voltage or extinction voltage. An arbitrary sensitivity of 10 pC is suggested until such time as a more definitive standard is established.Various specifications are already written specifying partial discharge-free transformers from 1.1 p.u. operating voltage to 2.0 p.u. operating voltage. It is the intent of this recommended practice to encourage manufacturers of dry-type transformers and users of dry-type transformers to investigate and report the results of factory tests and field experience of partial discharge in dry-type transformers. It is recognized that Paschen's Law applies to the partial discharge intensity and extinction voltage of dry-type transformers. It is conceivable that a dry-type transformer would test partial discharge-free at 1.65 p.u. voltage at room temperature and be barely partial discharge-free at operating temperature for a Class 220C. system. This correlation should be verified with field experience and reported.The guide specifies no particular discharge testing instruments and systems. Several commercially available units are being used. A measuring system of discreet components readily available has been used for measuring partial discharge. Most manufacturers' laboratories have partial discharge-free HV test sets and oscilloscopes. The only additional components required to complete the detection circuit are a partial discharge-free coupling capacitor composed of two 60 kV, .002 mfd capacitors in series, and an inductance composed of a coil of magnet wire. Calibration is accomplished using a calibrated square wave generator and a calibrated coupling capacitor of .0001 mfd.The following two test procedures are proposed:1)To test partial discharge between the coil and ground, normally accomplished during the applied voltage test.2)The test procedure takes place during the induced voltage test to detect partial discharge within a coil. It issuggested that partial discharge measurements be made in both modes. The partial discharge measurement may be made during the normal sequence of tests, while the applied and induced voltage tests are being made.An alternative sequence is to conduct the partial discharge test immediately following the applied voltage test and induced voltage test.The high-voltage bus bars of high-voltage transformers sometimes cause nondestructive partial discharge. This partial discharge in no way affects the reliability of the transformer coils. It may be necessary to disconnect the bus bar from the coils before conducting the partial discharge test on only the coils in order to test for partial discharge in the transformer coils. A note should be added to any test reports stating that the bus bar was removed for the test.At the time this document was submitted to the Standards Board, the Working Group on Recommended Practice for the Detection of Partial Discharges and the Measurement of Apparent Charge in Dry-Type Transformers had the following members:A.D. Kline, (Chairman)B. F. Allen Roy Bancroft D. A. Barnard A. Bimbiris M. Cambre O. R. Compton J. FrankE. Gearhart R. Hayes R. H. HollisterJ. W. HuppA.M. IversenA. J. JonnattiS. P. KennedyE. KoenigM. L. ManningR. A. MarekM. I. MitelmanJ. J. NayW. F. PattersonR. L. ProvostJ. RoddenV. ThenappanR. E. Uptegraff, Jr.G. H. VaillancourtH. J. WindischThe following persons were on the balloting committee that approved this document for submission to the IEEE Standards Board:E. J. Adolphsen L. C. Aicher D. J. AllanB. AllenR. Allustriarti M. S. Altman J. C. Arnold J. AubinR. Bancroft D. Barnard D. L. Basel P. L. Bellaschi S. Bennon W. B. Binder J. V. Bonucchi J. D. Borst C. V. Brown O. R. Compton F. W. Cook J. L. Corkran D. W. Crofts J. N. Davis D. J. Douglas R. F. Dudley J. C. Dutton J. K. Easley J. A. Ebert D. J. Fallon F. L. Foster M. Frydman H. E. Gabel R. E. Gearhart D. W. Gerlach D. A. Gillies R. S. Girgis R. L. GrubbF. J. GryszkiewiczG. HallJ. H. HarlowF. W. HeinrichsW. R. HenningD. R. HightonP. J. HoeflerC. HoeselR. H. HollisterC. C. HoneyE. HowellsC. HurryG. W. IliffY. P. IijimaR. G. JacobsenD.C. JohnsonD. L. JohnsonA. J. JonnattiC. P. KappelerR. B. KaufmanJ. J. KellyW. N. KennedyJ.P. KinneyB. KlaponskiA.D. KlineE. KoenigJ. G. LackeyR. E. LeeH.F. LightS.R. LindgrenL.W. LongL. A. LowdermilkR. I. LoweM. L. ManningH. B. MargolisT. MassoudaJ. W. MatthewsJ. McGillC. J. McMillenW. J. McNuttS. P. MehtaC. K. MillerC. H. MillianR. E. MinkwitzM. MitelmanH. R. MooreW. E. MorehartR. J. MuselW. H. MutschlerE. T. NortonR. A. OlssonB. K. PatelW. F. PattersonH. A. PearceD. PercoL. W. PierceJ. M. PollittC. P. RaymondC. A. RobbinsL. J. SavioW. E. SaxonD. N, SharmaV. ShenoyW. W. SteinL. R. StenslandD. SundinL. A. SwensonD. S. TakachV. ThenappanR. C. Thomas J. A. Thompson T. P. Traub D. E. Truax W. B. Uhl R. E. Uptegraff, Jr.G. H. VaillancourtA. VeitchL. B. WagenaarR. J. WheartyA. L. WilksW. E. WrennA. C. WurdackE. J. YasudaAt the time this recommended practice was published, it was under consideration for approval as an American National Standard. The Accredited Standards Committee on Transformers, Regulators, and Reactors, C57, had the following members at the time this document was sent to letter ballot:Leo J. Savio, ChairJohn A. Gauthier, SecretaryOrganization of Representative Electric Light and Power Group...............................................................................................P.E. OrehekS. M. A. RizviF. StevensJ. SullivanJ. C. ThompsonM.C. Mingoia (Alt.) Institute of Electrical and Electronics Engineers......................................................................J. D. BorstJ. DavisJ. H. HarlowL. SavioH. D. SmithR. A. VeitchNational Electrical Manufacturers Association........................................................................G. D. CoulterP. DeweverJ. D. DouglasA. A. GhafourianK. R. LinsleyR. L. PlasterH. RobinR. E. Uptegraff, Jr.P. J. Hopkinson (Alt.)J. Nay (Alt.) Tennessee Valley Authority.......................................................................................................F. A. Lewis Underwriters Laboratories, Inc.................................................................................................W. T. O'GradyUS Department of Agriculture, REA........................................................................................J. BohlkUS Department of Energy, Western Area Power Administration.............................................D. R. TorgersonUS Department of the Interior, Bureau of Reclamation............................................................F. W. Cook, Sr.US Department of the Navy, Civil Engineering Carps.............................................................H. P. StickleyWhen the IEEE Standards Board approved this standard on June 27, 1991, it had the following membership:Marco W. Migliaro, ChairDonald C. Loughry, Vice ChairAndrew G. Salem, SecretaryDennis BodsonPaul L. BorrillClyde CampJames M. Daly Donald C. Fleckenstein Jay Forster*David F. Franklin Ingrid Fromm Thomas L. HannanDonald N. HeirmanKenneth D. HendrixJohn W. HorchBen C. JohnsonIvor N. KnightJoseph Koepfinger*Irving KolodnyMichael A. LawlerJohn E. May, Jr.Lawrence V. McCAllT. Don Michael*Stig L. NilssonJohn L. RankineRonald H. ReimerGary S. RobinsonTerrance R. Whittemore*Member EmeritusAlso included are the following nonvoting IEEE Standards Board liaisons:Fernando Aldana Satish K. AggarwalJames Beall Richard B. EngelmanStanley Warshaw Deborah A. CzyzIEEE Standards Project EditorCLAUSE PAGE1. Scope (1)2. Purpose (1)3. References (1)4. Definitions (2)5. Partial Discharge Detection System (3)5.1High-Voltage Coupling Circuit (3)5.2Measuring Impedance Unit (Z m) (4)5.3Filter Characteristics (5)5.4Display Unit (5)5.5Discharge Meter (6)5.6Basic Sensitivity (6)5.7Partial Discharge Detector Basic Sensitivity Test (6)6. Calibrator Characteristics (6)6.1Calibrating Capacitor Value (C q) (7)6.2Pulse Generator Rise Time and Decay Time (7)6.3Pulse Generator Amplitude (U o) (7)6.4Pulse Generator Output Impedance (Z o) (7)6.5Calibrator Output Level Adjustment (7)6.6Pulse Generator Frequency (7)7. Tests (7)7.1General Requirements (7)7.2Conditioning (8)7.3Requirements for the Test Voltage (8)7.4Transformer Connections (8)7.5Significance of Various Test Connections (8)7.6Choice of Test Procedure (9)7.7Disturbances (10)8. Bibliography (18)Annex (informative) Partial Discharge Recognition (24)IEEE Recommended Practice for the Detection and the Measurement of Partial Discharge in Dry-Type Transformers1. ScopeThis recommended practice applies to the detection of partial discharges occurring in the insulation of dry-type transformers or their components, and to the measurement of the associated apparent charge at the terminals when an alternating test voltage is applied.2. PurposePartial discharge measurements in dry-type transformers may preferably be made on the basis of measurement of the apparent charge. Relevant measuring systems are classified as narrow-band or wide-band systems. Both systems are recognized and widely used. Without giving preference to one or the other, it is the object of this document to describe the wide-band method. General principles of partial discharge measurements, including the narrow-band method, are covered in IEEE Std 454-19731 [8]2, IEC 270 (1981) [6]3and IEC 76-3 (1980) [5].3. ReferencesThe following publications should be used in conjunction with this document. When the standards referred to in this guide are superseded by a new revision approved by the relevant standards authority, the latest revision should apply.[1] ANSI C68.1-1968, Standard for Measurement of Voltage In Dielectric Tests.41This standard has been withdrawn, however, copies are available from the Institute of Electrical and Electronics Engineers, Inc., Service Center, 445 Hoes Lane, Piscataway, N.J. 08855, U.S.A.2The numbers in brackets refer to those listed in Section 4.3IEC publications are available from IEC Sales Department, Case Postale 131, 3 rue de Varembé, CH 1211, Genève 20, Switzerland/Suisse. IEC publications are also available in the United States from the Sales Department, American National Standards Institute, 11 West 42nd Street, 13th Floor, New York, NY 10036, USA.4ANSI publications are available from the American National Standards Institute, 11 West 42nd Street, 13th Floor, New York, NY 10036.IEEE Std C57.124-1991IEEE RECOMMENDED PRACTICE FOR THE DETECTION OF PARTIAL DISCHARGE [2] ASTM D1868-81 (1990-E01), Method for Detection and Measurement of Partial Discharge (Corona) Pulses in Evaluation of Insulation Systems.5[3] ASTM STP-669, Engineering Dielectrics, Vol. 1, Corona Measurement and Interpretation.[4] CIGRE Working Group 21-03, "Recognition of Discharges," Electra, No. 11, pp. 61-98, Dec, 1969.6[5] IEC 76-3 (1980), Power Transformers, Part 3: Insulation Levels and Dielectric Tests.7[6] IEC 270 (1981), Partial Discharge Measurements.[7] IEEE Std 436-1991, IEEE Guide for Making Corona (Partial Discharge) Measurements of Electronics Transformers (ANSI).8[8] IEEE Std 454-1973, IEEE Recommended Practice for the Detection and Measurement of Partial Discharge (Corona) During Dielectric Tests.[9] IEEE Std C57.113-1991, IEEE Guide for Partial Discharge Measurement in Oil-Filled Power Transformers and Shunt Reactors (ANSI).4. Definitionspartial discharge: A partial discharge within the terms of this document is an electric discharge that only partially bridges the insulation between conductors. The term “corona ” has also been used frequently with this connotation. Such usage is imprecise and is gradually being discontinued in favor of the term “partial discharge.”apparent charge (terminal charge): The apparent charge (q) of a partial discharge is that charge which, if it could be injected instantaneously between the terminals of the test object, would momentarily change the voltage between its terminals by the same amount as the partial discharge itself. The apparent charge should not be confused with the charge transferred across the discharging cavity in the dielectric medium. Apparent charge within the terms of this document is expressed in coulombs, abbreviated C. One pC is equal to 10-12 coulombs.repetition rate(n).: The partial discharge pulse repetition rate (n) is the average number of partial discharge pulses per second measured over a selected period of time.acceptable terminal partial discharge level: The acceptable terminal partial discharge level is that specified maximum terminal partial discharge value for which measured terminal partial discharge values exceeding the said value are considered unacceptable. The method of measurement and the test voltage for a given test object should be specified with the acceptable terminal partial discharge level.voltage related to partial discharges: Voltage within the terms of this document is the phase-to-ground alternating voltage for applied tests (Fig 1) or terminal to terminal alternating voltage for induced voltage tests (Fig 8). Its value is expressed by its peak value divided by the square root of two.partial discharge inception voltage: The lowest voltage at which partial discharges exceeding a specified level are observed under specified conditions when the voltage applied to the test object is gradually increased from a lower value. This voltage is expressed as the peak value divided by the square root of two.5ASTM publications are available from the American Society for Testing and Materials, Customer Service Dept., 916 Race Street, Philadelphia, P.A. 19103, U.S. A.6CIGRE publications are available from the International Conference on Large-Voltage Electric Systems, 112 Boulevard Haussman, F-75008 Paris, France.7IEC publications are available from IEC Sales Department, Case Postale 131, 3 rue de Varembé, CH 1211, Genève 20, Switzerland/Suisse. IEC publications are also available in the United States from the Sales Department, American National Standards Institute, 11 West 42nd Street, 13th Floor, New York, NY 10036, USA.8IEEE publications are available from the Institute of Electrical and Electronics Engineers, Inc., Service Center, 445 Hoes Lane, Piscataway, N.J. 08855, U.S. A.partial discharge extinction voltage: The voltage at which partial discharges exceeding a specified level cease under specified conditions when the voltage is gradually decreased from a value exceeding the inception voltage. This voltage is expressed as the peak value divided by the square root of two.partial discharge-free test voltage: The partial discharge-free test voltage is a specified voltage, applied in accordance with a specified test procedure, at which the test object should not exhibit partial discharges above the acceptable energized background noise level.energized background noise level: The energized background noise level stated in pC is the residual response of the partial discharge measurement system to background noise of any nature after the test circuit has been calibrated and the test object is energized at a maximum of 50% of its nominal operating voltage.acceptable energized background noise level: The acceptable energized background noise level present during test should not exceed 50% of the acceptable terminal discharge level, and in any case, should be below 100 pC (5 pC if an acceptable terminal discharge level of 10 pC is required.)5. Partial Discharge Detection System(Figs 9 and 10 taken from ASTM STP669)The partial discharge detection system comprises the following components:1) a high-voltage coupling circuit (C1, C v)2) a measuring impedance unit (Z m consisting of R m, C2 and L)3)an amplifier and filter circuit4) a display unit5) a discharge meter6) a calibrator (C q, V l)7) a source filter (Z optional)5.1 High-Voltage Coupling CircuitThe purpose of the high-voltage coupling circuit is to allow the connection of the measuring impedance (Z m) to the high-voltage terminal of the transformer under test. In other types of high-voltage equipment, a single, low-capacitance high-voltage capacitor is usually used for this purpose, but in transformers, the equivalent terminal capacitance is usually very low, so a substantial amount of signal is normally produced by partial discharge of only a few pC, and measurement sensitivity is not a problem. At the same time, due to standing waves within the winding, a certain amount of signal cancellation may occur if the bandwidth is not sufficiently wide. Therefore, to insure that the input circuit of the partial discharge detection system does not act as a differentiator and reduce the total bandwidth of the system, it has been found that a high value for the coupling capacitor is necessary to produce a long-time constant of the input circuit. Even then, however, the low equivalent terminal capacitance of the transformer, usually less than 500 pF, will limit this time constant and it may not be possible to make it sufficiently long. Therefore, the use of a single coupling capacitor is not recommended. On the other hand, a satisfactory time constant can always be obtained by using a second capacitor (C2) as part of the measurement impedance, and this is the recommended method.As shown in Figs 9 and 10, high-voltage capacitor C1, and low-voltage capacitor C2, will form a voltage divider. This voltage divider will reduce the sensitivity of the measurement. To make sure that both the sensitivity is still sufficient and the time constant is sufficiently long, the values of C1, C2, and L should respect the conditions below:(1)≥C1100pF(2)where R m is the parallel resistive part of the measuring impedance unit (Z m ). The value of R m should be determined from the particular partial discharge instrument that is used.Example:For f L = 70kHz ,R m = 2.5k Ω and C 1 = 100pFA value of 1000 pF may be chosen for C 2 since 1000 pF ≥ 909 pF and (3)In cases where an RLC measurement impedance is used, the value of L should satisfy the equation below. This will ensure that the measurement bandwidth is unaffected by the presence of L .(4)If the transformer to be tested is fitted with a capacitive bushing tap, then this can be used directly as the high-voltage coupling circuit, and a separate coupling capacitor C 1 is not needed.5.2 Measuring Impedance Unit (Z m )The measuring impedance unit (Z m ) is located physically close to the high-voltage coupling circuit and serves two main purposes:1)It attenuates the test voltage present on the high-voltage coupling circuit to a safe value for measurement of partial discharge signals;2)It matches the amplifier and filter circuit to the high-voltage coupling circuit in insuring a fiat frequencyresponse across the full measurement bandwidth.The measuring impedance unit (Z m )should be configured in such a way as to permit test voltage level monitoring and to observe the phase relationship between the test voltage and the partial discharge pulses; this technique helps to identify the nature of the discharges.As shown in Figs 9 and 10, capacitor C v , whose capacitance value should be chosen to be at least 500 times that of C 1,may be placed in either one of the following two positions:1)In series with inductor L (Fig 9 or 10), or 2)In series with the low-voltage side of high-voltage capacitor C 1 (Fig 10.)C 212llf L R m -------------------≥C 216.28x 70000x 2500,----------------------------------------------909pF ==1000100-----------1015≤=L 25006.28x 70000,------------------------------- 5.7mH ==C 2C 1------15≤L R m 2llf L -----------≥Figure 9 shows the preferred position for C v since the input impedance of the display unit does not shunt the measurement impedance and can be neglected. However, some voltage will reach the input of the amplifier and filter circuit and may cause it to saturate. This voltage can be decreased by increasing the value of C v until it is less than 5V . If saturation occurs, a 20 nF low-voltage capacitor may be placed in series with the input of the amplifier and filter circuit as shown to decouple voltage at the excitation frequency.In the cases where one can not be absolutely certain that saturation of the amplifier will not occur, it is then advised to place C v in the position shown in Fig 10. The impedance of the display circuit will now shunt the measurement impedance and its input capacitance will need to be considered to calculate the value of C 2, as it will add to it.5.2.1 Lower and Upper Cut-Off Frequencies (f L and f H )The lower and upper cut-off frequencies f L and f H , respectively, are the frequencies at which the response to a constant sinesoidal input voltage has fallen by 6 db from the maximum value occurring inside the recommended bandwidth.f L should be located in the range from 70 to 120 kHz to minimize the effect of winding attenuation on partial discharge signals, and at the same time, to provide adequate rejection of SCR-generated noise present in manufacturing plants.An upper limit on f H of 300 kHz is usually necessary to prevent broadcast stations from interfering with the partial discharge measurement.5.3 Filter CharacteristicsThe filter characteristics of the partial discharge detection circuit should be such as to provide attenuation of at least 50db at 25 kHz, of at least 60 db at 15 kHz and below, and of at least 20 db at 500 kHz and above, with respect to the response at the geometric mean frequency (f C ) of the system pass bandwidth that is given by:The filter may be combined with an amplifier to form an active filter. Care should be taken to prevent the saturation of the filter input by the presence of the applied test voltage.5.3.1 Bandwidth ∆fThe bandwidth ∆f is defined as:∆f = f H − f LThe bandwidth should not be less than 100 kHz. A wider bandwidth provides a response whose level is less sensitive to the location of a partial discharge pulse along a transformer winding and is, therefore, more uniform. A bandwidth wider than 100 kHz is preferable, but may lead to background noise problems.5.3.2 LinearityThe instrument circuit, display unit, and discharge meter should be linear within plus or minus 10% of full scale in the range of 50 to 1000 pC.5.4 Display UnitThe display unit should be a cathode ray oscilloscope with a linear, rectangular, or an elliptical time-base. In all cases,the time-base should be synchronized to the test voltage, and at least 98% of a full cycle should be displayed. The phase relationship of the partial discharges to the test voltage should be easy to determine. A suitable graticule should be provided.f c f L f H •().5=。

纳米管制作皮肤感应器 翻译 中英

纳米管制作皮肤感应器 翻译 中英

最后译文:纳米管弹性制作出皮肤般的感应器美国斯坦福大学的研究者发现了一种富有弹性且透明的导电性能非常好的薄膜,这种薄膜由极易感触的碳纳米管组成,可被作为电极材料用在轻微触压和拉伸方面的传感器上。

“这种装置也许有一天可以被用在被截肢者、受伤的士兵、烧伤方面接触和压迫的敏感性的恢复上,也可以被应用于机器人和触屏电脑方面”,这个小组如是说。

鲍哲南和他的同事们在他们的弹透薄膜的顶部和底部喷上一种碳纳米管的溶液形成平坦的硅板,覆盖之后,研究人员拉伸这个胶片,当胶片被放松后,纳米管很自然地形成波浪般的结构,这种结构作为电极可以精准的检测出作用在这个材料上的力量总数。

事实上,这种装配行为上很像一个电容器,用硅树脂层来存储电荷,像一个电池一样,当压力被作用到这个感应器上的时候,硅树脂层就收紧,并且不会改变它所储存的电荷总量。

这个电荷是被位于顶部和底部的硅树脂上的纳米碳管测量到的。

当这个复合膜被再次拉伸的时候,纳米管会自动理顺被拉伸的方向。

薄膜的导电性不会改变只要材料没有超出最初的拉伸量。

事实上,这种薄膜可以被拉伸到它原始长度的2.5倍,并且无论哪种方向不会使它受到损害的拉伸它都会重新回到原始的尺寸,甚至在多次被拉伸之后。

当被充分的拉伸后,它的导电性喂2200S/cm,能检测50KPA的压力,类似于一个“坚定的手指捏”的力度,研究者说。

“我们所制作的这个纳米管很可能是首次可被拉伸的,透明的,肤质般感应的,有或者没有碳的纳米管”小组成员之一Darren Lipomi.说。

这种薄膜也可在很多领域得到应用,包括移动设备的屏幕可以感应到一定范围的压力而不仅限于触摸;可拉伸和折叠的几乎不会毁坏的触屏感应器;太阳能电池的透明电极;可包裹而不会起皱的车辆或建筑物的曲面;机器人感应装置和人工智能系统。

其他应用程序“其他系统也可以从中受益—例如那种需要生物反馈的—举个例子,智能方向盘可以感应到,如果司机睡着了,”Lipomi补充说。

显微粒子图像测速技术Micro-PIV研究进展

显微粒子图像测速技术Micro-PIV研究进展

M =10 NA =0.25
2.9 4.3 9.4 18 55
M =20 NA =0.5
2.3 2.8 5.0 9.3 27
M =40 NA =0.6
2.2 2.6 4.3 7.9 23
M =40 NA =0.75
2.2 2.4 3.7 6.4 18
M =60 NA =0.1.4
2.1 2.2 2.6 3.9 10
2
面,获得 2 维粒子图像,只能实现 2 维速度测量。
v
s t
(1)
但由于微器件特征尺度十分微小,对其内部流动进行测量提出了许多新的挑战。因此,与传统 PIV 相 比,Micro-PIV 在粒子图像获取方法、示踪粒子和图像处理三方面存在较大差别[12],下面将对这三方面差 别及关键技术的最新研究进展进行介绍。
1 引言
近年来,微流动器件(Microfluidic devices)应用成为国内外研究的热点[1,2],微流动器件结构和功能也越 来越复杂,微流动器件内部的流动行为问题成为了研究的重要内容[3,4],引起来了广泛关注[5,6]。已有研究 表明,由于流动特征尺度的微小,流体分子间作用力、静电力等表面力效应相对增强,同时流动还受到微 流体器件构型、壁面粗糙度和浸润性等因素影响,微尺度流动行为表现得非常复杂,目前还无法对这些复 杂流动现象进行合理的解释[7,8]。流动可视化技术是微尺度流动研究的重要实验方法[9,10]。显微粒子图像测 速技术(Micro-scale particle image velocimetry, Micro-PIV or µ PIV)是 20 世纪 90 年代发展起来的一种微尺度 流动测量与显示技术[11],可以实现无干扰、整场、瞬态、定量的微尺度速度场测量,有效测量的尺度范围 为 0.1-100µm[12],目前已达到相当高的分辨率(<1µ m) ,测速范围从每秒数纳米到数米,成为重要的微流 动研究手段,受到了研究者的广泛关注。目前,其它的微尺度流场测速技术,如磷光显示测速[13]、光漂白 测速[14]、分子标记测速[15]和拉曼散射技术[16]等,只能获得定性研究结果,也被称作流动定性可视化研究, 其分辨率和测量精度都无法和 Micro-PIV 相比。 Micro-PIV 是在 PIV 技术基础上发展起来的,两者基本原理相同,都是通过观测流场中散布的示踪粒 子,获得两副或多幅粒子图像,并对这些粒子图像进行空间相关性分析得到流场速度[17-24]。但是由于两者 的光路设计及组成部件有重大不同,一般认为这是两种相互独立流场测速技术[12,25,26]。最早的利用示踪粒 子进行微流动可视化研究的技术也被称作微尺度粒子跟踪技术(Micro-scale particle tracking, PTV)[27],最初 被应用于生物和医学研究中[28]。利用 PTV 技术,Taylor 等人[29]和 Brody 等人[30]使用超荧光显微镜对直径 900nm 的荧光示踪粒子进行长时间照明观测,获得粒子运动轨迹图像。这种方法获得的速度场具有不准确

Amorphous SiC as a structural layer in microbridge-based RF MEMS switches

Amorphous SiC as a structural layer in microbridge-based RF MEMS switches

Amorphous SiC as a structural layer in microbridge-based RF MEMS switches for use in software-defined radioRocco J.Parro a ,Maximilian C.Scardelletti b ,Nicholas C.Varaljay b ,Sloan Zimmerman c ,Christian A.Zorman a,*aDepartment of Electrical Engineering and Computer Science,Case Western Reserve University,715A Glennan Building,10900Euclid Avenue,Cleveland,OH 44106,USA bElectron and Optical Device Branch,NASA Glenn Research Center,USA cSolon High School,Solon,OH,USAa r t i c l e i n f o Article history:Received 29May 2008Accepted 7June 2008Available online 25July 2008Review of this manuscript was arranged by A.Iliadis,C.Richter,and A.Zaslavsky Keywords:PECVD silicon carbide RF MEMS MicrobridgeYoung’s modulus Residual stressa b s t r a c tThis paper reports an effort to develop amorphous silicon carbide (a-SiC)films for use in shunt capacitor RF MEMS microbridge-based switches.The films were deposited using methane and silane as the precur-sor gases.Switches were fabricated using 500nm and 300nm-thick a-SiC films to form the microbridges.Switches made from metallized 500nm-thick SiC films exhibited favorable mechanical performance but poor RF performance.In contrast,switches made from metallized 300nm-thick SiC films exhibited excel-lent RF performance but poor mechanical performance.Load-deflection testing of unmetallized and met-allized bulk micromachined SiC membranes indicates that the metal layers have a small effect on the Young’s modulus of the 500nm and 300nm-thick SiC MEMS.As for residual stress,the metal layers have a modest effect on the 500nm-thick structures,but a significant affect on the residual stress in the 300nm-thick structures.Ó2008Elsevier Ltd.All rights reserved.1.IntroductionNASA is looking to replace existing communications architec-tures that make use of multiple receivers with a single,state-of-the-art software-defined radio (SDR)to meet the requirements of future missions to the Moon and Mars.This envisioned system will require RF front-end hardware to operate at S-and K-band fre-quencies.RF MEMS switches may be a viable technology option for such systems because RF MEMS switches have the potential to add configurability and improved RF performance to SDR sys-tems,while at the same time reduce system complexity,size,weight and cost.One of the most common RF MEMS switch designs for commu-nications systems is based on suspended microbridges.In its sim-plest form,a microbridge-based RF MEMS switch consists of two conductors,an anchored conductor that is fixed to the substrate and the microbridge conductor that is perpendicular to and spans the anchored conductor in regions where they intersect.The microbridge conductor is designed so that it has sufficient mechan-ical flexibility to support actuation upon application of an attrac-tive potential to the anchored conductor.The microbridge must have sufficient stiffness such that once the attractive potential is removed,the microbridge returns to its unactuated up position.The microbridge must be made of a material that resists plastic deformation so that it can maintain consistent performance over millions of switching cycles,resists charge trapping so as to sup-press transient attractive forces that might keep the bridge in the down ON-state,and resists chemical modification so that mois-ture-induced stiction does not chemically bond the microbridge to the anchored electrode when in the actuated (ON)state.More-over,the microbridge must be made from a material that is com-patible with the other materials of the microswitch in terms of material processing and device fabrication procedures.These four conditions make the selection of the microbridge material a very challenging task.Previous research in the area of microbridge-based RF MEMS switches has focused on the use of Au for both the fixed electrodes and the suspended microbridges.This approach is particularly appealing from the design and fabrication perspectives since de-vices made in this manner involve the fewest number of processing steps and materials.Unfortunately,Au microbridges lack the high spring constant (related to Young’s modulus)required to overcome stiction and electrostatic charging effects so that they can snap back to the unflexed (OFF)state when the applied force is removed.0038-1101/$-see front matter Ó2008Elsevier Ltd.All rights reserved.doi:10.1016/j.sse.2008.06.004*Corresponding author.Tel.:+12163686117.E-mail addresses:rocco.parro@ (R.J.Parro),christian.zorman@ (C.A.Zorman).Solid-State Electronics 52(2008)1647–1651Contents lists available at ScienceDirectSolid-State Electronicsj o u r n a l ho m e p a g e :w w w.e l sevier.com/locate/sseAu microbridge-based switches suffer from mechanical deforma-tion during long-term use due to the fact that Au is a ductile mate-rial.These two key materials-related issues inhibit the deployment of such systems in applications that require reliable operation over long periods of time.Due to its outstanding mechanical properties,SiC is an appeal-ing material for microbridge-based RF MEMS switches.Crystalline SiC has a high Young’s modulus(>350GPa)which translates to a high spring constant when the material is configured into aflexible microbridge.This property ensures that aflexed microbridge will return to its unflexed state when the applied actuation force is re-moved.Crystalline SiC is a brittle material that can be elastically deformed;therefore a SiC microbridge should retain its original shape after many actuation cycles.SiC surfaces are extremely hydrophobic,making them highly resistant to stiction,a mois-ture-related phenomenon that causes MEMS-based switches made from hydrophilic materials(e.g.,Si)to remain in contact with the actuation electrode after the switching force is removed.Unfortu-nately,the high thermal processing temperatures associated withsingle crystalline and polycrystalline-SiC growth(>900°C)are par-ticularly problematic for electrostatically-actuated,SiC micro-bridge-based RF MEMS switches,especially capacitive shunt switches,since such structures require that the microbridge span afixed metallic electrode typically made from Au,which is intoler-ant to such high processing temperatures.High temperature alter-natives such as Ni or W can possibly be used in place of Au; however,Au is the preferred metal due to its inertness,very high electrical conductivity,and ease of processing.Plasma-enhanced chemical vapor deposition(PECVD)can be used to produce SiC at much lower deposition temperatures than conventional CVD,typically below400°C[1,2].SiCfilms can be deposited by PECVD using a wide variety of precursors,including SiH4and CH4[3],SiH4and C2H4[4],(CH3)3SiH[5],CH3SiCl3[6], and C6H18Si12[7].SiCfilms deposited by PECVD are typically amor-phous in microstructure and electrically insulating.The low depo-sition temperatures and insulating properties make PECVD SiC films particularly attractive for microbridge-based actuators[8–10].The amorphous microstructure generally corresponds to a lower Young’s modulus than found in crystalline SiC,ranging from 56±1GPa[10]and88±10GPa[11]to196±10GPa[12]depend-ing on precursor and processing conditions.At the high end of this range,the Young’s modulus is considerably higher than Au (78GPa)and thus would lead to microbridges with higher spring constants and thus more favorable mechanical switching proper-ties.The purpose of this study was to develop amorphous SiCfilms for use as structural layers in shunt capacitor-based RF MEMS switches,to explore the trade-offs between RF and mechanical per-formance,and to examine the influence of metal layers on the mechanical properties of micromachined SiC structures.2.Experimental2.1.RF MEMS switch designThe amorphous SiC-based RF MEMS shunt capacitor switch developed in this study is shown schematically in Fig.1.This sin-gle-pole single-throw(SPST)RF MEMS switch incorporates a dou-bly anchored SiC microbridge fabricated on a sapphire substrate with a dielectric constant of9.4.The MEMS switch is implemented in coplanar waveguide(CPW)transmission line media with a cen-ter conductor width(S),a slot width(W)and a ground width(G)of 130,60and300l m,respectively,which equates to a characteristic impedance(Zo)of approximately50X.The doubly anchored microbridge consists of an insulating a-SiC structural layer and a chrome/gold(Cr/Au)conductor to allow actuation.In this study, two switch designs that differ only by the SiC structural layer thickness were investigated.Thefirst switch design incorporated a500nm-thick a-SiC microbridge and the second a300nm-thick microbridge.2.2.Switch fabricationThe SiC microbridge switches were fabricated on500l m-thick, 75mm-diameter sapphire wafers.Thefirst step in device fabrica-tion involved the creation of the Cr/Au ground plane and center conductors by electron-beam deposition and lift-off patterning. The thickness of the Cr layer was25nm and that of the Au layer was100nm.Follow the patterning step,a2l m-thick SiO2sacrifi-cial layer was deposited by PECVD.The oxide layer was patterned by reactive ion etching to form vias for the microbridge anchors.An a-SiCfilm was then deposited by PECVD in a Technics Micro-PD Series900reactor using silane(SiH4)and methane(CH4)as precur-sors atflow rates of10.0±0.1sccm and40.0±0.1sccm,respec-tively.Films were deposited at a substrate temperature of 300±5°C,a chamber pressure of900±5mTorr and an RF power of100±1W at30kHz.Following SiC deposition,a low-tempera-ture anneal at450°C in a Tempress horizontal diffusion furnace was performed to convert the stress state of the as-depositedfilms from compressive to tensile[4].The300nm-thickfilm was an-nealed for30min,while the500nm-thick membrane wafer was annealed for60min.The furnace ramp time was set to22min at an average temperature ramp rate of19°C/min.The annealing step results in a modest densification of the as-depositedfilm as indi-cated by a decrease of only8nm for the300nm-thickfilm and 15nm for the500nm-thickfilm.Following the annealing step,a second Cr/Au layer totaling 125nm in thickness(25nm of Cr and100nm of Au)was deposited by electron-beam evaporation and patterned into the microbridge conductor by lift-off.The underlying SiC layer was then patterned by reactive ion etching(RIE)using the Cr/Au layer as an etch mask. After the RIE step,the wafer was diced and each microbridge was released by immersion in buffered HF.Following release,the de-vice was subjected to critical-point drying to eliminate the possi-bility of stiction.At this point the devices are ready for testing. Fig.2is an SEM micrograph of a fabricated device after release.2.3.Switch testingThe SiC-based SPST RF MEMS switches were characterized using an HP8510Vector Network Analyzer(VNA).The two-port MEMS switches were tested on a Cascade Microtech Model42 Microwave R and D Probe Station and two150l m-pitch ground-signal-ground(GSG)probes from GGB Industries were usedto Fig.1.Schematic diagram of an amorphous SiC-based SPST RF MEMS switch.1648R.J.Parro et al./Solid-State Electronics52(2008)1647–1651make contact to the switch via microwave test cables to ports 1and 2on the VNA.The system was calibrated with on wafer thru-reflect-line (TRL)calibration standards to ensure accuracy.The return loss (S 11)and insertion loss (S 21)in the OFF-state (un-flexed)as well as the isolation and S 11in the ON-state (flexed)for the MEMS switches were measured to determine the effective-ness of each switch design.2.4.Membrane preparationTo characterize the mechanical properties of the a-SiC films used in the microbridge switches,bulk micromachined mem-branes of the same material as the structural layers of the micro-bridges were fabricated.In this case,300nm and 500nm SiC films were deposited onto 350l m-thick (100)Si wafers.The films were deposited on both sides of each wafer and annealed using the process described above.The film on the backside of each wafer was patterned into etch windows by RIE using photoresist as an etch mask.Each wafer was then subjected to a 1:5(wt%)aqueous solution of KOH heated to 55°C to selectively remove Si.Under these conditions,the etch could be completed in roughly 18hours.The resulting square membranes had dimensions of 750l m Â750l m.2.5.Load-deflection testingIn order to characterize the mechanical properties of a-SiC,load-deflection measurements,which have previously focused on materials such as silicon nitride (Si 3N 4)[13,14]and 3C-SiC [15]as well as a-SiC [4,5,11],were conducted.Each membrane chip was mounted to a specialized chuck that enables a differential pressure to be applied to the membrane under test.This pressure was measured by an electronic transducer and recorded by a PC.The pressure caused the membrane to deflect,and this deflection was measured using a microscope-mounted interferometer.Con-centric rings were generated from the resulting interference pat-tern,and the spacing between each ring translated to 270nm of membrane deflection.Pressure was slowly released from the membrane and a pressure reading was made each time a ring dis-appeared.Measurements continued until the differential pressure reached zero.The Young’s modulus and residual stress of the a-SiC films were extracted from the load-deflection test results by using the follow-ing equation [16]:P ¼t W O C 1r O þf ðv ÞE W 2O ð1Þwhich when linearized with respect to W 2O becomes:P O ¼t C 1r O þf ðv ÞE W 2Oð2Þwhere P is the applied differential pressure,W O is the total deflec-tion,t is the membrane thickness,a is one-half of the membrane side length,C 1is a constant equal to 3.41[17],r O is the residual stress of the film f (v )is a function equal to 1.37Â(1.446–0.427Áv )[17],v is Poisson’s ratio,and E is Young’s modulus.In this work,v is assumed to be equal to 0.23,which is consistent with val-ues for a-SiC reported in the literature [10–12].Through curve-fitting,coefficients for the first and third-order deflection terms were found,and with knowledge of the other parameter values,the residual stress and Young’s modulus were calculated.As the Young’s modulus is in the coefficient of the third-order deflection term,a sufficient differential pressure must be applied to the membrane in order to extract an accurate value.It has been previously reported [15]that the greatest contribution of error in load-deflection testing using the aforementioned tech-nique is due to the operator’s ability to determine when a pressure reading should be made,so a minimum of ten trials was per-formed,for which the average and standard deviation were calculated.After performing the aforementioned tests,the membrane chip was then coated on a single side with the Cr/Au metallization used in the microbridge switches and retested following the procedure detailed above.In this case,the values of the residual stress and Young’s modulus extracted from the data were those of the multi-layered film.According to the law of mixtures,the Young’s modu-lus of the multilayer is related to the equivalent properties of the metal and SiC film by the following relationship [13]:E C ¼t S t S þt ME S þt Mt S þt ME M ð3ÞThe subscripts in both equations denote the multilayered com-posite (C),the metal layer (M),and the a-SiC layer (S).The totalthickness of the composite layer (t C )is simply equal to the sum of the metal and a-SiC layer thicknesses,t S +t M .Likewise,t M is equal to the sum of the Cr and Au thicknesses,t Cr +t Au .For this study,the normalized film thicknesses of Cr and Au were 0.192and 0.808,respectively.By substituting the previously extracted values for E S and measured values for t S and t M ,expected values for E C were calculated.A comparison of the measured values with those calculated using the aforementioned equation and published values for both Cr and Au were used to determine if the law of mix-tures appropriately describes mechanical properties of the multi-layer,perhaps leading to an explanation of microbridge performance data as discussed later in this paper.3.Results3.1.RF MEMS switchThe return loss (S 11),insertion loss (S 21),isolation and S 11(down-position)for the MEMS switch with the 500nm-thick SiC microbridge is shown in Fig.3.In the OFF-state (up-position),S 11is better than –20dB and S 21is less than 0.2dB for frequencies be-tween 0and 30GHz.To actuate the switch into the ON-state (down-position)a 45volt square wave was applied to the DC bias pads.S 11in the ON-state (down-position)and the isolation do not reach the values necessary to ensure adequate RF response of a shunt capacitor (i.e.,beyond –10dB).However,this same MEMS switch performs very well as a mechanical actuator in that itdoesFig.2.SEM micrograph of an amorphous SiC microbridge-based switch.R.J.Parro et al./Solid-State Electronics 52(2008)1647–16511649not suffer from stiction and can be actuated repeatedly.The RF re-sponse in the ON-state for the300nm-thick SiC microbridge switch is illustrated in Fig.4.The S11is less than0.2dB and the iso-lation is better than–10dB above5GHz.The isolation actually im-proves as the frequency increases resulting in an isolation of better than–30dB at40GHz.Although the300nm-thick a-SiC micro-bridges provide superior RF performance in the on-state,they suf-fer from a lack of mechanical reliability as they fail to actuate after a few cycles if at all,suggesting that the mechanical properties of the bilayer made from300nm-thick a-SiC are somehow adversely affected by the metal conductor.3.2.Young’s modulus and residual stressFig.5shows load-deflection plots of an unmetallized and met-allized500nm-thick a-SiC membrane.From these data,as well as data from the other two membranes,the average Young’s mod-ulus of the500nm-thick SiCfilm was129±5GPa and the average residual stress was162±6MPa.Metallizing these membranes hada modest affect on the mechanical properties.The average effective Young’s modulus of the multilayer was118±4GPa and the aver-age residual stress was170±5MPa.Fig.6shows load-deflection plots of an unmetallized and metallized300nm-thick membrane. In this case,the average Young’s modulus of the300nm-thick SiCfilm was133±6GPa and the average residual stress was 175±10MPa.By comparison,the metallized bilayer had an aver-age effective Young’s modulus of122±6GPa and an average resid-ual stress of204±22MPa.It should be mentioned that for both the unmetallized and metallized membranes,the Young’s modulus and residual stress values for individual trials were randomly dis-tributed about the mean values for each membrane,indicating that neither the SiC nor the metal layers were damaged by the loads ap-plied during the experiments and the resulting deflections.4.AnalysisThe performance data from the two RF switch designs indicate that as the SiC microbridge thickness is reduced to improve its per-formance as a shunt capacitor,its mechanical properties are mod-ified in an adverse manner,possibly by the increasing influence of the metal conductor that it is supporting.The load-deflection data indicate that the metal layer has a modest effect on the effective Young’s modulus for both the300and500nm-thickfilms,reduc-ing the modulus by only11GPa in both cases.In terms of residual stress,however,the metal layer has a significant effect on struc-tures made from the300nm-thick SiCfilm,increasing it by 29MPa as compared with only8MPa for structures made from 500nm-thickfilms.To determine if the law of mixtures provides an appropri-ate means to estimate the Young’s modulus of the multilayeredFig.5.Load-deflection behavior of unmetallized and metallized500nm-thick SiCmembranes.Fig.6.Load-deflection behavior of unmetallized and metallized300nm-thick SiCmembranes.1650R.J.Parro et al./Solid-State Electronics52(2008)1647–1651Au/a-SiC structures,a comparison of measured values to those cal-culated using Eq.(3)were ing measured thickness values for the Cr and Au layers and literature values for the Young’s mod-uli of Cr and Au,an effective Young’s modulus of the metallization layer was calculated to ing this value and the mea-sured thickness and Young’s modulus of the a-SiCfilms,the effec-tive modulus for each multilayer was calculated.Table1 summarizes the results of these calculations and compares these values to the measured values.The measured values for the multi-layered membranes made from the300and500nm-thickfilms differ from the calculated values by only2.4%and6.3%,respec-tively.These differences are roughly comparable with the uncer-tainties in the measured Young’s modulus values.Although the calculated values are,at best,estimates since they rely on litera-ture values for the Young’s moduli of the Cr and Au layers as op-posed to actual measured values for thesefilms,this analysis indicates that the law of mixtures does indeed provides a means to estimate the impact of the metalfilm on the mechanical proper-ties of the multilayered structure and thus can be a useful tool when designing a-SiC microbridge-based switches.The results presented herein do not explain the poor mechani-cal performance of the shunt capacitor-based RF MEMS switches made from the300nm-thick a-SiCfilms because the effective Young’s modulus was nominally the same as the switch made from the500nm-thickfilm and the larger residual stress in the switch made from the300nm-thickfilm should actually improve the off-state performance.Thefindings presented herein do,however, show that the residual stress in the metallization layer should be taken into account when designing multilayered microswitches based on a-SiC microbridges.To reduce the influence of the metal layer on the mechanical properties of the microbridge,the thick-ness and/or the residual stress in the metal layer should be mini-mized.However,if the residual stress in the metalfilm cannot be reduced,the post-deposition annealing step for the SiCfilm could be omitted,yielding a net compressive stress in the SiC micro-bridge which may counter the tensile stress in the metal layer, resulting in a multilayered microbridge with a low tensile stress.5.ConclusionShunt capacitor-based RF MEMS switches have been fabricated using amorphous SiC as the principal structural layer in the micro-bridge.Thick microbridges(500nm)exhibit favorable mechanical performance but poor RF performance.The opposite is true for switches made from300nm-thickfilms.Load-deflection testing of unmetallized and metallized SiC membranes indicates that the Young’s modulus of the amorphous SiCfilms is insensitive tofilm thickness and is only slightly reduced when metallized.The resid-ual stress of the500nm-thickfilm is only slightly affected by the metal layer,whereas that of the300nm-thick SiCfilm is signifi-cantly affected.Thesefindings suggest that the residual stress of the metal layer should be taken into account when designing microbridge-based switches using submicron-thick amorphous SiC structural layers,especially as the structural layer thickness ap-proaches that of the metallization layer.AcknowledgementsThe authors would like to thank the NASA Graduate Student Re-search Program,the NASA Lewis Educational and Research Collab-orative Internship Program,and The NASA Glenn Summer Faculty Fellowship program for funding this research.References[1]Tong L,Mehregany M,Tang WC.Amorphous silicon carbidefilms by plasma-enhanced chemical vapor deposition.In:Proceedings of IEEE micro electro mechanical systems;February1993.p.242–7.[2]Solomon I,Schmidt MP,Tran-Quoc H.Selective low-power plasmadecomposition of silane-methane mixtures for the preparation of methylated amorphous silicon.Phys Rev B1988;38(14):9895–901.[3]Jean A,Chaker M,Diawara Y,Leung PK,Gat E,Mercier PP,et al.Characterization of a-SiC:Hfilms produced in a standard plasma enhanced chemical vapor deposition system for X-ray mask application.J Appl Phys 1992;72(7):3110–5.Oct.[4]Windischmann H.Intrinsic stress and mechanical properties of hydrogenatedsilicon carbide produced by plasma-enhanced chemical vapor deposition.J Vac Sci Technol A1991;9(4):2459–63.[5]Du J,Singh N,Summers J,Zorman CA.Development of PECVD SiC for MEMSusing3MS as the precursor.In:Materials research society symposium proceedings,vol.919;2005.p.283–8.[6]Ivashchenko VI,Dub SN,Porada OK,Ivashchenko LA,Skrynskyy PL,Stegniy AI.Mechanical properties of PECVD a-SiC:H thinfilms prepared from methyltricholorosilane.Surf Coat Technol2006;200(22–23):6533–7.[7]Klumpp A,Schaber U,Offerins HL,Kuhl K,Sandmaier H.Amorphous siliconcarbide and its application in silicon micromachining.Sens Actuat A:Phys 1994;41(1–3):310–6.[8]Sarro PM,deBoer CR,Korkmaz E,Laros JMW.Low-stress PECVD thinfilms forIC-compatible microstructures.Sens Actuat A:Phys1988;67(1–3):175–80. [9]Carreño MNP,Lopes AT.Self-sustained bridges of a-SiC:Hfilms obtained byPECVD at low temperatures for MEMS applications.J Non-Cryst Solids2004;338–340:490–5.[10]Flannery AF,Mourlas NJ,Storment CW,Tsai S,Tan SH,Heck J,et al.PECVDsilicon carbide as a chemically resistant material for micromachined transducers.Sens Actuat A:Phys1988;70(1–2):45–8.[11]El Khakani MA,Chaker M,Jean A,Boily S,Kieffer JC,O’Hern ME,et al.Hardnessand Young’s modulus of amorphous a-SiC thinfilms determined by nanoindentation and bulge tests.J Mater Res1994;9(1):96–103.[12]Cros B,Gat E,Saurel JM.Characterization of the elastic properties ofamorphous silicon carbide thinfilms by acoustic microscopy.J Non-Cryst Solids1997;209:273–82.[13]Tabata O,Kawahata K,Sugiyama S,Igarashi I.Mechanical propertymeasurements of thinfilms using load-deflection of composite rectangular membranes.In:Proceedings of IEEE micro electro mechanical systems,1989.An investigation of micro structures,sensors,actuators,machines and robots, February20–22;1989.p.152–6.[14]Bromley EI,Randall JN,Flanders DC,Mountain RW.A technique for thedetermination of stress in thinfilms.J Vac Sci Technol B1983;1(4):1364–6.[15]Mitchell JS,Zorman CA,Kicher T,Roy S,Mehregany M.Examination of bulgetest for determining residual stress,Young’s modulus,and Poisson’s ratio of 3C-SiC thinfilms.J Aerospace Eng2003;16(2):46–54.[16]Vlassak JJ,Nix WD.A new bulge test technique for the determination ofYoung’s modulus and Poisson’s ratio of thinfilms.J Mater Res 1992;7(12):3242–9.[17]Pan JY,Lin P,Maseeh F,Senturia SD.Verification of FEM analysis of load-deflection methods formeasuring mechanical properties of thinfilms.In:IEEE solid-state sensor and actuator workshop,1990.4th Technical digest,June4–7;1990.p.70–3.Table1Summary of the Young’s modulus values for metallized300nm-thick and500nm-thick a-SiCfilmsa-SiCfilm thickness (nm)Calculated effectiveYoung’s modulus(GPa)Measured effectiveYoung’s modulus(GPa)Percentdifference(%)300125122±6 2.4500126118±4 6.3The calculated values were determined using the law of mixtures.R.J.Parro et al./Solid-State Electronics52(2008)1647–16511651。

MIFS IH

MIFS IH
1 Complicated three dimensionally winding channel system is difficult to make 2 Assemble process is necessary 3 Packaging problem
a)Bending micro pipe with square cross section (1 OOx 1OOx 1000 pm) b) Branch iiiicro pipes (30pin) 2. Micro valve with passive movement Micro Venous Valve (One-way valve 8Okm)
Fig.3
On the other hand, IH process can solve above issues. As we demonstrated at MEMS'93, various type of micro fluidic parts made of UV polymer with 3D shape have been fabricated successfully by IH process.
1.INTRODUCTION
The new process for 3D micro fabrication proposed in MEMS'93 is shown as Fig.1.[1][2][3] This process temporally named "IH Process" (Integrated Harden Polymer Process) can fabricate two type of materials such as polymer and metals. In order to make polymer structure, thin layers made of UV solidified polymer are stacked from bottom to up to create complicated three dimensional shape such as base as shown in Fig. l(a). In addition to polymer, three dimensional micro structure made of metal can be obtained by using metal molding process combined with former polymer process as shown in Fig. l(b). In this case, the polymer structure is utilized as a cast and the metal is electroplated into this cast. And then the polymer is removed by solvent or the appropriate chemical process. Finally micro three dimensional metal structure is obtained. This process can applicablc not only for any electroplatable metals but also lor other materials which can be molded into polymer s tr tic t lire. This apparatus has following specifications. I . UV light source: Xenon lamp 2 . Spot size of UV beam: 5pm 3 . Position accuracy: 0.25 ~ 0 . 2 5.O 1 ~ [~tm] (X,Y,Z> 4. Minimum size of solidification unit 5 s 5 x3 [pin] (X,Y,Z)

Series 925 MicroPirani

Series 925 MicroPirani

Series 925MicroPirani™ TransducerThe Series 925 MicroPirani™ transducer is a thermal conductivity gauge based on a unique, MEMS-based (Micro-Electro-Mechanical Systems) sensor. The 925 is used for vacuum pressure measurement and offers analog voltage output, digital interface and set point relays for process controlling.The 925 Transducer offers a wide measurement range from 1x10measurement of thermal conductivity. The MicroPirani sensor consists of a silicon chip with a heated resistive element forming one surface of a cavity. A cover on top of the chip forms the other surface of the cavity. Dueto the geometry of the sensor, convection cannot take place within the cavity and consequently, the sensor is insensitive to the mounting position. Gas molecules are passed by diffusion only to the heated element wherethe heat loss of the gas is measured.ApplicationsThe 925 can be used in many differentvacuum applications within the semiconductor,analytical, and coating industries:General vacuum pressure measurementForeline and roughing pressure measurementGas backfilling measurement and controlMass spectrometer controlActivation of UHV gaugesSystem process controlControl system pressureLike all thermal conductivity sensors, the 925 is sensitive to Array gas type. To compensate for gas dependency, the MicroPiranihas a number of common gas calibrations that can beselected via the digital interface. This makes it a simplesolution for locating medium to fine leaks in vacuum systems.The 925 has RS232, RS485, and EtherCAT digitalcommunication interface for setup of transducer parametersand to provide real time pressure measurement.The 925 also has a analog pressure output of 1 VDC/decadethat can be interfaced to external analog equipment forpressure readout or controlling. Other analog outputs andcurves can be selected via the digital user interface.The 925 has up to three mechanical relays which can beused for process control, examples are interlocking valvesor pumps. The 925 compact design significantly reducesthe amount of space occupied by a vacuum gauge. This isparticularly appealling to system designers and allows for amore compact vacuum system.Dra winga lDimensionNote: Unless otherwise specified, dimensions are nominal values in inches(mm referenced).PinOutsThree (3) set point relays and dual Aout, 15 pin D Subminiature and RJ45 EtherCAT IN/OUT ConnectorsSpecificationsSensor Type MicroPirani (MEMS Thermal Conductivity) Measuring Range 1.0 x 10-5 Torr to AtmosphereSet Point Range 5.0 x 10-4 Torr to 500 TorrCalibration Gas Air, Argon, Helium, Nitrogen, Hydrogen, H2O vapor, CO2, Xenon, NeonOperating Temperature Range 0° to 40°C (32° to 104°F)Maximum Bakeout Temperature 80°C (176°F), non-operatingCommunication RS485 / RS232 (4800 to 230400 Baud)Controls Zero adjust, atmosphere adjust, pressure units, baud rate, address, factory default, gas type;set point functions: value, hysteresis, direction, enable analog output transducer status, switch,LEDtestStatus Pressure reading and units, set point, operating time, transducer temperature, user tag, model,device type, serial number, firmware and hardware versions part number, manufacturer Analog Output 1 to 9 VDC, 100W maximum output impedance, 1 volt/decadeAnalog Output Resolution 16 bitRelays (Optional) 925 - 3 relays SPDTRelay Contact Rating 1 A @ 30VAC/DC, resistiveRelay Response<100 msec maximumPower Requirements 9 to 30 VDC, < 1.5 watts maxAccuracy (Typical)1 5 x 10-4 to 10-3 Torr ±10% of Reading10-3 to 100 Torr ±5% of Reading100 Torr to atm ±25% of ReadingRepeatability (Typical)110-3 to 100 Torr ±2% of ReadingOverpressure Limit 3000 Torr absoluteInstallation Orientation AnyInternal Volume (KF16) 2.80 cm3Materials Exposed to Vacuum 304 stainless steel, Silicon, SiO2, Si3N4, Gold, Viton®,Low out gassing epoxy resinElectronic Casing and Flange 304 stainless steelWeight (KF 16) 170 gCompliance CE, ETG.5003.2080 Vacuum Pressure Gauge 1 Accuracy and repeatability are typical values measured with Nitrogen gas at ambient temperature after zero adjustment.Ordering InformationOrdering Code Example: 925-11030Code Configuration925 with Displ a yThe optional integrated touch-screen display is user configurable; the user can change pressure units, orientation and has access to set point parameters as well as gas type. The display also indicates the status of the available set point relays. Displayed reading can be seen from >5 meters away on the high contrast display.PDR900 Power Supply and DisplayThe PDR900 power supply and readout unit is a stand alone, single channel controller for use with the Series 900 digital vacuum transducers. It can be used as a stand-alone power supply readout unit or as a tool for configuration, calibration and diagnostics of system integrated transducers in OEM applications.+1-978-645-5500 I +1-800-227-8766MKS products provided subject to the US Export Regulations. Export, re-export, diversion or transfer contrary to US law (and local country law) is prohibited.mksinst ™ and MicroPirani ™ are trademarks of MKS Instruments, Inc., VCR ® is a registered trademark of Swagelok Co. Viton ® is a registered trademark of E.I Dupont Co., Inc. EtherCAT ® is a registered trademark and patented technology, licensed by Beckhoff Automation GmbH, Germany. U.S. Patent No. 6,672,171. Other patents pending.925_01/20©2020 MKS Instruments, Inc.Specifications are subject to change without notice.。

silicon waveguide01Front

silicon waveguide01Front

Fabrication of micro-mirrors in silicon optical waveguidesOlly Powell B.Sc.Hons.School of Microelectronic EngineeringGriffith UniversityandThe CRC for micro TechnologySubmitted in fulfilment of the requirementsof the degree of Doctor of PhilosophyJuly5,2004iiAbstractThe conventional large radii bends used in large cross section silicon-on-insulator waveguides were replaced with novel wet etched corner mirrors,potentially allowing much smaller devices,therefore lower costs.If such corners had been based on reactive ion etch techniques they would have had the disad-vantage of rougher surfaces and poor alignment in the vertical direction.Wet etching overcomes these two problems by providing smooth corner facets aligned precisely to the vertical{100}silicon crystallographic planes.The waveguides obtained had angled walls,and so numerical analysis was undertaken to establish the single mode condition for such trapezoidal structures.To show the relationship between fabrication tolerances and optical losses a three dimensional simulation tool was developed,based on expansion of the incident mode into plane waves.Various new fabrication techniques were are proposed,namely:the use of titanium as a mask for deep silicon wet anisotropic etching,a technique for aligning masks to the crystal plane on silicon-on-insulator wafers,a corner compensation method for sloping sidewalls,and the suppression of residues and pyramids with the use of acetic acid for KOH etching.Also,it was shown that isopropyl alcohol may be used in KOH etching of vertical walls if the concentration and temperature are sufficiently high.As the proposed corner mirrors were convex structures the problem of undercutting by high order crystal planes arose.This was uniquely overcome by the addition of some structures to effectively convert the convex structures into concave ones.The corner mirrors had higher optical losses than were originally hoped for,similar to those of mirrors in thinfilm waveguides made by RIE.The losses were possibly due to poor angular precision of the lithography process.The design also failed to provide adequate mechanisms to allow the etch to be stopped at the optimal time.The waveguides had the advantage over thinfilm technology of large,fibre-compatible cross sections.However the mirror losses must be reduced for the technology to compete with existing large cross section waveguides using large bends.Potential applications of the technology are also discussed.The geometry of the crystal planes places fundamental limits on the proximity of any two waveguides.This causes some increase in the length of MMI couplers used for channel splitting.The problem could possibly be overcome by integrating one of the mirrors into the end of the MMI coupler to form an L shaped junction.“Any third rate researcher or engineer can increase complexity;but it takes a certainflair of real insight to make things simple again.”E.F.Schumacher,from Small is BeautifulStatement of originalityThe content of this thesis has not been previously submitted for any other degree,and to the best of my knowledge it has not been published or developed by others,except where due credit has been given.Olly PowellDateCommonly used acronymsAR Anti reflectionBESOI Bond and etch back silicon on insulator BHF Buffered hydrofluoric acidBPM Beam propagation methodCMOS Complimentary metal oxide semiconductor CVD Chemical vapour depositionCW/CCW Clock wise/counter clock wiseCZ(wafers)CzochralskiDI(water)De-ionisedDRIE Deep reactive ion etchingEIM Effective index methodFZ(wafers)Float zoneHF Hydrofluoric acidIPA Iso-propyl alcohol(propan-2-ol)KOH Potassium hydroxideMEMS Micro-electro-mechanical-systemsMMF Multi-modefibreMMI(coupler)Multi mode interferenceMZ Mach ZehnderPVD Physical vapour depositionRIE Reactive ion etchingSIMOX(wafers)Separation of implanted oxygenSMF Single-modefibreSOI Silicon on insulatorTE Transverse electricTEM Transverse electromagneticTMAOH Tetra-methyl ammonium hydroxideTM Transverse magneticWDM Wavelength division multiplexingContents1Introduction11.1Internally reflecting corner mirrors (1)1.2Structure of this thesis (2)2Background32.1Integrated optics (3)2.2Silicon for optical waveguiding-a brief history (3)2.3Optical properties of silicon (4)2.4Fabrication of silicon-on-insulator waveguides (6)2.5Competing integrated optics technologies (7)2.6Applications of silicon based integrated optics (9)2.7Analysis of planar optical waveguides (9)2.7.1Maxwell’s equations (10)2.7.2Modal analysis (10)2.7.3Separation of variables (11)2.7.4The vector Helmholtz equations (13)2.7.5Scalar paraxial beam propagation method (13)2.7.6Full vector,wide angle and bi-directional BPM (14)3Conditions for single-mode operation153.1Introduction (15)3.2Simulation by the beam propagation method (16)3.3Analysis of the vertical-walled rib structure (18)3.4Analysis of the trapezoidal rib structure (19)3.4.1Integration of waveguide power (19)3.4.2Effective index/matrix transfer method (20)3.4.3Marcatili’s method (22)3.5Conclusion (23)4Analysis of corner mirrors254.1Introduction (25)4.2Theory (26)4.2.1Decomposition into angular representation (26)4.2.2Decomposition into perpendicular and parallel components (28)4.2.3Phase changes between reference frames (30)4.2.4Projection onto the output mode (32)4.3Results (32)4.3.1Comparing the three plane wave models (32)4.3.2Effects of fabrication tolerances (33)4.4Conclusion (35)iiCONTENTS iii 5Silicon micromachining techniques375.1Standard processes (37)5.1.1Thermal oxide growth (37)5.1.2Chemical vapour deposition (38)5.1.3Physical vapour deposition (38)5.1.4Lithography (38)5.1.5Wet isotropic etching (39)5.1.6Wet anisotropic etching (39)5.1.7Dry isotropic etching (41)5.1.8Dry anisotropic etching (41)5.2KOH etching techniques (42)5.2.1Experimental details (42)5.2.2Results (43)5.3Etching in TMAOH (46)5.3.1TMAOH/Methanol/IPA/H2O (46)5.3.2TMAOH/H2O/IPA (46)5.3.3TMAOH/H2O (48)5.3.4Corner compensation in TMAOH (48)5.4Fabrication of Olly’s microhouse (49)5.4.1Application of photo-resist on high aspect ratio3D structures (49)5.5Precise alignment to the crystal direction (50)5.6Non-conventional masking materials (52)5.6.1Introduction (52)5.6.2Sputtered metalfilms as masks (53)5.6.3Titanium masks (54)5.6.4Titanium dioxide masks (58)5.6.5Patterning of TiO2films (60)5.6.6Electrical resistivity of Ti/TiO2 (61)5.6.7Metal masks for etching SOI (61)5.7Conclusion (63)6Fabrication of corner mirrors656.1Design issues (65)6.1.1Wet etched vertical walls for corner facets (65)6.1.2The undercutting of convex corners (65)6.1.3Alignment precision (66)6.1.4Masks and lithography (67)6.2Waveguide dimensions (71)6.2.1The SOI wafer specifications (72)6.2.2Thefinal process list (74)6.3Testing (75)6.3.1Sample preparation (75)6.3.2Experimental setup (76)6.3.3Waveguide and coupling losses (77)6.3.4Corner mirror losses (78)6.4Conclusion (79)7Possible applications827.1Introduction (82)7.2Channel splitting (82)7.2.1Y junctions,star couplers and evanescent couplers (82)7.2.2MMI couplers (83)7.2.3T junctions (86)7.2.4L junctions (86)7.3Unbalanced Mach-Zehnder interferometers (87)7.4Ring resonator (88)7.5An alternative ring resonator (90)iv CONTENTS7.6Conclusion (92)8Conclusion938.1Conditions for single-mode operation (93)8.2Corner simulations (93)8.3Micromachining techniques (94)8.4Fabrication of the corner mirrors,and applications (94)8.5Areas for further research (95)A Publications arising from this work96B Trigonomic proofs97B.1Calculation of W and H of the equivalent waveguide (97)B.2Transformations between the three coordinate systems (97)B.3Vector projections (100)B.4Calculation of reflected vectors (100)C Linear algebra102C.1Operation of a1×2MMI splitter/combiner (102)D TMAOH etching solutions103List of Figures2.1ASOC T M transceiver from Bookham Technology (4)2.2Optical absorption of single crystal silicon (5)2.3SOI rib waveguides (6)2.4Rectangular dielectric waveguides (11)2.5The effective index method (12)3.1Trapezoidal rib waveguide (16)3.2First order mode in rib waveguide (17)3.3BPM simulation in rib waveguide (18)3.4Cut-offconditions in vertical rib waveguides (18)3.5Cut-offconditions in trapezoidal rib waveguides (19)3.6Matrix transfer method (20)3.7Mode-width estimates in trapezoidal structures (22)3.8Thefinal curves for both types of rib structure (23)4.1The general problem (25)4.2Plane wave expansion method (27)4.3Projection of the tangential component of E x onto the perpendicular plane (29)4.4Comparing the plane wave models in the corner mirror situation (33)4.5Comparing different plane wave expansion models by BPM (34)4.6The effect of horizontal displacement,and the Goos-H¨a nchen shift (34)4.7The effects of structure size on the losses due to fabrication tolerances (35)4.8The effects of angular misalignment (36)5.1The self passivation of p type silicon during alkaline etching (40)5.2Structure types producible on a(100)silicon wafer (41)5.3Corner fabricated by DRIE (42)5.4Vessels used for wet etching (43)5.5The dependence on temperature and KOH concentration (44)5.6Dependence on concentration and IPA addition (45)5.7Undercutting at corners for the two different structures (45)5.8Structures etched in TMAOH (47)5.9Results in TMAOH (48)5.10Olly’s micro-house (49)5.11Microhouse fabrication process (49)5.12General limitation for alignment methods (50)5.13Alignment method (51)5.14Alignment trenches (51)5.15Alignment trenches (51)5.16Exposure of buried layer with under-etch (52)5.17Schematic of proposed improved method (53)5.18Mesas using nickel and tungsten as masks (54)5.19Structure using a titanium mask in TMAOH etching (55)5.20Mesas using titanium in KOH etching (57)5.21Structures made with TiO2masks (58)vvi LIST OF FIGURES5.22Thermal oxidation of silicon and titanium (59)5.23Through holes with an off-aligned TiO2mask (60)5.24Deep etch pits with titanium and TiO2 (61)5.25Etch rates as a function of mask coverage (62)6.1Corner compensation schemes (66)6.2Realignment scheme (68)6.3Full fabrication process (69)6.4SEM image of a corner mirror (69)6.5SEM image,top view (70)6.6Photoresist buildup under wings (70)6.7Fibre to waveguide coupling loss (71)6.8Beam scope2D representation (72)6.9E beam mask layout (73)6.10SEM image of the polished end facet (75)6.11Experimental setup for the testing of waveguide devices (76)6.12A series of corner mirrors (77)6.13Straight waveguide losses (78)6.14Blocks of four and six mirrors (78)6.15Predicted losses from larger misalignments (80)6.16Undercutting of the mirror features in TMAOH (81)6.17Improved test structure (81)7.1Waveguide splitting schemes (83)7.2Ring resonator pattern (84)7.3Junctions at the end of MMI couplers (84)7.4MMI splitters and combiners (85)7.5Defining L and T junction parameters (86)7.6L junction applications (87)7.7Unbalanced Mach Zehnder structure (88)7.8Cascaded2×2MZ structures (89)7.9Ring resonator (90)7.10Transfer function of MMI based ring resonator (91)7.11Alternative ring resonator structure (91)7.12Simulation of narrow bandfilter (92)B.1The geometrical structure of the problem (98)B.2Rotations (98)B.3Diagram of the incident and reflected vectors (100)。

微流控芯片-质谱联用接口的研究进展

微流控芯片-质谱联用接口的研究进展

综述生命科学仪器2020第18卷/10月刊微流控芯片-质谱联用接口的研究进展张荣楷谭聪睿徐伟*(北京理工大学生命学院北京100081)摘要:微流控芯片由于具有尺寸小、集成程度高、结构功能多样化和样品用量少等优势被广泛应用于化学、生命科学和医学等多个领域:质谱具有灵敏度高、检测速度快和便于定性定量分析等优点:微流控芯片与质谱的联用充分结合了二者各自的优势,通过简便的操作实现对微量样品的快速分析检测:接口的研究是二者联用的前提和关键.经过20余年的发展,微流控芯片与质谱的接口技术逐渐成熟,实现了高效稳定的离子化效果,保证了分析的效率和准确性。

本文总结了基于电喷雾和基质辅助激光解吸两种离子化方式中.微流控芯片与质谱接口的主要类型和相关应用并分析了目前存在的问题及未来发展方向。

关键词:微流控芯片;质谱;接口;电喷雾电离源;基质辅助激光解吸电离源中图分类号:0657文献标识码:A DOI:10.11967/2020181002Recent Advances of Microfluidic Chip-Mass Spectrometry InterfacesZhang Rongkai Tan Congrui Xu Wef(School of L ife Science,Beijing Institute of Technology,Beijing100081,China)Abstract:Microfluidic chips are widely used in many fields such as chemistry,life sciences and medical science due to their small size,high integration,diversified functions and little sample usage.Mass spectrometry has the advantages of high sensitivity,fast detection speed,and convenient qualitative and quantitative analysis.The combination of microfluidic chip and mass spectrometry fully combines their respective advantages and achieves the rapid analysis of trace samples through simple operation..The research on the interface is the key of microfluidic chip-mass spectrometry.After more than20years' development,the interface technology between the microfluidic chip-mass spectrometry has gradually matured,achieving an efficient and stable ionization effect,ensuring the efficiency and accuracy of the analysis.In this review,the main types and related applications of the interface based on ESI and MALDI between microfluidic chip-mass spectrometry,as well as the current problem and future development were discussed.Key Words:Microfluidic chip;Mass spectrometry;Interface;ESI;MALDI|CLC Number]0657[Document Code]A DOI::10.11967/20201810021、引言统基于流动注射分析、色谱和电泳的理论,通过减小通道内径、缩短通道长度以实现分离性能更微流控芯片,又称芯片实验室(lab on achip),是一种将生物、化学、医学分析过程的样品制备、反应、分离和检测等基本操作单元集成到微小尺寸芯片的技术。

参考文献标准格式

参考文献标准格式

参考文献标准格式参考文献类型:专著[M],论文集[C],报纸文章[N],期刊文章[J],学位论文[D],报告[R],标准[S],专利[P],论文集中的析出文献[A]电子文献类型:数据库[DB],计算机[CP],电子公告[EB]电子文献的载体类型:互联网[OL],光盘[CD],磁带[MT],磁盘[D K]A:专著、论文集、学位论文、报告[序号]主要责任者.文献题名[文献类型标识].出版地:出版者,出版年.起止页码(可选)[1]刘国钧,陈绍业.图书馆目录[M].北京:高等教育出版社,1957.15 -18.B:期刊文章[序号]主要责任者.文献题名[J].刊名,年,卷(期):起止页码[1]何龄修.读南明史[J].中国史研究,1998,(3):167-173.[2]OU J P,SOONG T T,et al.Recent advance in research on appli cations of passive energy dissipation systems[J].Earthquack Eng,199 7,38(3):358-361.C:论文集中的析出文献[序号]析出文献主要责任者.析出文献题名[A].原文献主要责任者(可选).原文献题名[C].出版地:出版者,出版年.起止页码[7]钟文发.非线性规划在可燃毒物配置中的应用[A].赵炜.运筹学的理论与应用——中国运筹学会第五届大会论文集[C].西安:西安电子科技大学出版社,1996.468.D:报纸文章[序号]主要责任者.文献题名[N].报纸名,出版日期(版次)[8]谢希德.创造学习的新思路[N].人民日报,1998-12-25(10).E:电子文献[文献类型/载体类型标识]:[J/OL]网上期刊、[EB/OL]网上电子公告、[M/CD]光盘图书、[DB/OL]网上数据库、[DB/MT]磁带数据库[序号]主要责任者.电子文献题名[电子文献及载体类型标识].电子文献的出版或获得地址,发表更新日期/引用日期[12]王明亮.关于中国学术期刊标准化数据库系统工程的进展[EB/OL]. /pub/wml.html,1998-08-16/1998-10-01.[8]万锦.中国大学学报文摘(1983-1993).英文版[DB/CD].北京:中国大百科全书出版社,1996.英文参考文献标准格式:论文参考文献格式规范计算机论文写作2010-04-10 13:38:59 阅读1372 评论0 字号:大中小订阅也可以在标点.之后加上一个空格,但一定要保证所有的项目空格个数一致一、参考文献的类型参考文献(即引文出处)的类型以单字母方式标识,具体如下:[M]--专著,著作[C]--论文集(一般指会议发表的论文续集,及一些专题论文集,如《***大学研究生学术论文集》[N]-- 报纸文章[J]--期刊文章:发表在期刊上的论文,尽管有时我们看到的是从网上下载的(如知网),但它也是发表在期刊上的,你看到的电子期刊仅是其电子版[D]--学位论文:不区分硕士还是博士论文[R]--报告:一般在标题中会有"关于****的报告"字样[S]-- 标准[P]--专利[A]--文章:很少用,主要是不属于以上类型的文章[Z]--对于不属于上述的文献类型,可用字母"Z"标识,但这种情况非常少见常用的电子文献及载体类型标识:[DB/OL] --联机网上数据(database online)[DB/MT] --磁带数据库(database on magnetic tape)[M/CD] --光盘图书(monograph on CDROM)[CP/DK] --磁盘软件(computer program on disk)[J/OL] --网上期刊(serial online)[EB/OL] --网上电子公告(electronic bulletin board online)很显然,标识的就是该资源的英文缩写,/前面表示类型,/后面表示资源的载体,如OL表示在线资源二、参考文献的格式及举例1.期刊类【格式】[序号]作者.篇名[J].刊名,出版年份,卷号(期号)起止页码.【举例】[1] 周融,任志国,杨尚雷,厉星星.对新形势下毕业设计管理工作的思考与实践[J].电气电子教学学报,2003(6):107-109.[2] 夏鲁惠.高等学校毕业设计(论文)教学情况调研报告[J].高等理科教育,2004(1):46-52.[3] Heider, E.R.& D.C.Oliver. The structure of color space in naming and memory of two languages [J]. Foreign Language Teaching and Research, 1999, (3): 62 67.2.专著类【格式】[序号]作者.书名[M].出版地:出版社,出版年份:起止页码.【举例】[4] 刘国钧,王连成.图书馆史研究[M].北京:高等教育出版社,1979:15-18,31.[5] Gill, R. Mastering English Literature [M]. London: Macmillan, 1985: 42-45.3.报纸类【格式】[序号]作者.篇名[N].报纸名,出版日期(版次).【举例】[6] 李大伦.经济全球化的重要性[N]. 光明日报,1998-12-27(3).[7] French, W. Between Silences: A Voice from China[N]. Atlantic Weekly, 1987-8-15(33).4.论文集【格式】[序号]作者.篇名[C].出版地:出版者,出版年份:起始页码.【举例】[8] 伍蠡甫.西方文论选[C]. 上海:上海译文出版社,1979:12-17.[9] Spivak,G. "Can the Subaltern Speak?"[A]. In C.Nelson & L. Grossberg(eds.). Victory in Limbo: Imigism [C]. Urbana: University of Illinois Press, 1988, pp.271-313.[10] Almarza, G.G. Student foreign language teacher's knowledge growth [A]. In D.Freeman andJ.C.Richards (eds.). Teacher Learning in Language Teaching [C]. New York: Cambridge University Press. 1996. pp.50-78.5.学位论文【格式】[序号]作者.篇名[D].出版地:保存者,出版年份:起始页码.【举例】[11] 张筑生.微分半动力系统的不变集[D].北京:北京大学数学系数学研究所, 1983:1-7.6.研究报告【格式】[序号]作者. 篇名[R].出版地:出版者,出版年份:起始页码.【举例】[12] 冯西桥.核反应堆压力管道与压力容器的LBB分析[R].北京:清华大学核能技术设计研究院, 1997:9-10. 7.专利【格式】[序号]专利所有者.题名[P].国别:专利号,发布日期.【举例】[13] 姜锡洲.一种温热外敷药制备方案[P].中国专利:881056073, 1989 07 26.8.标准【格式】[序号]标准编号,标准名称[S].【举例】[14] GB/T 16159-1996, 汉语拼音正词法基本规则[S].9.条例【格式】[序号]颁布单位.条例名称.发布日期【举例】[15] 中华人民共和国科学技术委员会.科学技术期刊管理办法[Z].1991-06-0510.电子文献【格式】[序号]主要责任者.电子文献题名.电子文献出处[电子文献及载体类型标识].或可获得地址,发表或更新日期/引用日期.【举例】[16] 王明亮.关于中国学术期刊标准化数据库系统工程的进展[EB/OL].http:///pub/wml.txt/980810 2.html, 1998 08 16/1998 10 04.[17] 万锦.中国大学学报论文文摘(1983 1993).英文版[DB/CD]. 北京: 中国大百科全书出版社, 1996. 11.各种未定义类型的文献【格式】[序号] 主要责任者.文献题名[Z].出版地:出版者, 出版年.特别说明:凡出现在"参考文献"项中的标点符号都失去了其原有意义,且其中所有标点必须是半角,如果你的输入法中有半角/全解转换,则换到半角状态就可以了,如果你的输入法中没有这一转换功能,直接关闭中文输入法,在英文输入状态下输入即可.其实,很多输入法(如目前比较流行的搜狐输入法)都提供了四种组合:(1)中文标点+ 全角:这时输入的标点是这样的,:【1】-(而这时,我没有找到哪个键可以输入/ 符号)也就是说,这些符号是一定不能出现在"参考文献"中的;(2)中文标点+半角:这时输入的标点是这样的,:【1】-(这时,我还是没有找到哪个键可以输入/ 符号)也就是说,这些符号也不能出现在"参考文献"中的;上面列出的符号,中间没有任何的空格,你能看出它们有什么区别吗?我看只是-的宽度有一点点不同,其它都一样(3)英文标点+全角:这时输入的标点是这样的,.:[1]-/(4)英文标点+半角:这时输入的标点是这样的,.:[1]-/从这两项可以明显的看出,半角和全角其实最大的差别是所占的宽度不一样,这一点对于数字来说最为明显,而英文标点明显要比中文标点细小很多(也许因为英文中,标点的功能没有中文那么复杂,就是说英文中标点符号的能力没有中文那么强大)所以,很多人在写"参考文献" 时,总是觉得用英文标点+半角很不清楚,间距也太小,其实这点完全不用担心如果你觉得真的太小不好看,就用英文标点+全角吧而在[1] 之后,一般也都有一个空格更为详细的内容,大家可以从附件中下载国家标准《文后参考文献著录规则GB/T 7714-2005》查看,不过,很长很烦,拿出点耐心看吧对于英文参考文献,还应注意以下两点:①作者姓名采用"姓在前名在后"原则,具体格式是:姓,名字的首字母. 如:Malcolm Richard Cowley 应为:Cowley, M.R.,如果有两位作者,第一位作者方式不变,&之后第二位作者名字的首字母放在前面,姓放在后面,如:Frank Norris 与Irving Gordon应为:Norris, F. & I.Gordon.②书名、报刊名使用斜体字,如:Mastering English Literature,English Weekly.三、注释注释是对论文正文中某一特定内容的进一步解释或补充说明注释应置于本页页脚,前面用圈码①、②、③等标识国外英文参考文献规范(国外英文参考文献格式)国外英文参考文献规范(国外英文参考文献格式)在处理英文参考文献时应尽量参考英语原版书的常用格式,以符合英语惯用法,使自己写的英语看起来像英美人写的英语。

29.Modeling_growth_rate

29.Modeling_growth_rate

Modeling growth rate of HfO 2thin films grown by metal–organicmolecular beam epitaxyMyoung-Seok Kim a ,Young-Don Ko a ,Tae-Houng Moon b ,Jae-Min Myoung b ,Ilgu Yun a,*aDepartment of Electrical and Electronic Engineering,Yonsei University,134Shinchon-Dong,Seodaemun-Gu,Seoul 120-749,South Korea bDepartment of Materials Science and Engineering,Yonsei University,134Shinchon-Dong,Seodaemun-Gu,Seoul 120-749,South KoreaReceived 30December 2004;received in revised form 6April 2005;accepted 12April 2005Available online 13June 2005AbstractHfO 2dielectric layers were grown on the p-type Si (100)substrate by metal–organic molecular beam epitaxy (MOMBE).Hafnium-tetra-butoxide,Hf(O $t -C 4H 9)4was used as a Hf precursor and Argon gas was used as a carrier gas.The thickness of the HfO 2film and intermediate SiO 2layer were measured by scanning electron microscopy (SEM)and high-resolution transmission electron microscopy (HRTEM).The properties of the HfO 2layers were evaluated by X-ray diffraction (XRD),X-ray photoelectron spectroscopy (XPS),high frequency (HF)capacitance–voltage (C–V)measurement,and current–voltage (I–V)measurement.C–V and I–V measurements have shown that HfO 2layer grown by MOMBE has a high dielectric constant (k )of 20–22and a low-level of leakage current density.The growth rate is affected by various process variables such as substrate temperature,bubbler temperature,Ar and O 2gas flows and growth time.Since the ratio of O 2and Ar gas flows are closely correlated,the effect of variations in O 2/Ar flow ratio on growth rate is also investigated using statistical modeling methodology.q 2005Elsevier Ltd.All rights reserved.Keywords:HfO 2;Gate dielectric;Thin film;Process modeling;MOMBE1.IntroductionThe rapid progress of complementary metal-oxide-semiconductor (CMOS)integrated circuit technology since the late 1980s resulted in the reduction of the gate oxide thickness.With the reduction of gate dielectric thickness to a few nanometers,higher dielectric constant material than the conventional SiO 2(i.e.high-k material)is needed to overcome the problem of an exponential increase in the leakage current level due to direct tunneling.Many researches for high-k dielectric materials such as ZrO 2,Ta 2O 5,Al 2O 3,HfO 2,TiO 2,silicates (ZrSi x O y and HfSi x O y ),STO,and BST have been performed [1–5].Among these candidates,HfO 2is one of the most highlighted high-k gate insulators because of its high dielectric constant (25–30),wide band-gap energy (5.68eV),high breakdown field (15–20MV/cm 2),and good thermal stability on Si substrate.The growth method of high-k materials is also an important factor to determine the property of gate dielectric layer.In our experiments,we searched appro-priate experimental conditions and examined the character-istics of HfO 2films using metal–organic molecular beam epitaxy (MOMBE)system.MOMBE is one of the powerful techniques obtaining abrupt interface and controlled thickness of films,mainly due to source evaporation at a controlled rate under ultra high vacuum condition [6].In this paper,the relationships between the growth rate of HfO 2films grown by MOMBE and process variables of MOMBE systems are investigated.The electrical characteristics of HfO 2films are investigated by high frequency (HF)capacitance–voltage (C–V)and current–voltage (I–V)measurements.X-ray photoelectron spectroscopy (XPS),scanning electron microscopy (SEM),high-resolution transmission electron microscopy (HRTEM),and X-ray diffraction (XRD)were also performed to analyze the HfO 2films grown by MOMBEsystems.Microelectronics Journal 37(2006)98–106/locate/mejo0026-2692/$-see front matter q 2005Elsevier Ltd.All rights reserved.doi:10.1016/j.mejo.2005.04.055*Corresponding author.Tel.:C 8221234619;fax:C 823132879.E-mail address:iyun@yonsei.ac.kr (I.Yun).2.ExperimentHfO2thinfilm was grown on a p-type Si(100)substrate, of which the native oxide was chemically eliminated by (50:1)H2O/HF solution prior to growth by MOMBE. Hafnium-tetra-butoxide[Hf(O$t-C4H9)4]was chosen as the MO precursor because it has a appropriate vapor pressure and relatively low decomposition temperature.High-purity (99.999%)oxygen gas was used as the oxidant.Hf-t butoxide was introduced into the main chamber using Ar as a carrier gas through a bubbling cylinder.The bubbler was maintained at a constant temperature to supply the constant vapor pressure of Hf-source.The apparatus of the system is schematically shown in Fig.1.High-purity Ar carrier gas passed through the bubbler containing the Hf-source.The gas line from the bubbler to the nozzle was heated to the same temperature.The mixture of Ar and metal–organic gases heated at the tip of the nozzleflows into the main chamber.The introduced Hf-source decomposed into Hf and ligand parts when it reached a substrate maintained at high temperature and the Hf ion was combined with O2gas supplied from another nozzle.The base pressure and working pressure were w10K9and w10K7Torr,respect-ively.Annealing at7008C for2min was carried out following the growth of thefilms to diminish the density of the interfacial charged particles[7].Detailed experimental conditions are listed in Table1.3.Result and discussionThe metal-oxide-semiconductor(MOS)capacitor struc-ture Au/HfO2/p-type Si was fabricated to measure its electrical characteristics.The high frequency(1MHz)C–V curve and I–V curve of HfO2films grown by MOMBE are shown in Fig.2.From measured capacitance and physical thickness,the dielectric constant and the equivalent oxide thickness can be obtained.Actually,the total capacitance must be calculated as series connection of HfO2layer capacitance and SiO2layer capacitance because an unexpected SiO2layer between HfO2and p-Si substrate shown in Fig.3affects the total capacitance strongly[8]. The thickness of SiO2layers was measured by HRTEM and had a value of10–20A˚with experimental conditions.It can be identified by TEM images that the formation of SiO2was affected by experimental conditions such as the growth time [9],O2/Ar gas ratio,and substrate temperature(not shown in here).The formation of SiO2layer is attributed to the excess oxygen during thefilm growth and should be eliminated or minimized because of its low dielectric constant.In addition,HfO2samples grown at substrate temperature of 4008C showed an amorphous nature in Fig.3(a).However, when the substrate temperature was at4508C,a large portion of the HfO2film was crystallized and some of the grain boundaries were observed in Fig.3(b).Ignoring the depletion region effect,the dielectric constant(k Z20–22) can be calculated.This relative low dielectric constantFig.1.Schematic illustration of MOMBE system.Table1Experimental conditionsProcess variables RangeSubstrate temperature300–6008CBubbler temperature70–1308CNozzle temperature2708C(Fixed)Base pressure10K9TorrWorking pressure10K7TorrGasflow(Ar)2–6sccmGasflow(O2)1–8sccmGrowth time20,30,40,50,60minRapid thermal annealing temp.7008C(2min in N2ambient) M.-S.Kim et al./Microelectronics Journal37(2006)98–10699compared with the bulk HfO 2dielectric constant (k Z 25–30)may be due to the polycrystalline nature of the grown HfO 2films and Hf–silicate interfacial layer (not identified in HRTEM image but other research showed that HfO 2/SiO 2interface are marginally unstable with respect to formation of silicates [10])that has lower dielectric constant compared with the bulk HfO 2.Generally,the thickness resulted from C–V curve is slightly higher than that from SEM.This may be due to two major quantum effects:(i)additional band bending because of surface electrons (or holes)above the edge of conduction band and (ii)presence of charge centroid [7].Fixed oxide charge density (w 8!1011/cm K 2)and interface state density (w 1!1012eV K 1cm K 2),attributing to the interface trap density in the oxide layer,were calculated using the flat-band voltage shift [11],and Lehovec’s equation [12].The level of leakage current densities of HfO 2film at positive gate bias (0w 2V)and at negative gate bias (0w K 2V)is approximately 10K 9A/cm 2and 10K 7A/cm 2,respectively,which are relatively low levels compared with that of the conventional SiO 2.Through screening experiments at different experimental conditions,it was identified that the growth rate of HfO 2films grown by MOMBE system was affected by many process variables.Among these process variables,the substrate temperature,bubbler temperature,growth time,Ar and O 2gas flow were chosen as main process variables to determine the properties and the growth rate of HfO 2films.It is shown the relationship between the growth rate and the substrate temperature in Fig.4.Through experiments in various conditions,we have found that there were two regions showing a different trend in growth rate (region a and region b).At below 3008C,there has been found almost no growth of HfO 2layer.When the substrate temperature reached at 4508C,the growth rate had a maximum value,and decreased with increasing substrate temperature until at 6008C.To investigate the region a,SEM images of HfO 2films grown by MOMBE at different substrate temperature (3008C w 4508C)were shown in Fig.5.As shown in Fig.5(a),it was found that HfO 2film was not grown under the temperature of 3008C because Hf source was not effectively decomposed.At the temperature of 3508C,HfO 2films began to grow,and the crystallization of HfO 2films were enhanced as the substrate temperature increased,shown in Fig.5(b)–(d).In our MOMBE system,HfO 2filmsFig.2.High frequency C–V and I–V characteristics of HfO 2film grown by MOMBE:(a)C–V curve and (b)I–Vcurve.Fig.3.HRTEM images of HfO 2film grown by MOMBE:(a)4008C and (b)4508C.M.-S.Kim et al./Microelectronics Journal 37(2006)98–106100grown at4508C of the substrate temperature showed the thickfilm thickness(rge growth rate)and fully crystallized structure.In order to examine the region b, XRD spectra of HfO2films grown at450and5508C of the substrate temperature are shown in Fig.6.HfO2films have been found to exist in monoclinic phase,tetragonal phase,cubic phase,and amorphous structure[13–16].This crystal structure depends on the growth method and experimental condition of HfO2films.It was found that there were monoclinic(m)and tetragonal(t)phase in our HfO2films grown by MOMBE system,and monoclinic phase(m)was dominant.Through XRD spectra,it was shown that the crystal structure of HfO2films was changed with the substrate temperature.At4508C,various crystal phase and crystal direction were shown in XRD spectrum. However,at5508C,the overall XRD spectrum became simple and some peaks disappeared.So,it can be assumed that the excess thermal energy from higher substrate temperature does not contribute to the growth offilms but only change the crystal structure of HfO2films.Another explanation for region b is related to the small crystallite size at higher substrate temperature.From each XRD spectrum,the crystallite size of HfO2films in different substrate temperature was ing Scherrer’s formula[17],the calculated crystallite size is approximately 7.2nm at4508C and4.6nm at5508C,respectively.It shows that the substrate temperature beyond a certain level limits the growth of crystallite size.In other research related to ZrO2film[18],it was shown that the incorporation of the hydrocarbon-rich environments limits the crystallite size. The decomposition rate of Hf-t-butoxide source increases with increasing substrate temperature.Fully decomposed source due to the high substrate temperature makes a chamber of MOMBE the hydrocarbon-rich circumstances. The incorporation of the hydrocarbon-rich circumstances limits the crystallite size,andfinally limits the growth rate. If we remind that HfO2and ZrO2is II–VI compound which have similar properties,this explanation can be persuasive. In addition,the observed XRD peak locations are not shifted to any particular direction,so it is possible tosuppose Fig.5.SEM images of HfO2films at different substrate temperature:(a)3008C,(b)3508C,(c)4008C,and(d)4508C.that there is no uniform tensile or compressive stress in the film[17].The relationship between the growth rate and the bubbler temperature is shown in Fig.7.It shows that the growth rate of HfO2films is a linear function of the bubbler temperature. If we remind of the fact that the bubbler temperature in MOMBE process is the major control parameter of transferred volume of Hf precursor,this phenomenon can be easily understood.The relationships between the growth rate and gasflows are shown in Figs.8and9.It is shown that the growth rate of HfO2films increases with increasing gasflow of carrier gas (Ar).They also showed that the growth rate of HfO2films increases with increasing gasflow of oxidant(O2).But the tendency of increasing growth rate is not linear,and the effects of O2/Ar gasflow ratio must be also taken into account[19,20].As the O2/Ar ratio was varied,an ionic species that dominate the deposition process can be also varied,and it can affect characteristics of HfO2films such as dielectric constant,surface morphology,fixed charge concentration and the growth rate.To investigate the effect of O2/Ar gas ratio more,XPS analysis was performed. Fig.10(a)shows the XPS spectra for Hf4f level that were calibrated from C1s peak at284.5eV.It was shown each spectrum at different O2gasflow(2–8sccm and Ar gasflow wasfixed at2sccm).As shown in Fig.10(a),the Hf4f5/2 and Hf4f7/2peaks,which have binding energies of16.05eV and17.76eV,respectively,related to Hf–O bonding in HfO2,shifted to the higher binding energy with increasing of O2gasflow.The origins of binding energy shiftare Fig.7.Growth rate of HfO2films in different bubblertemperature.Fig.8.Growth rate of HfO2films in different gasflow(Ar).Fig.9.Growth rate of HfO2films in different gasflow(O2). Fig.6.XRD spectra of HfO2films at different substrate temperature(450and5508C).M.-S.Kim et al./Microelectronics Journal37(2006)98–106102suggested as a number of factors such as charge transfer effect,presence of electric field,environmental charge density,and hybridization.Among these,charge transfer is regarded as a dominant mechanism causing a binding energy shift.According to the charge transfer mechanism,removing an electron from the valence orbital generates increase in core electron’s potential and finally leads a chemical binding energy shift [21].Therefore,it is considered that the Hf 4f 5/2and Hf 4f 7/2peaks shift (D BE Z 0.6eV)originated from the enhanced charge transfer with increasing O 2gas flow,i.e.the larger portion of Hf atoms was fully oxidized with increasing O 2gas flow.Fig.10(b)shows the O 1s core level peaks also demonstrated binding energy shift with changing of O 2gas flow.Each peak can be split into two sub-peaks by Gaussian fitting which represent the Hf–O bonding at w 531eV and O–C or O–Si bonding at w 532.5eV [22,23].The relative quantities of Hf and O elements incorporated in the layer can be obtained by comparing the areas of Hf 4f peak and O 1s sub-peak for Hf–O bonding.The relationship between growth time and growth rate is shown in Fig.11.Growth time is critical and direct factor to determine the growth rate.Through the Fig.11,it can be identified that the growth rate increases with increasing growth time.It shows that growth rate and growth time have the linear relationship over 20min of growth time.However,there were no experimental results about the growth rate below the 20min of growth time.Considering other research related with the growth rate of thin films,it can be assumed that the growth rate has a logarithm function of growth time,and more experiments for the growth rate below 20min of the growth time willbeFig.10.XPS spectra of HfO 2film grown by MOMBE:(a)Hf and (b)O.Fig.11.Growth rate of HfO 2films at different growthtime.Fig.12.XRD spectra of HfO 2films at different growth time.M.-S.Kim et al./Microelectronics Journal 37(2006)98–106103needed to fully understand the relationship between the growth rate and the growth time.XRD spectra and SEM images of HfO 2films grown at different growth time are shown in Figs.11and 12.It can be also identified that the growth time affects the crystal structure of HfO 2films (Fig.13).The growth rate in the deposition process is one of the important factors,which can determine optical and electrical characteristics.Therefore,the statistical modeling methodology is used to predict the response factor and analyze the relationship between input factors to reduce the time and the cost for the manufacturing process.Prior to the modeling,the experimental design matrix is generated by using factorial design to reduce the runs of the experiment and then add to the additional points,center point and star point,to cover the weak ranges for the predicted model [24].Response surface methodology (RSM)is a collection of mathematical and statistical techniques that are useful for the modeling and analysis of problem in which a response is influenced by several variables and the objective is to optimize the response [25].Response surface models maybe represented as the full quadratic model:y Z b 0CX n i Z 1b i X i CX n j Z i C 1X n i Z 1b ij X i X j CX n i Z 1b ij X 2i Where y is the response variable,n is the number ofindependent process factor,b are model coefficient,and X i are process factor values.The order in which experiments were performed has been randomized to avoid statistically the effect of irrelevant factors,which may be present,but not considered in this paper.The residual plot describes the difference between the predicted value and the measured value.One of the assumptions of this analysis is that the residuals are both normally and randomly distributed with zero mean and unit variance.The predicted model based on the statistical methodology can characterize the certain process factors without many experiments and identify the accuracy and validity of our model.In our paper,we performed the statistical modeling about the growth rate in different O 2/Ar gas ratio to verify the effect of gas flows on growth rate.The design matrix using the 2-factor factorial method and additional one center point and two star points is summarized in Table 2.At first,we performed the analysis of variance (ANOVA)to calculate the P -values that indicate whether the process variables are statistically significant on the response.Based in theresultsFig.13.SEM images of HfO 2films at different growth time (30,40,and 50min.)Table 2Design of experiment for the effect of O 2/Ar ratio Run O 2flow rate Ar flow rate 1K 1K 12K 112-factor31K 1Factorial design 411500Center point 601Star point 7K 1Star pointP -value of the simple regression modelTable 3Results of experimental design Factor Statistical significance Growth rate O 2gas flow 0.0027Ar gas flow0.0185M.-S.Kim et al./Microelectronics Journal 37(2006)98–106104shown in Table 3,it is verified that O 2and Ar gas flows are statistically significant on the growth rate.In addition,we calculated the regression coefficients by means of the least square estimation method.In this way,we could extract the simple linear regression model for O 2/Ar ratio effect on the growth rate of HfO 2film.Fig.14shows that the plot of residual versus fitted value.We assumed that the residuals are both normally and randomly distributed.[24].It is observed that for randomized run orders,the residuals are scattered about zero and there is no special features or patterns in residuals.The contour plot of response surface is shown in Fig.15.The plot shows that when the one of gases (Ar or O 2)is insufficient compared to the other,the growth rate is restricted,and when both gas flows are sufficient,the growth rate has a high level.Also we could identify that Argas flow and O 2gas flow are closely correlated.When O 2/Ar ratio is about unity,the growth rate had a tendency to increase steeply.4.ConclusionThe characteristics of the HfO 2dielectric layer on the p-type Si substrate by MOMBE process were investigated.HfO 2films grown by MOMBE had a high dielectric constant (k Z 20–22).It was identified that there existed the fixed oxide charge (w 8!1011cm K 2)and interface state density (w 1!1012eV K 1cm K 2)in the HfO 2layer.Leakage current density of HfO 2film grown by MOMBE is about 10K 9–10K 7A/cm 2in K 2to 2V gate voltage.It is observed that the growth rate of HfO 2film was affected by the substrate temperature due to changing of growth direction,crystal structure,and crystallite size.It was also revealed that the HfO 2film grown by MOMBE had monoclinic and tetragonal phase and there was no uniform strain.The growth rate increased with increasing the bubbler tempera-ture.The effects of gas flows on the growth rate were also investigated by SEM and XPS.It is observed that the low level of gas flows limited the growth of films.In addition,it was founded that the growth time greatly affected the growth rate of HfO 2fiing the statistical modeling methodology,it can be concluded that the low level of gas flows limited the growth rate,and the growth rate was appeared to be increased when O 2gas flow rate is almost comparable to Ar gas flow rate with sufficient flowrate.Fig.14.Residual plot of simple regressionmodel.Fig.15.Contour plot of response surface for growth rate.M.-S.Kim et al./Microelectronics Journal 37(2006)98–106105AcknowledgementsThis research was supported by the MIC(Ministry of Information and Communication),Korea,under the ITRC (Information Technology Research Center)support pro-gram supervised by the IITA(Institute of Information Technology Assessment)References[1]M.Copel,M.Gribelyuk,E.Gusev,Applied Physics Letter76(4)(2000)436.[2]G.B.Alers,R.M.Fleming,Y.H.Wong,B.Dennis,A.Pinczuk,G.Redinbo,R.Urdahl,E.Ong,Z.Hasan,Applied Physics Letter72(11)(1998)1308.[3]Stefan Jakschik,Uwe Schroeder,Thomas Hecht,Dietmar Krueger,Guenther Dollinger,Andreas Bergmaier,Claudia Luhmann,Johann W.Bartha,Journal of Applied Surface211(2003)352.[4]Q.Fang,J.Y.Zhang,Z.M.Wang,J.X.Wu, B.J.O’Sullivan,P.K.Hurley,T.L.Leedham,H.Davies,M.A.Audier,C.Jimenez, J.-P.Senateur,Ian W Boyd,Thin Solid Films427(2003)391.[5]J.P.Chang,Y.S.Lin,Appled Physics Letter79(22)(2001)3666.[6]Q.Fang,Organometallic Vapor-Phase Epitaxy:Theory and Practice,second ed,Academic Press,San Diego,USA,1999.[7]Alvin Chi-hai Ng,Jun Xu,J.B.Xu,W.Y.Cheung,Electron DeviceMeeting,IEEE Hong-Kong,2001,101.[8]G.D.Wilk,R.M.Wallace,J.M.Anthony,Journal of Applied Physics89(2001)5243.[9]Tae-Hyoung Moon,Moon-ho Ham,Myoung-Seok Kim,Ilgu Yun,Jae-Min Myoung,Applied Surface Science240(2005)105.[10]Maciej Gusowski,John E.Jaffe,Chun-Li Liu,Matt Stoker,RamaI.Hegde,Raghaw S.Rai,Philip J.Tobin,Applied Physics Letter80(11)(2002)1897.[11]Q.Fang,E.H.Nicollian,MOS(metal-oxide-semiconductor)physicsand technology,Wiley Interscience,New York,1981.[12]K.Lehovec,A.Slobodskoy,Solid-State Electronics7(1964)59.[13]M.Alvisi,S.Scaglione,S.Martelli,A.Rizzo,L.Vasanelli,Thin SolidFilms354(1999)19.[14]K.Kukli,J.Ihanus,Applied Physics Letter68(1996)3737.[15]J.Wang,Journal of Material Science27(1992)5397.[16]J.Aarik,A.Aidla,Applied Surface Science173(2001)15.[17]K.Lehovec,B.D.Cullity,S.R.Stock,X-ray diffraction,third ed.,Prentice Hall,Englewood Cliffs,NJ,2001.[18]K.Lehovec,Byeong-Ok Cho,Jane P.Chang,Jae-Ho Min,SangH.Moon,Yil W.Kim,Igor Levin,Journal of Applied Physics93(1)(2003)745.[19]K.Lehovec,Byeong-Ok Cho,Jianjun Wang,Jane P.Chang,Journalof Applied Physics92(8)(2002)4238.[20]K.Lehovec,Jong-ho Yun,Shi-Woo Rhee,Thin solid Films292(1997)324.[21]P.S.Bagus,F.Illas,G.Paccghioni,F.Parmigiani,Journal of ElectronSpectroscopy Related and Phenomena100(1999)215.[22]H.Y.Yu,X.D.Feng,D.Grozea,Z.H.Lu,R.N.S.Sodhi,A.M.Hor,H.Aziz,Applied Physics Letter78(2001)2595.[23]J.P.Chang,Y.S.Lin,Journal of Applied Physics90(2001)2964.[24]K.Lehovec,Introduction:Design and analysis of experiments,fifthed.,Wiley,New York,2001.[25]K.Lehovec,R.H.Myers, D.C.Montgomery,Response SurfaceMethodology,Wiley,New York,1995.M.-S.Kim et al./Microelectronics Journal37(2006)98–106 106。

Microstrip Antenna

Microstrip Antenna

1An Introduction to Microstrip Antennas1.1IntroductionDeschamps first proposed the concept of the MSA in1953[1].However, practical antennas were developed by Munson[2,3]and Howell[4]in the 1970s.The numerous advantages of MSA,such as its low weight,small volume,and ease of fabrication using printed-circuit technology,led to the design of several configurations for various applications[5–9].With increas-ing requirements for personal and mobile communications,the demand for smaller and low-profile antennas has brought the MSA to the forefront.An MSA in its simplest form consists of a radiating patch on one side of a dielectric substrate and a ground plane on the other side.The top and side views of a rectangular MSA(RMSA)are shown in Figure1.1.However, other shapes,such as the square,circular,triangular,semicircular,sectoral, and annular ring shapes shown in Figure1.2,are also used.Radiation from the MSA can occur from the fringing fields between the periphery of the patch and the ground plane.The length L of the rectangular patch for the fundamental TM10mode excitation is slightly smaller than␭/2,where␭is the wavelength in the dielectric medium,which in terms of free-space wavelength␭o is given as␭o/√⑀e,where⑀e is the effective dielectric constant of a microstrip line of width W.The value of ⑀e is slightly less than the dielectric constant⑀r of the substrate because thefringing fields from the patch to the ground plane are not confined in the12Broadband Microstrip AntennasFigure1.1MSA configuration.Figure1.2Different shapes of microstrip patches.dielectric only,but are also spread in the air.To enhance the fringing fields from the patch,which account for the radiation,the width W of the patch is increased.The fringing fields are also enhanced by decreasing the⑀r or by increasing the substrate thickness h.Therefore,unlike the microwave integrated circuit(MIC)applications,MSA uses microstrip patches with larger width and substrates with lower⑀r and thicker h.The details of various substrates used for MSA are given in Appendix A.For MSA applications in the microwave frequency band,generally h is taken greater than or equal to 1/16th of an inch(0.159cm).A typical comparison of MSA with MIC in the microwave frequency range is given in Table1.1.An Introduction to Microstrip Antennas3Table1.1A Comparison of MIC and MSAMIC MSAh≤0.159cm≥0.159cm⑀r≥9.8≤9.8W Small LargeRadiation Minimized Maximized1.2Characteristics of MSAsThe MSA has proved to be an excellent radiator for many applications because of its several advantages,but it also has some disadvantages.The advantages and disadvantages of the MSA are given in Sections1.2.1and 1.2.2.1.2.1AdvantagesMSAs have several advantages compared to the conventional microwave antennas.The main advantages of MSAs are listed as follows:•They are lightweight and have a small volume and a low-profileplanar configuration.•They can be made conformal to the host surface.•Their ease of mass production using printed-circuit technology leadsto a low fabrication cost.•They are easier to integrate with other MICs on the same substrate.•They allow both linear polarization and CP.•They can be made compact for use in personal mobile communica-tion.•They allow for dual-and triple-frequency operations.1.2.2DisadvantagesMSAs suffer from some disadvantages as compared to conventional micro-wave antennas.They are the following:•Narrow BW;•Lower gain;•Low power-handling capability.4Broadband Microstrip AntennasMSAs have narrow BW,typically1–5%,which is the major limiting factor for the widespread application of these antennas.Increasing the BW of MSAs has been the major thrust of research in this field,and broad BW up to70%has been achieved[9,10].Various broadband MSA configurations are summarized in this chapter,and they are detailed in the following chapters.1.2.3Applications of MSAsThe advantages of MSAs make them suitable for numerous applications [6–11].The telemetry and communications antennas on missiles need to be thin and conformal and are often MSAs.Radar altimeters use small arrays of microstrip radiators.Other aircraft-related applications include antennas for telephone and satellite communications.Microstrip arrays have been used for satellite imaging systems.Patch antennas have been used on commu-nication links between ships or buoys and satellites.Smart weapon systems use MSAs because of their thin profile.Pagers,the global system for mobile communication(GSM),and the global positioning system(GPS)are major users of MSAs.Some of the applications of MSAs are listed in Table1.2.1.3Feeding TechniquesThe MSA can be excited directly either by a coaxial probe or by a microstrip line.It can also be excited indirectly using electromagnetic coupling or aperture coupling and a coplanar waveguide feed,in which case there is noTable1.2Typical Applications of MSAsSystem ApplicationAircraft and ship antennas Communication and navigation,altimeters,blindlanding systemsMissiles Radar,proximity fuses,and telemetrySatellite communications Domestic direct broadcast TV,vehicle-basedantennas,communicationMobile radio Pagers and hand telephones,man pack systems,mobile vehicleRemote sensing Large lightweight aperturesBiomedical Applicators in microwave hyperthermiaOthers Intruder alarms,personal communication,and soforthAn Introduction to Microstrip Antennas5direct metallic contact between the feed line and the patch[9–12].Feeding technique influences the input impedance and characteristics of the antenna, and is an important design parameter.The coaxial or probe feed arrangement is shown in Figure1.1.The center conductor of the coaxial connector is soldered to the patch.The main advantage of this feed is that it can be placed at any desired location inside the patch to match with its input impedance.The disadvantages are that the hole has to be drilled in the substrate and that the connector protrudes outside the bottom ground plane,so that it is not completely planar.Also, this feeding arrangement makes the configuration asymmetrical.A patch excited by microstrip line feed is shown in Figure1.3(a).This feed arrangement has the advantage that it can be etched on the same substrate,so the total structure remains planar.The drawback is the radiation from the feed line,which leads to an increase in the cross-polar level.Also, in the millimeter-wave range,the size of the feed line is comparable to the patch size,leading to increased undesired radiation.For thick substrates,which are generally employed to achieve broad BW,both the above methods of direct feeding the MSA have problems.In the case of a coaxial feed,increased probe length makes the input impedance more inductive,leading to the matching problem.For the microstrip feed, an increase in the substrate thickness increases its width,which in turn increases the undesired feed radiation.The indirect feed,discussed below, solves these problems.An electromagnetically coupled RMSA is shown in Figure1.3(b).The electromagnetic coupling is also known as proximity coupling[9,12,13].The feed line is placed between the patch and the ground plane,which is separated by two dielectric media.The advantages of this feed configuration include the elimination of spurious feed-network radiation;the choice between two different dielectric media,one for the patch and the other for the feed line to optimize the individual performances; and an increase in the BW due to the increase in the overall substrate thickness of the MSA.The disadvantages are that the two layers need to be aligned properly and that the overall thickness of the antenna increases.Another method for indirectly exciting a patch employs aperture cou-pling[14].In the aperture-coupled MSA configuration,the field is coupled from the microstrip line feed to the radiating patch through an electrically small aperture or slot cut in the ground plane,as shown in Figure1.3(c). The coupling aperture is usually centered under the patch,leading to lower cross-polarization due to symmetry of the configuration.The shape,size, and location of the aperture decide the amount of coupling from the feed line to the patch[15–17].The slot aperture can be either resonant or6Broadband Microstrip AntennasFigure1.3Rectangular MSA fed by(a)microstrip line,(b)electromagnetic coupling,(c)aperture coupling,and(d)coplanar waveguide(CPW).An Introduction to Microstrip Antennas7nonresonant[10,11].The resonant slot provides another resonance in addition to the patch resonance thereby increasing the BW at the expense of an increase in back radiation.As a result,a nonresonant aperture is normally used.The performance is relatively insensitive to small errors in the alignment of the different layers.Similar to the electromagnetic coupling method,the substrate parameters of the two layers can be chosen separately for optimum antenna performance.This feeding method gives increased BW as described in Chapter4.The coplanar waveguide feed,shown in Figure1.3(d),has also been used to excite the MSA[18].In this method,the coplanar waveguide is etched on the ground plane of the MSA.The line is excited by a coaxial feed and is terminated by a slot,whose length is chosen to be between0.25 and0.29of the slot wavelength.The main disadvantage of this method is the high radiation from the rather longer slot,leading to the poor front-to-back ratio.The front-to-back ratio is improved by reducing the slot dimension and modifying its shape in the form of a loop[19].1.4Methods of AnalysisThe MSA generally has a two-dimensional radiating patch on a thin dielectric substrate and therefore may be categorized as a two-dimensional planar component for analysis purposes.The analysis methods for MSAs can be broadly divided into two groups.In the first group,the methods are based on equivalent magnetic current distribution around the patch edges(similar to slot antennas).There are three popular analytical techniques:•The transmission line model;•The cavity model;•The MNM.In the second group,the methods are based on the electric current distribution on the patch conductor and the ground plane(similar to dipole antennas,used in conjunction with full-wave simulation/numerical analysis methods).Some of the numerical methods for analyzing MSAs are listed as follows:8Broadband Microstrip Antennas•The method of moments(MoM);•The finite-element method(FEM);•The spectral domain technique(SDT);•The finite-difference time domain(FDTD)method.This section briefly describes these methods.1.4.1Transmission Line ModelThe transmission line model is very simple and helpful in understanding the basic performance of a MSA.The microstrip radiator element is viewed as a transmission line resonator with no transverse field variations(the field only varies along the length),and the radiation occurs mainly from the fringing fields at the open circuited ends.The patch is represented by two slots that are spaced by the length of the resonator.This model was originally developed for rectangular patches but has been extended for generalized patch shapes.Many variations of this method have been used to analyze the MSA[9,20–22].Although the transmission line model is easy to use,all types of configu-rations can not be analyzed using this model since it does not take care of variation of field in the orthogonal direction to the direction of propagation.1.4.2Cavity ModelIn the cavity model,the region between the patch and the ground plane is treated as a cavity that is surrounded by magnetic walls around the periphery and by electric walls from the top and bottom sides.Since thin substrates are used,the field inside the cavity is uniform along the thickness of the substrate[23–25].The fields underneath the patch for regular shapes such as rectangular,circular,triangular,and sectoral shapes can be expressed as a summation of the various resonant modes of the two-dimensional resonator.The fringing fields around the periphery are taken care of by extending the patch boundary outward so that the effective dimensions are larger than the physical dimensions of the patch.The effect of the radiation from the antenna and the conductor loss are accounted for by adding these losses to the loss tangent of the dielectric substrate.The far field and radiated power are computed from the equivalent magnetic current around the periphery.An alternate way of incorporating the radiation effect in the cavity model is by introducing an impedance boundary condition at the walls of the cavity.The fringing fields and the radiated power are not included insideAn Introduction to Microstrip Antennas9the cavity but are localized at the edges of the cavity.However,the solution for the far field,with admittance walls is difficult to evaluate[9].1.4.3MNMThe MNM for analyzing the MSA is an extension of the cavity model[9, 26,27].In this method,the electromagnetic fields underneath the patch and outside the patch are modeled separately.The patch is analyzed as a two-dimensional planar network,with a multiple number of ports located around the periphery.The multiport impedance matrix of the patch is obtained from its two-dimensional Green’s function.The fringing fields along the periphery and the radiated fields are incorporated by adding an equivalent edge admittance network.The segmentation method is then used to find the overall impedance matrix.The radiated fields are obtained from the voltage distribution around the periphery.Appendix C details this method.The above three analytical methods offer both simplicity and physical insight.In the latter two methods,the radiation from the MSA is calculated from the equivalent magnetic current distribution around the periphery of the radiating patch,which is obtained from the corresponding voltage distribution.Thus,the MSA analysis problem reduces to that of finding the edge voltage distribution for a given excitation and for a specified mode. These methods are accurate for regular patch geometries,but—except for MNM with contour integration techniques—they are not suited for arbitrary shaped patch configurations.For complex geometries,the numerical tech-niques described below are employed[9].1.4.4MoMIn the MoM,the surface currents are used to model the microstrip patch, and volume polarization currents in the dielectric slab are used to model the fields in the dielectric slab.An integral equation is formulated for the unknown currents on the microstrip patches and the feed lines and their images in the ground plane[28].The integral equations are transformed into algebraic equations that can be easily solved using a computer.This method takes into account the fringing fields outside the physical boundary of the two-dimensional patch,thus providing a more exact solution.This book makes extensive use of a commercially available software IE3D[29] based on MoM to analyze various MSA configurations.10Broadband Microstrip Antennas1.4.5FEMThe FEM,unlike the MoM,is suitable for volumetric configurations.In this method,the region of interest is divided into any number of finite surfaces or volume elements depending upon the planar or volumetric structures to be analyzed[30].These discretized units,generally referred to as finite elements,can be any well-defined geometrical shapes such as triangular elements for planar configurations and tetrahedral and prismatic elements for three-dimensional configurations,which are suitable even for curved geometry.It involves the integration of certain basis functions over the entire conducting patch,which is divided into a number of subsections.The problem of solving wave equations with inhomogeneous boundary conditions is tackled by decomposing it into two boundary value problems,one with Laplace’s equation with an inhomogeneous boundary and the other corre-sponding to an inhomogeneous wave equation with a homogeneous boundary condition[10].1.4.6SDTIn the SDT,a two-dimensional Fourier transform along the two orthogonal directions of the patch in the plane of substrate is employed.Boundary conditions are applied in Fourier transform plane.The current distribution on the conducting patch is expanded in terms of chosen basis functions, and the resulting matrix equation is solved to evaluate the electric current distribution on the conducting patch and the equivalent magnetic current distribution on the surrounding substrate surface.The various parameters of the antennas are then evaluated[31].1.4.7FDTD MethodThe FDTD method is well-suited for MSAs,as it can conveniently model numerous structural inhomogenities encountered in these configurations[10].It can also predict the response of the MSA over the wide BW witha single simulation.In this technique,spatial as well as time grid for the electric and magnetic fields are generated over which the solution is required. The spatial discretizations along three Cartesian coordinates are taken to be same.The E cell edges are aligned with the boundary of the configuration and H-fields are assumed to be located at the center of each E cell.Each cell contains information about material characteristics.The cells containing the sources are excited with a suitable excitation function,which propagates along the structure.The discretized time variations of the fields are deter-mined at desired ing a line integral of the electric field,the voltage across the two locations can be obtained.The current is computed by a loop integral of the magnetic field surrounding the conductor,where the Fourier transform yields a frequency response.The above numerical techniques,which are based on the electric current distribution on the patch conductor and the ground plane,give results for any arbitrarily shaped antenna with good accuracy,but they are time-consuming.These methods can be used to plot current distributions on patches but otherwise provide little of the physical insight required for antenna design.1.5Review of Various Broadband Techniques for MSAsAs mentioned earlier,the most serious limitation of the MSA is its narrow BW.The BW could be defined in terms of its VSWR or input impedance variation with frequency or in terms of radiation parameters.For the circularly polarized antenna,BW is defined in terms of the axial ratio(AR)[9]. Therefore,before describing the various methods for increasing the BW,the various definitions of the BW are described.1.5.1Definition of BWThe VSWR or impedance BW of the MSA is defined as the frequency range over which it is matched with that of the feed line within specified limits. The BW of the MSA is inversely proportional to its quality factor Q and is given by[32]BW=VSWR−1Q√VSWR(1.1)where VSWR is defined in terms of the input reflection coefficient⌫as:VSWR=1+|⌫|1−|⌫|(1.2)The⌫is a measure of reflected signal at the feed-point of the antenna. It is defined in terms of input impedance Z in of the antenna and the characteristic impedance Z0of the feed line as given below:⌫=Z in −Z 0Z in +Z 0(1.3)The BW is usually specified as frequency range over which VSWR is less than 2(which corresponds to a return loss of 9.5dB or 11%reflected power).Sometimes for stringent applications,the VSWR requirement is specified to be less than 1.5(which corresponds to a return loss of 14dB or 4%reflected power).Conversion of BW from one VSWR level to another can be accomplished byBW 1BW 2=(VSWR 1−1)√VSWR 1√VSWR 2(VSWR 2−1)(1.4)where BW 1and BW 2correspond to VSWR 1and VSWR 2,respectively.The variation of percentage BW for VSWR ≤2and efficiency ␩of a square MSA with normalized substrate thickness h /␭0for two different values of ⑀r (2.2and 10)are given in Figure 1.4(a).Also,the variation of percentage BW with frequency for three commonly used values of h and ⑀r =2.32is given in Figure 1.4(b).The BW of a single-patch antenna increases with an increase in the substrate thickness and a decrease in the ⑀r of the substrate [9,33,34].The BW is approximately 15%for ⑀r =2.2and h =0.1␭0.The ⑀r can be chosen close to 1to obtain a broader BW.The larger thickness of the substrate givesrise to an increase in probe reactance Figure 1.4(a)Variation of percentage BW and efficiency of a square MSA versus h /␭0.(——)⑀r =2.2,(---)⑀r =10and (b)variation of percentage BW with frequency for three values of h and ⑀r =2.32:(——)0.318,(---)0.159,(–-–)0.079cm.for the coaxial feed and the excitation of surface waves,which reduces the efficiency␩of the antenna as can be seen from Figure1.4(a).The efficiency of the MSA is defined in Appendix C.The BW enhancement of the single regularly shaped MSA is discussed in Chapter2.The expressions for approximately calculating the percentage BW of the RMSA in terms of patch dimensions and substrate parameters is given by%BW=Ah␭0√⑀r√W L(1.5)whereA=180forh␭0√⑀r≤0.045A=200for0.045≤h␭0√⑀r≤0.075A=220forh␭0√⑀r≥0.075where W and L are the width and length of the RMSA.With an increase in W,BW increases.However,W should be taken less than␭to avoid excitation of higher order modes.For other regularly shaped patches,values of equivalent W can be obtained by equating the area with that of the RMSA as described in Chapter2[35,36].Another simplified relation for quick calculation of BW(in megahertz) for VSWR=2of the MSA operating at frequency f in gigahertz,with h expressed in centimeters,is given by[37]BW≅50hf2(1.6) The BW can also be defined in terms of the antenna’s radiation parame-ters.It is defined as the frequency range over which radiation parameters such as the gain,half-power beamwidth(HPBW),and sidelobe levels are within the specified minimum and maximum limits.This definition is more complete as it also takes care of the input impedance mismatch,which also contributes to change in the gain.The expression for approximately calculating the directivity D of the RMSA is given byD≅0.2W+6.6+10log΀1.6/√⑀r΁dB(1.7) For other geometries,the values of equivalent W can be obtained by equating its area with that of the RMSA as described in Chapter2[35,36].The above definitions for BW are mainly for a linearly polarized MSA. For a circularly polarized MSA,the BW is generally limited by its AR.This BW is the frequency range over which AR is less than a maximum limit (e.g.,3or6dB).There are various techniques for increasing the BW of the MSAs.The main techniques used to increase the BW are presented briefly in the following sections.1.5.2Modified Shape PatchesThe regular MSA configurations,such as rectangular and circular patches have been modified to rectangular ring[38]and circular ring[39],respectively,to enhance the BW.The larger BW is because of a reduction in the quality factor Q of the patch resonator,which is due to less energy stored beneath the patch and higher radiation.When a U-shaped slot is cut inside the rectangular patch,it gives a BW of approximately40%for VSWR≤2[40]. Similar results are obtained when a U-slot is cut inside a circular or a triangular MSA[41,42].These configurations are discussed in detail in Chapter6.1.5.3Planar Multiresonator ConfigurationsThe planar stagger–tuned coupled multiple resonators yield wide BW in the same way as in the case of multistage tuned circuits.Several configurations are available yielding BW of5–25%[43–49].Various parasitic patches like narrow strips,shorted quarter-wavelength rectangular patches,and rectangu-lar resonator patches have been gap-coupled to the central-fed rectangular patch.Three combinations of gap-coupled rectangular patches are shown in Figure1.5.To reduce the criticality of the gap coupling,direct coupling as depicted in Figure1.6has been used to obtain broad BW.Both gap and direct(hybrid)coupling have been used with circular MSAs(CMSAs)and equilateral triangular MSAs(ETMSAs)to yield broad BW.Figure1.5Various gap-coupled multiresonator RMSA configurations:(a)three RMSAs gap-coupled along radiating edges,(b)three RMSAs gap-coupled along non-radiating edges,and(c)five gap-coupled RMSAs.These planar multiresonator configurations yield broad BW but have the following disadvantages:•The large size,which makes them unsuitable as an array element;•The variation in the radiation pattern over the impedance BW.A modification of the multiresonator patches—to avoid the above-mentioned problems—entails using five or six narrow strips that are gap-coupled along the width[49].This yielded wide BW with a relatively small variation in pattern over the BW.Various broadband planar multiresonator configurations are covered in Chapter3.1.5.4Multilayer ConfigurationsIn the multilayer configuration,two or more patches on different layers of the dielectric substrate are stacked on each other.Based on the coupling mechanism,these configurations are categorized as electromagnetically cou-pled or aperture-coupled MSAs.Figure1.6Various direct-coupled multiresonators:(a)three RMSAs direct-coupled along radiating edges,(b)three RMSAs direct-coupled along nonradiating edges,and(c)five direct-coupled RMSAs.1.5.4.1Electromagnetically Coupled MSAsIn the electromagnetically coupled MSA,one or more patches at the different dielectric layers are electromagnetically coupled to the feed line located at the bottom dielectric layer as shown in Figure1.3(b).Alternatively,one of the patches is fed by a coaxial probe and the other patch is electromagnetically coupled.Either the bottom or top patch is fed with a coaxial probe as shown in Figure1.7.The patches can be fabricated on different substrates,and accordingly the patch dimensions are to be optimized so that the resonance frequencies of the patches are close to each other to yield broad BW.TheseFigure1.7An electromagnetically coupled MSA,in which(a)the bottom patch is fed and(b)the top patch is fed.two layers may be separated by either air-gap or foam yielding BW of15–30% [50–56].1.5.4.2Aperture-Coupled MSAsIn the aperture-coupled MSA,the field is coupled from the microstrip feed line placed on the other side of the ground plane to the radiating patch through an electrically small aperture/slot in the ground plane,as shown in Figure1.3(c).Two different dielectric substrates could be chosen,one for the patch and the other for the feed line to optimize the individual perfor-mances.The coupling to the patch from the feed line can be maximized by choosing the optimum shape of the aperture[14–16].Two patches of rectangular or circular shapes,which are stacked on each other in different dielectric layers yield around30%BW[57–60].A BW of nearly70%has been obtained by stacking patches with resonant apertures[61].The multilayer broadband MSAs,unlike single-layer multiresonator configurations,show a very small degradation in radiation pattern over the complete VSWR BW.The drawback of these structures is the increased height,which is not desirable for conformal applications and increased back radiation for aperture-coupled MSAs.Multilayered configurations using both electromagnetic as well as aperture coupling are described in Chapter4.1.5.5Stacked Multiresonator MSAsThe planar and stacked multiresonator techniques are combined to further increase the BW and gain.A probe-fed single rectangular or circular patch located on the bottom layer has been used to excite multiple rectangular or circular patches on the top layer,respectively[62,63].Besides increasing the BW,these configurations also provide an increase in gain as described in Chapter5.1.5.6Impedance-Matching Networks for Broadband MSAsThe impedance-matching networks are used to increase the BW of the MSA. Some examples that provide about10%BW are the rectangular MSA with a coplanar microstrip impedance-matching network and an electromagnetically coupled MSA with single-stub matching as shown in Figure1.8[12,64–66].1.5.7Log-Periodic MSA ConfigurationsThe concept of log-periodic antenna has been applied to MSA to obtain a multi-octave BW.In this configuration,the patch dimensions are increased。

基于PB相位的等离子体超透镜设计

基于PB相位的等离子体超透镜设计

新技术新工艺2020年第8期基于PB相位的等离子体超透镜设计#夏习成,姚赞(中国科学技术大学精密机械与精密仪器系,安徽合肥230027)摘要:相较于传统光学透镜,超透镜具有对光的可操纵性强、设计灵活和易于集成化等众多优点。

然而基于电介质的超透镜需要亚波长尺度的高深宽比结构,对加工技术要求十分苛刻&提出了一种基于PB相位的等离子体超透镜设计方式,通过控制结构单元光轴方向可以实现对圆偏振光的调控,设计中采用的在金膜上刻蚀矩形孔的方式可以大大降低对加工条件的要求。

基于时域有限差分(FDTD)的仿真计算表明,超透镜的焦距与设计值偏差约为2.5%,焦点半高宽(FWHM)与衍射极限偏差约为2.7%,具有较好的吻合度。

关键词:超表面;超透镜;相位调控;PB相位;时域有限差分;圆偏振光中图分类号:TH74文献标志码:ADesign of Plasma Metalens Based on PB PhaseXIA Xicheng,YAO Zan(Department of Precision Machinery and Instrumentation,University of Science and Technology of China,Hefei230027 ,China) Abstract:Compared with traditional optical lens,the metalens had many advantages such as strong maneuverability to light ,flexible design,easy integration and so on.However,dielectric-based metalens required a high aspect ratio structure atthesub-wavelengthscale whichdemandedhighprocessingtechnology APBphase-basedplasmametalensdesignmethod wasproposed bycontro l ingthedirectionoftheopticalaxisofthestructuralunit itwaspossibletoadjustthephaseofthe circularly polarized light,the method of etching rectangular holes in the gold film used in this design could greatly reduce pDocessingDequiDements.The simulation calculation based on finite di f eDence in time domain(FDTD)showed that the devi-ationofthefocallengthofthemetalensfDomthedesignvaluewasabout2.5%and the deviation between the FWHM with thedi D actionlimitwasabout2.7%ithadagoodagDeement.Keywords:metasuDface metalens phasecontDol PBphase finitedi f eDenceintimedomain ciDculaDlypolaDizedlight传统的聚焦透镜对光线的调控依赖于沿着光路的相位积累,因此会受到自然材料折射率的限制’此外,对制造工艺的要求也会很高,想要加工高精度的透镜十分困难’超表面的优越特性吸引了国内外学者的极大兴趣,其概念最早由Yu等提出,他们提出了一种V形纳米天线组成的超表面山,通过改变天线的开口方向可以实现对圆偏振光的调控,并提出了广义斯涅尔定律来解释。

Memory expansion and chip scale stacking system an

Memory expansion and chip scale stacking system an

专利名称:Memory expansion and chip scale stackingsystem and method发明人:Russell Rapport,James W. Cady,JamesWilder,David L. Roper,James DouglasWehrly, Jr.,Jeff Buchle申请号:US10963867申请日:20041012公开号:US07256484B2公开日:20070814专利内容由知识产权出版社提供专利附图:摘要:The present invention stacks chip scale-packaged integrated circuits (CSPs) intomodules that conserve PWB or other board surface area. In another aspect, the invention provides a lower capacitance memory expansion addressing system and method and preferably with the CSP stacked modules provided herein. In a preferred embodiment in accordance with the invention, a form standard is disposed between the flex circuitry and the IC package over which a portion of the flex circuitry is laid. The form standard provides a physical form that allows many of the varying package sizes found in the broad family of CSP packages to be used to advantage while employing a standard connective flex circuitry design. In a preferred embodiment, the form standard will be devised of heat transference material such as copper to improve thermal performance. In a preferred embodiment, a high speed switching system selects a data line associated with each level of a stacked module to reduce the loading effect upon data signals in memory access. This favorably changes the impedance characteristics exhibited by a DIMM board populated with stacked modules. In a preferred embodiment, FET multiplexers for example, under logic control select particular data lines associated with particular levels of stacked modules populated upon a DIMM for connection to a controlling chip set in a memory expansion system.申请人:Russell Rapport,James W. Cady,James Wilder,David L. Roper,James Douglas Wehrly, Jr.,Jeff Buchle地址:Austin TX US,Austin TX US,Austin TX US,Austin TX US,Austin TX US,Austin TX US 国籍:US,US,US,US,US,US代理机构:Fish & Richardson P.C.更多信息请下载全文后查看。

微位移放大结构英文

微位移放大结构英文

微位移放大结构英文Micro-displacement amplification structures are pretty cool gadgets that help us magnify tiny movements. You know, like when you need to detect the slightest shift or vibration in a device, these structures come into play.Just imagine, if you're working on a project where even the tiniest movement could make a big difference, having a micro-displacement amplifier is like having a superpower.It takes those almost imperceptible movements and boosts them to a level where they can be easily measured and analyzed.One thing I like about these structures is how they can be designed in so many different ways. You can have levers, gears, or even fancy materials that exhibit unique properties when it comes to amplifying small displacements. It's like a puzzle where you get to choose the best pieces to fit your needs.And let's not forget about the applications. From precision engineering to medical devices, these micro-displacement amplifiers are everywhere. They're helping us build better machines, diagnose diseases more accurately, and even improve the performance of everyday gadgets.So in a nutshell, micro-displacement amplification structures are handy tools that allow us to see and measure what was once too small to detect. They're revolutionizing the way we work with tiny movements, and I'm excited to see what new innovations they'll enable in the future.。

Internet www.itwm.fraunhofer.de

Internet www.itwm.fraunhofer.de

J. Kalcsics, S. Nickel, M. SchröderTowards a Unifi ed Territory Design Approach – Applications, Algorithms and GIS IntegrationBerichte des Fraunhofer ITWM, Nr. 71 (2005)© Fraunhofer-Institut für Techno- und Wirtschaftsmathematik ITWM 2005 ISSN 1434-9973Bericht 71 (2005)Alle Rechte vorbehalten. Ohne ausdrückliche, schriftliche Gene h m i g ung des Herausgebers ist es nicht gestattet, das Buch oder Teile daraus in irgendeiner Form durch Fotokopie, Mikrofi lm oder andere Verfahren zu reproduzieren oder in eine für Maschinen, insbesondere Daten v er a r b e i-t ungsanlagen, verwendbare Sprache zu übertragen. Dasselbe gilt für das Recht der öffentlichen Wiedergabe.Warennamen werden ohne Gewährleistung der freien Verwendbarkeit benutzt.Die Veröffentlichungen in der Berichtsreihe des Fraunhofer ITWMkönnen bezogen werden über:Fraunhofer-Institut für Techno- undWirtschaftsmathematik ITWMGottlieb-Daimler-Straße, Geb. 4967663 KaiserslauternGermanyTelefon: +49 (0) 6 31/2 05-32 42Telefax: +49 (0) 6 31/2 05-41 39E-Mail: info@itwm.fraunhofer.deInternet: www.itwm.fraunhofer.deVorwortDas Tätigkeitsfeld des Fraunhofer Instituts für Techno- und Wirt s chafts m a t he m a t ik ITWM um f asst an w en d ungs n a h e Grund l a g en f or s chung, angewandte For s chung so w ie Be r a t ung und kun d en s pe z i fis che Lö s un g en auf allen Gebieten, die für Tech-n o- und Wirt s chafts m a t he m a t ik be d eut s am sind.In der Reihe »Berichte des Fraunhofer ITWM« soll die Arbeit des Instituts kon t i -nu i er l ich ei n er interessierten Öf f ent l ich k eit in Industrie, Wirtschaft und Wis s en -schaft vor g e s tellt werden. Durch die enge Verzahnung mit dem Fachbereich Ma-t he m a t ik der Uni v er s i tät Kaiserslautern sowie durch zahlreiche Kooperationen mit in t er n a t i o n a l en Institutionen und Hochschulen in den Bereichen Ausbildung und For s chung ist ein gro ßes Potenzial für Forschungsberichte vorhanden. In die Be-r icht r ei h e sollen so w ohl hervorragende Di p lom- und Projektarbeiten und Dis s er -ta t i o n en als auch For s chungs b e r ich t e der Institutsmitarbeiter und In s t i t uts gäs t e zu ak t u e l l en Fragen der Techno- und Wirtschaftsmathematik auf g e n om m en werden. Darüberhinaus bietet die Reihe ein Forum für die Berichterstattung über die zahl-r ei c hen Ko o p e r a t i o ns p ro j ek t e des Instituts mit Partnern aus Industrie und Wirt-s chaft.Berichterstattung heißt hier Dokumentation darüber, wie aktuelle Er g eb n is s e aus mathematischer For s chungs- und Entwicklungsarbeit in industrielle An w en d un g en und Softwareprodukte transferiert wer d en, und wie umgekehrt Probleme der Pra-x is neue interessante mathematische Fragestellungen ge n e r ie r en.Prof. Dr. Dieter Prätzel-WoltersInstitutsleiterKaiserslautern, im Juni 2001Towards a Unified Territory Design Approach—Applications,Algorithms and GIS IntegrationJ¨o rg KalcsicsUniversit¨a t des Saarlandes,GermanyStefan NickelUniversit¨a t des Saarlandes and Fraunhofer Institutf¨u r Techno-und Wirtschaftsmathematik,GermanyMichael Schr¨o derFraunhofer Institut f¨u r Techno-und Wirtschaftsmathematik,GermanyJanuary24,2005AbstractTerritory design may be viewed as the problem of grouping small geographic areas into larger geographic clusters called territories in such a way that the latter are acceptableaccording to relevant planning criteria.In this paper we review the existing literature forapplications of territory design problems and solution approaches for solving these typesof problems.After identifying features common to all applications we introduce a basicterritory design model and present in detail two approaches for solving this model:aclassical location–allocation approach combined with optimal split resolution techniquesand a newly developed computational geometry based method.We present computationalresults indicating the efficiency and suitability of the latter method for solving large–scalepractical problems in an interactive environment.Furthermore,we discuss extensions tothe basic model and its integration into Geographic Information Systems.Keywords:territory desgin,political districting,sales territory alignment,optimization algorithms,Geographical Information Systems1IntroductionTerritory design may be viewed as the problem of grouping small geographic areas called basic areas(e.g.counties,zip code or company trading areas)into larger geographic clusters called territories in such a way that the latter are acceptable according to relevant planning criteria.Depending on the context,these criteria can either be economically motivated(e.g. average sales potentials,workload or number of customers)or have a demographic background (e.g.number of inhabitants,voting population).Moreover spatial restrictions(e.g.contiguity,11INTRODUCTION2 compactness)are often demanded.We note,that in the literature often the term alignment instead of design is used.As both expressions are interchangeable we will use the latter one throughout this paper.Territory design problems are motivated by quite different applications ranging from po-litical districting over the design of territories for schools,social facilities,waste collection or emergency services to sales and service territory design.However,the two main applications are political districting and sales and service territory design.In the former application,a governmental area,such as a city or state,has to be partitioned into a given number of territories.As each territory elects a single member to a parliamentary assembly,the main planning criteria is to have approximately the same number of voters in each territory,i.e. territories of similar size,in order to respect the principle of”one man–one vote”.The task of designing sales territories is common to all companies which operate a sales force and need to subdivide the market area into regions of responsibility.Closely related to this is the prob-lem of designing service territories for attending to customers or technical facilities.Typical planning requirements are to design territories which are similar in size,e.g.in terms of sales potentials or workload,or which reduce travel times within the territories needed to attend to customers or service incidents.Due to legal regulations,shifting markets or the introduction of new products,territory design decisions have to be frequently re–evaluated.Especially for a large number of basic areas and territories this is a lengthy task and therefore an algorithmic optimization approach for expediting the process is often desired.For sales territories,well–planned decisions enable an efficient market penetration and lead to decreased costs and improved customer service while in terms of political districting,an algorithmic approach protects against politically motivated manipulations during the territory design process.When reviewing the literature,one can observe that only few papers consider territory design problems independently from a concrete practical background.Hence the tendency in operations research to separate the model from the application and establish the model itself as a self–contained topic of research can not be observed.However,when taking a closer look at the proposed models for different applications,a lot of similarities can be observed.Indeed, the developed models can often be,more or less directly,carried over to other applications.Therefore,we will introduce a basic model for territory design and present in detail two approaches for solving the problem:a classical location–allocation approach combined with optimal split resolution techniques and a newly developed computational geometry based method.In the former one,which was already introduced in the mid sixties and has been extensively used since then,the territory design problem is modeled as a discrete capacitated facility location problem and solved by applying a location–allocation method.As territories of similar size are usually obtained in this approach byfirst solving a continuous,capacitated transportation problem and then rounding the fractional assignments in some non–optimal way in order to obtain non–overlapping territories,we present optimal techniques for resolving the non–integral assignments.Optimal in the sense that the maximal or average deviation of the territory size from the mean is minimal.Since this method is not suitable for the use in an interactive environment,as our computational tests showed,we developed a new method,which is based on computational geometry and utilizes the underlying geographical information of the problem.The idea of this method is to recursively partition the region under consideration geometrically into smaller and smaller subproblems taking the planning criteria into account until an elemental level has been reached where the territory designproblem can efficiently be solved for each of the elemental subproblems.Although the idea for this method was already briefly sketched in the literature,no details were given.We will present computational results indicating the efficiency and suitability of this method for solving large–scale practical problems in an interactive environment.The rest of the paper is organized as follows.In the next section,we will present various applications for territory design problems found in the literature and identify basic features, common to all applications.Based on these observations we introduce in Section3a basic model,which covers most of the previously identified features.In the following section we review the existing literature on models and solution techniques for solving territory design problems and in Sections5and6present in detail two methods for solving such problems:a classical location–allocation approach combined with optimal split resolution techniques and a newly developed computational geometry based method.In the next section,computational results for the two approaches are presented and in Section8we discuss extensions to the basic model and the integration of the presented methods into a Geographic Information System.The paper concludes with a summary and an outlook to future research.2Applications of territory design problemsIn what follows we will present several applications which all have in common the task of subdividing the region under inspection into a number of territories,subject to some side constraints.2.1Political districtingThe problem of determining political territories can be viewed as one of dividing a governmen-tal area,such as a city or a state,into subareas from which political candidates are elected. This problem,usually referred to as political districting,is particularly important in democ-racies where each territory elects a single member to a parliamentary assembly.This is for example the case in Canada,New Zealand,most states in the U.S.and Germany.We note that in this context,territories are usually called districts,census tracks or constituencies.In general,the process of redistricting has to be periodically undertaken in order to account for population shifts.The length of these periods varies from country to country,e.g.in New Zealand every5years,in Canada and the U.S.every decade.To aid this process, operations researchers have developed since the early sixties many different automatic and neutral districting procedures.For recent books on political districting,the reader is referred to Grilli di Cortona et al.[GdCMP+99]and Bozkaya et al.[BELN05].In the past,political districting has often beenflawed by manipulation aiming to favor some particular party or to discriminate against social or ethnic minorities.Since the respon-sibility for approving state and local districting plans usually falls to elected representatives, plans are likely to be shaped implicitly,if not overly,by political considerations, e.g.to keep them in power.A famous case arose in Massachusetts in the early nineteenth century when the state legislature proposed a salamander–shaped electoral district in order to gain electoral advantage.The governor of the state at that time was Elbridge Gerry and this practice became known as gerrymandering.See Lewyn[Lew93]for an interesting description of gerrymandering cases.To prevent political interference in the districting process,many states have set up a neutral commission,whose functions include the drawing up of political boundaries satisfying a number of legislative and common sense criteria.Depending on the country or jurisdiction involved,these criteria may be enforced by legislative directive,judicial mandate or historical precedent.However,in the scientific literature related to political science,law or geography there is no consensus on which criteria are legitimate for the districting process,i.e.satisfy the neutrality condition.Some are predetermined by constitutional laws,while others are based on common sense and can be disputed.Moreover,it is unclear how they should be measured(Williams[Wil95]).In the following,we will present some of the most commonly used,see e.g.Williams[Wil95],George et al.[GLW97]and Bozkaya et al.[BEL03].They can be placed under the three general headings:demographic,geographic and political. Demographic criteriaPopulation equality When designing electoral districts the main criteria is equity in order to respect the principle of”one man–one vote”,i.e.every vote has the same power,and achieve an equitable presence of elected officials.However,depending on the country, deviations from the equal population target are permitted in order to take other criteria into account.Allowed relative deviations from the average range from5%in New Zealand to25%in Germany up to,in exceptional cases,50%in Canada.In the U.S.however,population equality has been deemed by the courts to be very important and as a result the actual deviation now–a–days is in most cases less than1%.Minority representation The intention of this criterion is to ensure that minority voters have the same opportunity as other members of the electorate to participate in the political process and to elect representatives of their choice(Parker[Par90]).Especially in the U.S.,this criteria has become an important consideration during the last30years. Geographic criteriaCompactness A district is said to be geographically compact if it is somewhat round–shaped and undistorted.Although being a very intuitive concept,a rigorous definition of compactness does not exist.Niemi et al.[NGCH90]and Horn et al.[HHV93]propose several measures to asses the compactness of a district,none of which is comprehensive.Compact districts are desired since this reduces the possibility of gerrymandering.In fact,in the U.S.,compactness has been defined as simply the absence of gerrymandering.Some authors however argue that the importance of compactness is less than that of other criteria,since gerrymandering is usually not a problem when an algorithm does not use political data(Garfinkel and Nemhauser[GN70]).Contiguity In general,territories have to be geographically connected.First,to protect once more against gerrymandering and second simply because of administrative reasons. Boundaries and community integrity In many cases,districts should be designed such that they adhere to the boundaries of other political constituencies,like cities or coun-ties,or match as closely as possible the boundaries of the previously existing electoral districts.Moreover topological obstacles like mountain–ranges or large bodies of water should be taken into account(George et al.[GLW97]).Political criteriaPolitical data Although much disputed some authors decided to consider political data for the redistricting process.For example Bozkaya et al.[BEL03]employ a criteria to achieve socio–economic homogeneity across the districts to ensure a better rep-resentation of residents who share common concerns or views.Like Garfinkel and Nemhauser[GN70],they argue that through the use of computer based methods,the possible subjective influence is minimal.Table1provides a selection of articles for political districting problems solved with meth-ods from operations research.It is indicated which of the above mentioned criteria are con-sidered and in addition the country to which the article refers to.Since all authors,without exception,take population equity into account,this criterion is omitted.Reference Country Contiguity Boundaries Compact-Politicalness dataHess et al.[HWS+65]USA+-+-Garfinkel&Nemh.[GN70]USA+-+-Helbig et al.[HOR72]USA+++-Bodin[Bod73]USA+---Bourjolly et al.[BLR81]Canada+-++ Nygreen[Nyg88]Wales+++-Hojati[Hoj96]Canada--+-George et al.[GLW97]New Zealand+++-Ricca&Simeone[RS97]Italy+++-Mehrotra et al.[MJN98]USA+++-Cirincione et al.[CDO00]USA++++ Bozkaya et al.[BEL03]Canada++++Forman and Yue[FY03]USA+++-Table1:Selected operations research studies for political districting.2.2Sales and service territory designThe important but expensive task of designing sales territories is common to all companies which operate a sales force and need to subdivide the market area into regions of responsibility. Closely related is the problem of designing service territories for attending to customers or technical facilities.Here,often quite similar criteria are employed for the design of territories for service staff.Fleischmann and Paraschis[FP88]report on a German manufacturer of consumer goods who delivers products to several thousand wholesalers.Sales promotions and advertising amongst the retailers is very important in the considered business and is carried out by sales agents,where each agent is in charge of a certain territory.The study was motivated by the impression of the company,that the8year old territories seemed to be inappropriate for today’s business,mainly because of the uneven distribution of workload.(Hereby,theworkload of a customer was expressed as an internal score taking into account sales value and frequency of visits.)Blais et al.[BLL03]report on a home–care districting problem in the province of Quebec, Canada.In this case,local community health clinics are responsible for the logistics of home–care visits by health–care personnel,like nurses or physiotherapists,in a given area.If the area is too large,it has to be partitioned into territories,each looked after by a different multi–disciplinary team.Due to a change in the Quebec health care policies,the overall workload increased and became uneven between different territories.To alleviate this problem,the home–care services managers decided to increase their number and re–align them.In general,there are several motivations for aligning existing or designing new territories. First an increase or decrease in the number of sales–or service–men obviously requires some adjustment of the territories.Other reasons are to achieve better coverage with the existing personnel or to evenly balance workload among the them.Moreover customer shifts or the introduction of new products make it necessary to align territories.In the following we present several commonly used criteria for sales territory design prob-lems,see e.g.Zoltners and Sinha[ZS83].Organizational criteriaNumber of territories Often,the number of districts to be designed is predetermined by the sales force size designated by the company or planner,see e.g.Fleischmann and Paraschis[FP88].In case the size is not self–evident,several methods are proposed to compute suitable numbers.For an overview see Howick and Pidd[HP90].In the past,several authors pointed out,that there exists an interdependency between the sales force size and the territory design and proposed models which assume a variable number of territories,see e.g.Drexl and Haase[DH99].Basic areas Sales territories are in most cases not designed based on single customers.In fact,customers are usuallyfirst aggregated into small areas which in turn then serve asa basis for the territory design process.Typical examples for basic areas are counties,zip code areas,predefined prospect clusters or company trading areas.As a result, depending on the level of detail or aggregation,the complexity of the problem reduces considerably and in addition relevant planning data,like sales potentials or distances, is generally much easier to obtain or estimate.Especially the last characteristic has a strong impact on the effort of the planning process.Exclusive assignment of basic areas In most applications basic areas have to be exclu-sively assigned to a territory.This requirement is motivated by several factors.Most notably,unique allocations result in transparent responsibilities for the sales representa-tives avoiding contentions among them and allowing for the establishment of long–term customer relations.The latter aspect often goes along with the desire to minimize arbi-trary changes in territory boundaries.Moreover,staff–dependent performance reviews are easier to compile.Locations of sales representatives Since sales persons have to visit their territories reg-ularly,their location,e.g.office or residence,is an important factor to be considered in the territory design process.Here,one has to decide wether locations of representativesare predetermined and should be kept or are subject to the planning process.Zoltners and Sinha[ZS83]remark,that center–seeking territory design approaches have the practical shortcoming,that most sales persons have strong preferences for home–base cities.Fleischmann and Paraschis[FP88]however report in their case study,that management did not want sales persons residences to influence the definition of territo-ries heavily,because addresses can frequently change.Geographical criteriaThese criteria are mainly motivated by the fact that sales representatives have to travel within their territories.Contiguity Territories should be geographically connected.Accessability Often a good accessibility of territories,e.g.to highways,or within territories, for example by means of public transportation,is required.Moreover,sometimes non–traversable obstacles like rivers or mountain ranges have to be accounted for. Compactness In most applications,compact territories are an important design criterion.As Hess and Samuels[HS71]point out,one way to improve a salesman’s efficiency is to reduce his unproductive travel pact territories usually have geographically concentrated sales(service)activity,therefore less travel,more selling(service)time and hopefully higher sales(better service levels).In other words,the term compactness expresses the desire for territories with minimal total travel times.In several applications,travel times within territories are in addition considered as part of an activity–related design criterion,e.g.workload.Ronen[Ron83]describes the case of a sales territory design problem for sparse accounts of a distributor in the Midwestern United States,where travel time is even the major design criterion,as the market area of the distributor covers almostfive states.Activity–related criteriaGeographic requirements on sales territories mainly focus on the travel aspect.However travelling is only a means to an end for the actual work to do,namely selling products or providing service.Balance For sales and service territory alignment problems,often districts which are bal-anced relative to one or more attributes(called activity measures)are sought for.This criterion expresses a relation of territories among each other and is motivated by the desire of an even treatment of all sales persons.For example in order to evenly dis-tribute workload or travel times among the sales persons or service staffor for reasons of fairness in terms of potential prospects or profit.Although several different sales territory related attributes have been discussed in the lit-erature(see e.g.Hess and Samuels[HS71],Zoltners and Sinha[ZS83]),one can observe, that only few authors consider more than one criterion simultaneously for designing balanced territories(Deckro[Dec77],Zoltners[Zol79],Zoltners and Sinha[ZS83]).Apart from the desire for balanced territories,sometimes strict upper or lower bounds for the size of districts are given.For example on maximal travel times or minimal number of customers within the district.Maximizing profit Especially for sales companies,profit is a major aspect in the planning process.Generally a limited resource of call time or effort is available and has to be allocated in a profit–maximizing way amongst a number of sales entities such as cus-tomers or prospects.See Howick and Pidd[HP90]for a good overview of commonly used time–effort allocation methods.To this end,several authors(Lodish[Lod75],Shanker et al.[STZ75],Glaze and Wein-berg[GW79],Zoltners and Sinha[ZS83],Skiera and Albers[SA94]and Drexl and Haase[DH99])propose an integration of time–effort allocation and territory design methods in order to produce more profit and sales than can be obtained by merely using sales potential.However,Ronen[Ron83]claims,that changing the solution of the strategic territory design problem is much more complicated and expensive than that of the operational time–effort allocation problem and therefore addresses them separately.Although several methods for integrating time–effort allocation and territory design to maximize profit have been proposed,most procedures still consider the balancing re-quirement when designing districts.For example to evenly share workload,potential prospects or profit among their sales force.Skiera and Albers[SA94]and Drexl and Haase[DH99],among others,object that the balancing aspect,i.e.fairness or equity, is not the primary criterion for most companies.The main aim should be to maximize profits,regardless of any balancing aspect.Skiera[Ski97]reports of(randomly gen-erated)experiments,where the sales obtained by a pure profit–maximizing approach compared with one taking balance into account,were5%–14%higher.In Table2a selection of,in our opinion,important articles for sales and service territory design problems solved with methods from operations research is provided.An extensive overview of models before1990can be found in Howick and Pidd[HP90].For each reference we mark wether a selected criterion is considered in the proposed model (”+”)or not(”-”).For the organizational criteria”number of territories”and”locations of sales representatives”,a”f”or”v”indicates if the number or locations arefixed or variable.”Mult”stands for models taking more than one activity measure for balance into account. Since the organizational criterion”exclusive allocation”and the geographic”compactness”requirement were considered by all authors without exception,these two are omitted in the table.2.3Other applicationsBesides the most common problems of sales territory design and political districting,several authors report on various other closely related applications.2.3.1Territories for facilities providing service at afixed locationIn many cases,customers have to visit a(public)facility in order to obtain service,e.g.schools or hospitals.Reference anizational Geographic Activity–relatedNb.Loc.s Contig.Access.Bal.Profit Hess&Samuels[HS71]Sales/serv.f v--+-Easingwood[Eas73]Service f f+-+-Shanker et al.[STZ75]Sales v f++++ Segal&Wein-Service f f+-+-berger[SW77]Glaze&Weinberg[GW79]Sales f v-+-+ Zoltners[Zol79]Sales f v+++/mult+ Marlin[Mar81]Service f f--+-Ronen[Ron83]Service f f--+-Zoltners&Sinha[ZS83]Sales f f++mult-Sales f v-v+-Fleischmann&Paraschis[FP88]Skiera&Albers[SA94]Sales f f++-+ Drexl&Haase[DH99]Sales v v++-+ Blais et al.[BLL03]Service f-++-+ Table2:Selected operations research studies for sales and service territory design problems.School districtsPalermo et al.[PDGT77]and Ferland and Gu´e nette[FG90]deal with the problem of assigning residential areas to schools.As an outcome of the planning process,all residential areas in the region under consideration are partitioned into a number territories,one for each school. Criteria generally taken into account are capacity limitations on and equal utilization of the schools,maximal or average travel distances for students,good accessability and racial balance (the latter especially in the U.S.).Territories for social facilitiesWhen planning territories for social facilities,like hospitals or public utilities,administrative units have to be aggregated into territories.As a result,it is determined for every inhabitant to which facility he should go in order to obtain service.Typically the number of inhabitants of each territory has to be within predetermined bounds in order to account for a good utilization and a limited capacity of the social facility.Moreover territories should be contiguous and the facilities should be easily accessible for all inhabitants of the respective territory,for example by public transportation.See e.g.Andria et al.[ACP79]and Minciardi et al.[MPZ81].2.3.2On–site service territoriesSeveral(public)institutions provide their service not at afixed location but distributed over a geographic region or on–site where the service incident occurs.Winter services and solid waste collectionMuyldermans et al.[MCvOL02]deal with the planning of winter gritting and salt spreading services.On a superior planning level,the region under consideration has to be partitioned into territories,where each territory contains at least one vehicle depot.Afterwards,vehicle。

XenoWorks微机械手操作指南说明书

XenoWorks微机械手操作指南说明书

The XenoWorks Micromanipulatorhas been designed around our hugely-successful MP-285 electrophysiology micromanipulator,with the addition of a smooth-moving adjustableinverted joystick. Because the motor drive is basedon an electrophysiology design, the XenoWorks Micromanipulator is extremely stable andresistant to ambient vibration. This stability alsomakes this manipulator an ideal platform for use inconjunction with the PrimeTech Piezo Impact Drives(PMM-150FU and PMM-4G), for applications suchas mouse nuclear transfer and intracytoplasmicsperm injection.The joystick offers an unprecedented level ofuser comfort during operation. The use of aninverted joystick in conjunction with an almostzero-profile base allows the operator to resttheir hands and forearms on the bench surface,reducing the risk of repetitive strain injuries.The function keys can be located and activatedby touch, removing the need to look away fromthe microscope, which will save time.Another time saving feature is the “Setup”function. Activating the setup function places themicromanipulator in the optimum position prior to beginning work (halfway on the Y-axis, three-quarters down on the Z-axis and three-quarters inwards on the X-axis). Once the manipulator is in the default setup position, the micropipette tip will already be projecting into the microscope’s optical path, reducing the time taken to find the pipette tip through the microscope. This also ensures that there is adequate travel range in all axes of motion prior to starting work.The manipulator uses a pre-programmed “Home” position for fast micropipette exchange, and the ability to determine a “Work” position so the pipette can be repositioned with the press of a button. A Y-axis lockout function allows X-only axial movement for maximum control during injections, and a Z-limit memory position is used to prevent the pipette from colliding with the bottom of the dish.The XenoWorks joystick was carefully designed to allow any user to immediately develop an intuitive feel that makes it easy to precisely control position. Movement of the joystick is directly converted to a proportional movement of the manipulator. Control of the scale factor, the center of travel and programmable positions are right at your fingertips. XENOWORKS ™ MICROMANIPULATOR BRML XenoWorks Micromanipulator (Left) Includes: 3 axis motor drive (left-hand configuration),motor drive base plate, controller, joystick user-interface, connecting cables, and manual BRMR XenoWorks Micromanipulator (Right) Includes: 3 axis motor drive (right-hand configuration), motor drive base plate, controller, joystick user-interface, connecting cables, and manualOne Digital Drive • Novato • CA 94949 • Phone 415.883.0128 Fax415.883.0572••********************HEIGHT AND TENSION ADJUSTABLE FOR OPTIMAL ERGONOMICSEXCEPTIONALLY SMOOTH AND RESPONSIVE MICROPIPETTE MOVEMENTVARIABLE MOVEMENT RANGE – 6 SETTINGS FROM COARSE TO ULTRA-FINEUSER-DEFINED MEMORY POSITIONSAuthorized Dealer:Meyer Instruments, Inc.1304 Langham Creek Drive, Houston,TX 77084281-579-0342 。

  1. 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
  2. 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
  3. 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。

The Design of Microstrip Six-Pole Quasi-Elliptic Filter with Linear Phase Response UsingExtracted-Pole TechniqueKenneth S.K.Yeo,Michael ncaster,Member,IEEE,and Jia-Sheng Hong,Member,IEEEAbstract—The development of microstrip filters has been in great demand due to the rapid growth of wireless communica-tion systems in this decade.Quasi-elliptic response filters are very popular in communication systems because of their high selectivity,which is introduced by a pair of transmission zeros.A number of ways of implementing the quasi-elliptic response filter on microstrip have been studied over the last two decades,i.e., the cascaded quadruplet filter,canonical filter,and extracted-pole filter.However,there is very little information in the literature giving the design details for microstrip extracted-pole filters. In this paper,design equations of the extracted-pole filter for microstrip are reviewed.A new class of microstrip filter is also presented here.This class of filter will have a quasi-elliptic function response and at the same time linear phase in the pass-band.The linear phase of the filter is introduced by an in-phase cross coupling,while the transmission zero is realized using an extracted-pole technique.Experimental results,together with a theoretical comparison between the group delay of this design,and the conventional quasi-elliptic six-pole filter are also presented. Index Terms—Author,please supply index terms.E-mail key-words@ for information.I.I NTRODUCTIONT HE rapid growth of wireless and mobile communica-tions in this decade has catalyzed an increasing demand for a high-performance microstrip bandpass filter with high selectivity and linear phase or flat group delay in the pass-band.High-selectivity bandpass filters have been successfully achieved by introducing additional out-of-phase cross cou-plings in the filter structure—namely,the canonical filters [1]and cascade quadruplet(CQ)filters[2].Conventionally, linear phase is usually achieved by a reflection-type equalizer attached via a circulator to the output of the bandpass filter. Jokela[1]has shown that by using the canonical filter structure, both high selectivity and linear phase is achievable without an external equalizer.However,there are some disadvantages attached to the canonical structure.The cross couplings,both in-phase and out-of-phase,contribute to the transmission zeros both at real and imaginary axes.Therefore,canonical filters are very difficult to tune,which is potentially difficult for narrow-band bandpass filters.In Jokela’s design,as shown in Fig.1(a),three types of resonator topologies are used to realize the required cross coupling.For a single topology filter,it isManuscript received August24,1999.The authors are with the School of Electronic and Electrical Engineering, University of Birmingham,Edgbaston,Birmingham B152TT,U.K. Publisher Item Identifier S0018-9480(01)01072-9.Fig.1.Coupling structure and microstrip implementations for:(a)canonicalfilter,(b)cascaded quadruplet filter,and(c)extracted-pole filter.always difficult to arrange the resonators in order to achieveboth in-phase and out-of-phase cross couplings.For CQ filters,both high selectivity and linear phase can be achieved.CQfilters have more flexibility compared to the canonical filterbecause the transmission zero pair at the imaginary axis(realfrequency)is controlled by an out-of-phase cross coupling andthe real axis(imaginary frequency)is controlled by an in-phasecross-coupling independently.However,to achieve both realand imaginary frequency transmission zero pairs,a minimumof eighth order is required due to the arrangement of the CQfilter.In this paper,a new way to achieve both high selectivity andlinear phase in the passband for microstrip filters is proposed.The structure allows for a sixth-order minimum.This new struc-ture originatedfrom-mode waveguide filters[3].The transmission zero pair at the imaginary axis is extracted from thetransfer function and realized separately with a pair of bandstopfilters and phase shifters connected to the end of the main cou-pling structure.The real axis transmission zero,which flattens 0018–9480/01$10.00©2001IEEEthe group delay,is achieved by an in-phase cross coupling.Thus,the real and imaginary zero pairs are independently tuned.Thishas made the structure very attractive for practical design.Thisdesign can be achieved even for resonator topology,which canonly realize single type of coupling.The microstrip(or stripline)layout of the canonical,CQ,and extracted-pole filters are shownin Fig.1,together with the coupling configuration diagrams.The extracted pole for microstrip was introduced by Hedgesand Humphreys[4]by using microstrip hairpin topology.Theydemonstrated that the extracted-pole technique developed byRhodes[3]can be transferred to microstrip using a four-polebandpass filter.However,there was little information on howto obtain the parameters for the microstrip design from theextracted-pole synthesis.This paper will extend Rhodes’extracted-pole technique to microstrip by working through anexample of a six-pole quasi-elliptic response filter.A new set ofequations for determining the externalis the fractional band-width of the bandpass filter.The frequency invariant admittance-network of capacitors.To convert the circuit model of Fig.5into microstrip,the coupling between the resonator and feedline(the external-factor equations,i.e.,and the coupling co-efficients(3)YEO et al.:DESIGN OF MICROSTRIP SIX-POLE QUASI-ELLIPTIC FILTER WITH LINEAR PHASE RESPONSE323Fig.3.Extracted section of the filter.(a)Original extracted section.(b)Modified extracted section to a realizable circuit model.(c)Equivalent circuit for microstrip layout using a square-loopresonator.Fig.4.Plot of j S 21j of the modified extracted section compared to the originalcircuit.toth resonator.Conventionally,the low-pass parameters [8]are used to determine theexternal-factors and coupling coefficients would notbe perfectly accurate.This is because the conventionalequationsFig.5.Full-circuit model for the bandpass extracted-pole microstrip filter.for determiningexternal,and an imaginary frequency trans-mission zero pairat324IEEE TRANSACTIONS ON MICROWA VE THEORY AND TECHNIQUES,VOL.49,NO.2,FEBRUARY 2001TABLE ITABLE IIFig.6.Theoretical responses together with the circuit model simulationresults.The coupling coefficients and the external-factor of 250andthe ideal case.When the loss is added,the passband insertionloss will increase to about 3dB.To illustrate the linear phaseresponse of this filter,a comparison is made between a conven-tional six-pole quasi-elliptic filter (without in-phase cross cou-pling)and this example.It is clearly shown in Fig.7(b)that byYEO et al.:DESIGN OF MICROSTRIP SIX-POLE QUASI-ELLIPTIC FILTER WITH LINEAR PHASE RESPONSE325Fig.8.Plot of type-II mixed coupling of a square open-loop resonator. introducing an in-phase cross coupling,the group delay can be flattened significantly.By introducing the in-phase cross cou-pling,type-II mixed coupling is introduced.V.T YPE-II M IXED C OUPLINGThe type-II mixed coupling was first discussed by Hong et al.[9]for a microstrip hairpin structure.This type of coupling is not continuously decreasing with coupling distance,but in-creasing until one particular point then starts decreasing.Here, there are actually two types of coupling existing in the type-II mixed coupling.Fig.8shows a plot of the coupling coefficients, which are obtained from full EM[5]simulation,against the cou-pling distance for the microstrip square-loop resonators.The type-II mixed coupling is also the superposition of the electric and magnetic couplings,as in the type-I mixed coupling[2]. However,the magnetic coupling of the type-II mixed coupling is out-of-phase with respect to the electric coupling.This is in contract to the type-I mixed coupling where the magnetic cou-pling is in-phase with respect to the electric coupling.Therefore, the electric and magnetic couplings in type-I mixed coupling en-hance each other,whereas the electric and magnetic couplings of the type-II mixed coupling cancel out each other.When the resonators are placed very close to each other,the electric coupling dominates.The electric couplings are very strong,but decay very rapidly.At one particular spacing,i.e., about0.9mm for this case,there is no coupling between the two resonators.This is because the electric coupling is equal to the magnetic coupling.In Fig.8,marked“Section II,”the coupling increases with increasing coupling distance,which is not expected in most coupling structures.This happens because the net coupling has changed from electric to magnetic,and also because the magnetic couplings decay at a slower rate compared to the electric coupling.Therefore,the net mag-netic coupling appears to be increasing(more negative).The couplings will increase until one particular point and start decreasing again,as shown in Fig.8,marked“SectionIII.”Fig.9.Circuit model of the type-II mixed coupling.When the spacing between the resonators is very far apart,the electric fields between the resonators are negligible.Therefore, the coupling appears to be purely magnetic.To model the type-II mixed coupling,a circuit diagram,as shown in Fig.9,is used.Due to the arrangement of the coupling structure,the flow of currents in the two resonators is changed compared to the type-I mixed coupling[2].This will cause the mutualinductances326IEEE TRANSACTIONS ON MICROWA VE THEORY AND TECHNIQUES,VOL.49,NO.2,FEBRUARY2001 Fig.10(a)Microstrip circuit layout of an extracted-pole bandpassquasi-elliptic response filter with linear phase with square-loop resonators.(b)Photograph of the fabricated filter.TABLEIIIas shown in Fig.8.The best fit obtained when the ratio betweenthe electric coupling and the magnetic coupling are0.6–0.4,i.e.,(11)VI.E XPERIMENTAL F ILTERThe layout of the six-pole microstrip filter using square-looptopology is shown in Fig.10(a).A photograph of the fabri-cated filter is shown in Fig.10(b).Here,-factors corresponding to Fig.10(a)are shown inTable III.This filter is fabricated using copper microstrip on an RT/Du-riod substrate with relative dielectricconstant of10.8andthickness of1.27mm.The linewidth of the microstrip is1.1mmthroughout.The measured performance of the fabricated mi-crostrip filter,which is obtained using the HP8720network an-alyzer,is shown in Fig.11(a).Some tuning is performed inthis measured result to obtain the best response.The tuning isachieved by placing small dielectric materials at the appropriateposition to change the resonant frequencies of each resonatoraccordingly.Tuning is essential because of the unavoidable fab-rication errors.The midband insertion loss is measured at about3.3dB,which is mainly contributed by the conductor lossofFig.11.(a)Experimental results for j S11j and j S21j.(b)Comparisonbetween the experimental and theoretical group delay.the copper.Two attenuation poles at the rolloff frequency nearthe passband,which improve selectivity,are achieved.The mea-sured group delays are shown in Fig.11(b).The group delay ofthe experimental results is slightly lower compared to the pre-dicted circuit model because there are slight increases in thebandwidth of the fabricated filter.VII.C ONCLUSIONWe have presented the design procedure for the ex-tracted-pole technique for microstrip filter.A circuit model ofthe microstrip extracted pole is also presented.This model hasshown close correlation with the theory.This model makes thedesign of microstrip extracted-pole technique for quasi-ellipticfunction filter more straightforward.To demonstrate the validityof the circuit model,a six-pole quasi-elliptic microstrip filterhas been designed,fabricated,and tested.The measurementcircuit model,together with the theoretical response,has beenpresented.We have also shown that by introducing an in-phasecross coupling in the microstrip extracted-pole filter,a linearYEO et al.:DESIGN OF MICROSTRIP SIX-POLE QUASI-ELLIPTIC FILTER WITH LINEAR PHASE RESPONSE327 phase filter with a real transmission zero pair can be achieved.The unexpected phenomena of the type-II mixed coupling hasbeen discussed.A circuit model is also presented to explainthese phenomena.R EFERENCES[1]K.T.Jokela,“Narrow-band stripline or microstrip filters with transmis-sion zeros at real and imaginary frequencies,”IEEE Trans.MicrowaveTheory Tech.,vol.MTT-28,pp.542–547,1980.[2]J.S.Hong and ncaster,“Couplings of microstrip square open-loopresonators for cross-coupled planar microwave filters,”IEEE Trans.Mi-crowave Theory Tech.,vol.44,pp.2099–2109,Nov.1996.[3]J.D.Rhodes and R.J.Cameron,“General extracted pole synthesis tech-nique with applications to low-loss TE。

相关文档
最新文档