Reliability of surface mount technology printed ... - FernUni Hagen

09.10.2009 - chemical etch process (KOH) and a plasma etch process ("Deep .... can state that the higher the material input the worse the impact on the ...
617KB Größe 4 Downloads 579 Ansichten
POWER DEVICES AND SENSORS (AG Leistungsbauelemente und Sensorik) apl. Prof. Dr. rer. nat. Reinhart Job Universitätsstr. 27 (PRG) D-58084 Hagen Tel. +49 (0)2331 987-1183 Fax: +49 (0)2331 987-4105 E-Mail: [email protected] www.fernuni-hagen.de/ag_job/ Staff: Wolfgang Benner

+49 (0)2331 987-1149

[email protected]

Matthias Dau,

Flensburg,

[email protected]

Sebastian Döring,

Mittenwald,

[email protected]

Markus Freitagsmüller,

Castrop-Rauxel,

[email protected]

Tino Krause,

Berlin,

[email protected]

Ralf van de Sand,

Bottrop,

[email protected]

Thierry Tchoumi,

Denzlingen,

[email protected]

Maik Wolfkühler,

Fredenbeck,

[email protected]

+49 (0)2331 987-4101,

[email protected]

Students:

Secretary: Diana Schulz,

Summer School: Advanced Microsystems Technologies for Sensor Applications R. Job1), U. Hilleringmann2), P. Glösekötter3), G. Wirth4) 1)

University of Hagen, Mathematics and Computer Science, Hagen, Germany 2) University of Paderborn, Computer Science, Electrical Engineering and Mathematics, Paderborn, Germany 3) University of Applied Sciences, Electrical Engineering and Computer Science, Münster, Germany 4) Federal University of Rio Grande do Sol, Electrical Engineering Porto Alegre, Brazil In July 2009 for the first time a Summer School devoted to Advanced Microsystems Technologies for Sensor Applications was realized at the Universidade Federal do Rio Grande do Sul (UFRGS) in Porto Alegre, Brazil. The summer school was organized together with two colleagues from the University of Paderborn (Prof. Dr. Hilleringmann, who was the principal organizer of the summer school) and the University of Applied Sciences in Münster/Steinfurt (Prof. Glösekötter). The local host at the UFRGS was Prof. Wirth.

Fig. 1: The 2009 class of the Summer School on Advanced Microsystems Technologies for Sensor Applications held at the UFRGS in Proto Alegre, Brazil

The summer school comprehended three lectures of interrelated lessons, i. e. Semiconductor Technology (Prof. Hilleringmann), Sensor Technology (Prof. Job), and Analog and Digital Circuits (Prof. Glösekötter), which jointly presented a detailed and actual overview on the addressed subjects. The lectures also gave an outlook on nano-electronics and nano-sensor applications. The educational focus was aimed at students (preferably studying in higher semesters) and/or PhD students in Electrical Engineering, Physics, Information Technology, Computer Science, or Applied Materials Research. The summer school was adjusted to the regular lectures presented at the UFRGS; and the participating students had to expose themselves to written examinations in each of the three lectures. Each of the lectures covers 30 lessons including the final examinations. During several evening session cultural recitals were presented, which provided general geographical, cultural and economical information about Europe, Germany and the Ruhr area as well as detailed information about the school system, the academic education and especially the advantages and challenges of distance education at the University of Hagen. The summer school was financed by the German Academic Exchange Service (DAAD).

Life IT: IT meets Environmental and Sustainable Energy Technologies R. Job, J. Keller University of Hagen, Mathematics and Computer Science, Hagen, Germany In spring 2009 (May 15th – 16th), the German-Russian conference Life IT: IT meets Environmental and Sustainable Energy Technologies took place at the University of Hagen. The conference was organized by the University of Hagen and the International Academy of Management und Technology (INTAMT); and it was financially supported by the German Federal Ministry of Education and Research (BMBF). The authors of this brief report were decisively involved in the organisation, scientific program planning and execution of the conference. Basically, the meeting was separated into two parallel events, i. e. a scientific conference dedicated to the discussions of international experts in the envisaged research field and a student conference where students from the CIS (the former Soviet Union) and Germany could present and discuss their research activities. Main foci of the conference were networking and the establishment of project activities involving also the participation of industrial partners.

Fig. 1: Participants of the conference Life IT: IT meets Environmental and Sustainable Energy Technologies held at the University of Hagen in May 2009 (photo: Gerd Dapprich, University of Hagen)

The objectives and results of the conference could be summarized as follows: -

The envisaged multidisciplinary research field was apprehended as an open system and systemic inceptive research activities were emphasized.

-

A scientific platform for interdisciplinary research was discussed, which focused on widespread investigations on environmental and climate monitoring, sustainable energy supply with regard to the reduction of greenhouse gas emissions and especially the role of IT technologies in such research activities.

-

Specialized research fields (mainly from engineering, computer science, material science, physics, chemistry, geo- and bioscience, but also social and political sciences, philosophy, medicine, especially geo-medicine) were denominated, which might participate in the systemic approach of the multidisciplinary research field.

-

Intended national and/or international partnerships and research projects focused on interdisciplinary and systemic solutions for the envisaged field of multidisciplinary environmental research were debated.

-

Related lecture contents in high level education at the undergraduate and graduate university level and the definition of educational objectives were discussed.

The authors want to acknowledge the assistance of Mrs. Irmgard Broekmann and her team (University administration, Dept. 1.2); without their tremendous support we would have been lost.

Compatibility of CMOS and MEMS Process Steps: Comparison of the influence of KOH and DRIE Structuring on Monolithically Integrated Micro-Systems * M. Freitagsmüller1), 2), R. Job1), W. Schreiber-Prillwitz2) 1)

University of Hagen, Mathematics and Computer Science, Hagen, Germany 2) ELMOS Semiconductor AG, Dortmund, Germany

Micro-systems technology has faced a tremendous evolution during the last years; and the significance of micro-systems is still increasing. For a vast range of growing applications, different process technologies are needed for the development of a large variety of microsystems devices and products. Hence, in frame of a master thesis the influences of a wet chemical etch process (KOH) and a plasma etch process ("Deep React Ion Etching", DRIE) on a co-integrated pressure sensor system were analyzed. The basic specifications of the involved co-integrated pressure sensor system are: -

gauge pressure measurement intern piezoresistive Wheatstone-bridge 64 byte EEPROM for calibration data digital linearization analog output pin 5V power supply Fig. 1: Cross-section of a co-integrated system (example)

Within the master thesis an existing wet chemical process, employing potassium-hydroxide (KOH), was used to structure the cavity of the pressure sensor cell. The plasma etch process had to be developed. The boundary condition was that the DRIE process should perform as good as possible; i. e. the etching angles should approach 90° and the appearance of "black silicon" should be as less as possible. Certainly the etched depth-accuracy is important to get a well defined pressure sensor. Therefore, six wafers were necessary for tuning the (Bosch) etch process; before the CMOS test structures could be processed.

Fig. 2: SEM inspection of a KOH processed cavity

Fig. 3: SEM inspection of a DRIE processed cavity

To get a precise impression of components of the developed ASIC ("Application Specific Integrated Circuit"), a few characteristic attributes and properties were investigated, e. g. the oscillator, band gap, resistance of the pressure sensor cell, internal voltage controller, etc. The whole performances of the components were tested in a wide temperature range, and no significant differences could be measured for both employed etching processes. Both process steps worked well for the inspected micro-system. In conclusion, badly designed layouts resulting for instance in large metal antenna effects are a higher risk for the micro-systems performances than the investigated etching processes. * Master thesis of Markus Freitagsmüller (to be finished in June 2010)

Unbalance Recognition with Inertial Sensors in Washing Appliances * T. Krause1, 2), H. G. Albayrak1), R. Job2) 1)

BSH Bosch und Siemens Hausgeräte GmbH, Product Area Laundry, Berlin, Germany 2) University of Hagen, Mathematics and Computer Science, Hagen, Germany

Unbalance recognition is an important factor in the development of washing appliances. In fact, the unbalanced mass through inhomogeneous laundry cannot be prevented by the system. The thereby occurring acceleration forces are the direct factor for stress of the components, and especially, the oscillating system. The outcomes of unbalances are undesirable effects like • inefficiency • noise • safety problems • premature component abrasion Modern control systems in washing appliances work with more than one method to detect and distinguish different unbalance types and quantities. These methods were supported by sensors to calculate an appropriate reaction like speed reduction or spinning abruption.

Figure 1: Acceleration record set of x, y, z

Simplified, the oscillating group of a washing machine can be handled with the second order equation of motion to describe its movement. Aim of this thesis is to investigate triaxial acceleration sensors in washing appliances for movement detection and to compare the results with actual attached sensors as well as to a reference laser system. Predestinated for this intended purpose are micromachined acceleration sensors (so called MEMS, "Micro-Electro-Mechanical-Systems"). MEMS ac- Figure 2: Filtered signal & calculated motion celeration sensors are nowadays small, cheap and thereby competitive to other motion sensors. Because of the prevalence of these sensors in customer and automotive applications in the last years, the sensors became also attractive for the operation in highly competitive markets like household products. To evaluate these sensors a large number of measurements with different unbalance positions were accomplished. The analysis and evaluation of the sensor data was completed with MATLAB. The input data for the signal determination originated from the acceleration sensor; they were transformed into position signals by the following well-known equation: t t

t

0 0

0

x(t ) = ∫ ∫ a(t )dtdt + ∫ v0 dt + x0 Fig. 1 shows a typical acceleration record section with a special spin profile and unbalance mass of 400 g in the front tub area. By calculating the maximum elongation of the oscillating group according to equation 1, the position movement of the tub can be determined, as shown in Fig. 2. The basic problem is to determine unbalances and predict the maximum elongation of the system movement to prevent impacts and movements. Within this thesis different approaches for unbalance detection/prediction will be tested and proved with the recorded data. * Master thesis of Tino Krause (to be finished in Nov. 2010)

Systemic Analysis of Power Generation with Biomass * M. Wolfkühler, R. Job University of Hagen, Mathematics and Computer Science, Hagen, Germany In general, power generation using biomass as fuel is regarded as promising for a sustainable energy supply in the future. However, concerning ecological as well as moral aspects recently some doubts appeared, if biomass utilization really is favorable for large scale power generation. To obtain a better understanding, within a master thesis various types of biomass resources (energy crops) were comparatively studied in view of environmental compatibility. 60

A

B

C

D

E

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

1 2 3 4 5

F

Material Input (rel. units)

50 40 30 20 10 0

1 2 3 4 5

1 2 3 4 5

Biomass Fuel (A, B, C, D, E, F) Fig. 1: Summary of the material-input per service unit data in a comparative analysis of various biomass fuels

The investigated biomass resources were: bales of winter wheat (A), methane gas produced from winter wheat (B), methane gas produced from corn (C), chippings of poplar wood (D), bale of winter wheat straw (E), and chippings of spruce wood (F). The evaluation was carried out on base of the MIPS-concept ("Material Input Per Service Unit"), and based on the local geographical situation in northern Germany and typical biomass power plant features. Extensive lifecycle analysis including the materials inputs of the relevant categories like abiotic (1) and biotic raw materials (2), soil cultivation including also erosion (3), water (4) and air consumption (5) were considered for the analyses. The MIPS-concept was decisively developed at the Wuppertal Institute for Climate, Environment and Energy (www.wuppertalinst.org). The results of the study are summarized in Fig. 1 resolved with regard to the 5 categories of the MIPS-concept (indicated by numbers). The analyzed biomass fuels are indicated by the letters A – F. The particular material inputs were presented in relative units. In general one can state that the higher the material input the worse the impact on the environment. It can be concluded that in case of energy crops the methane refinement is more resource-efficient (B, C) than the direct use of the energy crop (A). This can be attributed to the better efficiency of power generation when using methane instead of simply burning the plants. Concerning energy crops, the best results could be obtained for chippings of poplar wood (D), since plantations are use for a long period of time and the cultivation and erosion based material input is rather low. In case of residuals like straw of winter wheat (E) or chippings of spruce wood (F) the resource efficiency is even higher, since for the MIPS calculation the resource consumption is already included in the calculations for the products like timber or food production. The analysis of power generation with biomass is part of research activities in the field of system engineering and sustainable energy production, which were recently started in the Power Devices and Sensors Group of Prof. Job. * Diploma thesis (D-I) of Maik Wolfkühler (completed in July 2009)

Mechanisms and Perspectives of a Sustainable Energy Supply by Ocean Thermal Energy Conversion * R. van de Sand, R. Job University of Hagen, Mathematics and Computer Science, Hagen, Germany Concerning sustainable energy supply, Ocean Thermal Energy Conversion (OTEC) is a rather unknown technology. However, obeying specific boundary conditions, OTEC exhibits some promising features so that it might play a significant role for the energy supply in the 21st century. An OTEC power plant is a thermodynamic heat engine, which is connected to a high and a low temperature reservoir. The heat energy flowing between the two reservoirs is partly converted into work energy; the underlying thermodynamic cycle is the Ranking cycle. Hence, OTEC power plants operate similar to conventional power plants, which are for instance fired by fossil fuel. For OTEC systems the high temperature reservoir is provided by warm surface ocean water. The low reservoir is the deep ocean water from a depth of about 1.000 m, which has a temperature of about 4 °C throughout the oceans independent on the degree of latitude. Since the efficiency of a heat engine is strongly related to the temperature difference between the high and the low thermal reservoir, in practice, OTEC systems should be located in tropical areas providing water temperatures well above 20 °C at the ocean surfaces. warm ocean water in

deaeration (optional)

non-condensible saturated gases water vapor (desalinated) vacuum chamber flash evaporation

non-condensible gases

unsaturated cold ocean water vapor water in (desalinated)

generator

warm ocean water discharge to sea

condensor

cold ocean water discharge to sea desalinated water (optional)

Fig. 1: Open-cycle OTEC system (see for instance: V. L. Bruch, SAND93-3946, Sandia National Labs, USA)

In frame of a diploma thesis the development of OTEC systems was reviewed enlightening the history and basic theoretical mechanisms as well as their technical realization and several practical applications. Various OTEC systems – open cycle OTEC (Fig. 1), closed cycle OTEC and hybrid OTEC systems – were analyzed with regard to their efficiencies. The main task of the study was the analyses and discussions of the usability of OTEC systems as a sustainable energy source with emphasis on ecological and economical aspects. Facility locations – especially discussing also land based and offshore systems – were also discussed. Although the technological principles are rather simple, OTEC systems are facing some challenges, which up to now prevented a successful breakthrough for this promising technology. Quite large investments have to be done to assemble OTEC systems. And due to still rather low prices for crude oil the provision of usable energy with OTEC is not price-competitive today. However, OTEC is based on solar energy (heating of ocean water); it is a sustainable and quasi boundless energy source. Hence, global warming and fossil fuel limitations make it reasonable to assume that in the future OTEC will play a role well beyond niche applications. The analyses OTEC systems are part of research activities in the field of system engineering and sustainable energy production, which were recently started in the Power Devices and Sensors Group of Prof. Job. * Diploma thesis (D-I) of Ralf van de Sand (completed in Jan. 2010)

Simulation of Ocean Thermal Energy Conversion Systems * T. Tchoumi, R. Job University of Hagen, Mathematics and Computer Science, Hagen, Germany In frame of a Master Thesis a graphic user interface for an Ocean Thermal Energy Conversion (OTEC) power plant was developed. With this tool the operation of OTEC-systems could be simulated under the assumption of realistic conditions. For the simulations, in a first order approximation the vertical temperature distribution of the open ocean was described by a two layer model. The upper layer is heated by the sun; it has a depth or thickness of about 100 m depending on the geographical location. The bottom layer, i. e. the deep ocean regions, consists of rather cold water (e. g. 4 °C). The temperature difference between the upper (warm) and bottom (cold) layers ranges from 10 °C to 25 °C, with the higher values found in equatorial waters. This temperature difference was also covered by the simulations.

Fig. 1: OTEC power plant simulation software; upper curve: simulation for ΔT < 20 °C (no net energy gain), lower curve: positive energy output for ΔT > 20 °C

The properties of OTEC systems could be analyzed for instance by the simulation of the gain of electric power in dependence on various pump settings, the simulation of the electric power in dependence on various temperature differences between the low and the high temperature reservoir of the system, or the simulation of the electric power in dependence of the quantity of water flow used within the applied thermodynamic cycle. For each simulation four major parameters describe the states of a working OTEC power plant. Hence, the generated energy of the actually simulated OTEC system is dependent on these four parameters, i. e. on the delivery (flow) rate of warm water to the pump, the difference between the warm surface water and the cold deep ocean water, the degree of generated vacuum rate in percent, and the flow rate of the cold water from the deep ocean (Fig. 1). The analyses of OTEC systems by simulation are part of research activities in the field of system engineering and sustainable energy production, which were recently started in the Power Devices and Sensors Group of Prof. Job. * Master thesis of Thierry Tchoumi (completed in Oct. 2009)

Detection of Vacancy Distributions by Decoration with Hydrogen * R. Job1), F.-J. Niedernostheide2), H.-J. Schulze2), H. Schulze3) 1)

University of Hagen, Mathematics and Computer Science, Hagen, Germany 2) Infineon Technologies AG, Munich, Germany 3) Infineon Technologies Austria AG, Villach, Austria

Concerning power device applications the detection of vacancy distributions in Float Zone (FZ) silicon is very important. In the literature only one detection method is described based on vacancy decoration by platinum, where Pt profiles can be analyzed by DLTS measurements. But two major drawbacks can be accented: 1. In-diffusion of Pt is a high-temperature step significantly altering the vacancy profiles; 2. DLTS analysis with suitable depth resolution is difficult. Therefore, we developed an analytical method, where vacancies were decorated by hydrogen, finally resulting in hydrogen-related donor-like defect states. The analyses of such donors, and hence, the indirect detection of vacancy distributions could be enabled by simple spreading resistance (SR) measurements with good spatial resolution and low effort. 14

15

15

500 °C

10

10 14 10 13 10 12 10

14

15

Carrier Concentration (cm )

450 °C

-3

-3

Carrier Concentrations (cm )

10 10

14

10

15

400 °C

10

14

10

15

14

a)

5

10

15

Depth (µm)

20

15

10 14 10 13 10 12 10

Tplasma = 450 °C

15

10 14 10 13 10 12 10

Tplasma = 400 °C

15

10

0

Tplasma = 500 °C

10 14 10 13 10 12 10

350 °C

10

-2

He-impl.: E = 3.75 MeV, F = 1 x 10 cm H-plasma: 15 min at various temperatures

+

H -Implantation & 15 min H-Plasma

Tplasma = 350 °C

0

25

b)

5

10

15

20

25

Depth (µm)

Fig. 1: a) Carrier concentration profiles of FZ Si treated by H+- (a) or He+-implantation (b) and subsequent 15min H-plasma exposure at T = 350 – 500 °C; the profiles were deduced from SR measurements.

FZ Si samples, treated by H+-implantation and subsequent H-plasma exposure, exhibited distinct doping profiles related to electrically active vacancy-hydrogen defect complexes (Fig. 1). Vacancies were created during ion implantation; and the SR profiles traced vacancy distributions within the samples. Near the projected ion range Rp, strong vacancy- and hydrogenrelated shallow donor formation occurred, since damage was high and vacancy concentrations were large. Large hydrogen concentrations originating from H-implantation and in-diffusion during the H-plasma exposure occurred also near Rp. Consequently, vacancy-hydrogen defect complexes assembled resulting in n-type doping. At process temperatures > 400 °C vacancies diffused also towards the wafer surface, and there vacancy concentrations were enlarged, too. Similar results were obtained for He+-implanted and H-plasma treated FZ Si; but some differences had to be considered: 1. Implantation damage was stronger at Rp, hence, multi-vacancy concentrations predominantly occurred; 2. hydrogen only derived from in-diffusion during plasma exposure, i. e. H-concentrations near Rp were lower than for H+-implanted FZ Si. In He+-implanted and H-plasma treated FZ Si, acceptor-like defects occurred near Rp overcompensating for the n-type doping, even creating a buried p-type layer (H-plasma exposure at 500 °C). We deduced that these defects were related to multi-vacancy-oxygen complexes. * R. Job, F. J. Niedernostheide, H. J. Schulze, H. Schulze, in: "Analytical Techniques for Semiconductor Materials and Process Characterization VI (ALTEC)", Editors: B. O. Kolbesen, C. L. Claeys, C. Fabry, M. Bersani, D. Giubertoni, G. Pepponi, ECS Transactions 25 (3), 35 (2009), invited talk and paper presented at "The 216th Meeting of the Electrochemical Society (ECS)", Oct. 4th – 9th, 2009, Vienna, Austria

Formation of Acceptor-Like Defect States in Helium Implanted and Plasma Hydrogenated Float Zone Silicon * R. Job1), F.-J. Niedernostheide2), H.-J. Schulze2), H. Schulze3) 1)

University of Hagen, Mathematics and Computer Science, Hagen, Germany 2) Infineon Technologies AG, Munich, German 3) Infineon Technologies Austria AG, Villach, Austria

Two-step processes based on H+- or He+-implantations and subsequent hydrogen-plasma treatments were studied for n-type FZ Si. In the wafers’ subsurface regions down to the projected ion range Rp surplus n-type doping occurred, which was caused by donor-like states related to hydrogenated vacancy defect complexes. Acceptor-like defect complexes occurred close to the surface and in He+-implanted samples also near Rp. Those states might be attributed either to hydrogenated vacancy complexes, e. g. V2-H2, or multi-vacancy-oxygen complexes, such as Vn-Om. Our experimental analyses enlightened the role of oxygen for a formation of acceptor-like defect states in case of He+-implanted and plasma hydrogenated FZ Si. 14

He-impl.: E = 11 MeV, F = 1 x 10

cm

-2

He-impl.: E = 11 MeV, F = 1 x 10

14

-2

cm

15

10

15

10

10

14

10

10

13

10

12

10

11

10

15

10

14

10

13

10

12

10

11

14

-3

Carrier Concentration (cm )

-3

Carrier Concentration (cm )

13

10

H-plasma: 15 min, 400 °C

12

10

H-plasma: 15 min, 400 °C

11

10

10

10

15

10

14

10

13

10

12

10 H-plasma: 60 min, 400 °C

10

10

0

a)

H-plasma: 60 min, 400 °C

11

10

20

40

60

Depth (µm)

80

100

b)

0

20

40

60

80

100

Depth (µm)

Fig. 1: a) Carrier concentration profiles of FZ Si – without (a) and with pre-oxidation (b) – treated by He+-implantation and 15 or 60min H-plasma exposure at 400 °C; the profiles were deduced from SR measurements

Fig. 1 shows carrier concentration profiles – deduced from spreading resistance (SR) measurements – of He+-implanted and subsequently H-plasma treated FZ Si wafers with and without previous pre-oxidation. After 15-min H-plasma exposure at 400 °C of pre-oxidized FZ Si, the carrier-concentration profile exhibited the same shape as for the sample without pre-oxidation. However, after 60-min post-implantation H-plasma exposure at 400 °C, significant alterations were observed (Fig. 1b). Near Rp two p-n junctions appeared indicating that a buried p-type layer was created between about 75 µm and 95 µm depth. This result suggested that the following mechanisms were involved in the formation of the doping profiles. During the applied H-plasma process, hydrogen was introduced into the wafer and rapidly diffused into deep wafer regions. The buried implantation damage layer acted as getter center and a diffusion barrier for H-atoms. After 60 min H-plasma exposure, the whole subsurface layer down to the buried implantation damage layer was flooded with hydrogen. Furthermore, hydrogen caused enhanced oxygen diffusion within the wafer, since it lowered the energy barrier for the interstitial oxygen migration in silicon. Therefore, oxygen could accumulate at Rp; and the probability was strongly enhanced that interstitial oxygen met appropriate defect sites for the formation of acceptor-like defect complexes. However, enough oxygen had to be available for this process. In FZ Si, which was not exposed to a pre-oxidation step, the oxygen concentration was too low to create sufficient acceptor-like defects for a buried p-type layer. * R. Job, F. J. Niedernostheide, H. J. Schulze, H. Schulze, "2009 Materials Research Society (MRS) Fall Meeting", Nov. 30th – Dec. 3rd, 2009, Boston, USA, to be published in: "Reliability and Materials Issues of Semiconductor Optical and Electrical Devices and Materials", Editors: O. Ueda, M. Fukuda, S. Pearton, E. Piner, P. Montanegro, MRS Symposium Proceedings Series, Vol. 1195¸1195-B11-02 (2010)