6月份「光電論壇」演講花絮
時間:97年6月13日下午4點30分~6點30分
講者:Dr.
Jean-Pierre Monchalin
(Group
Leader, Optical Diagnostics of Materials, Modelling and Diagnostics,
Industrial Materials Institute, National Research Council of Canada,
Canada)
講題:Optical
Coherence Tomography and Ultrasound Modulated Optical Imaging at the
Industrial Materials Institute of the National Research Council of
Canada
Dr. Jean-Pierre
Monchalin
於97年6月13日(星期五)蒞臨本所訪問,並於博理館101演講廳發表演說,講題為「Optical
Coherence Tomography and Ultrasound Modulated Optical Imaging at the
Industrial Materials Institute of the National Research Council of
Canada」,本所教師及學生皆熱烈參與演講活動,獲益良多。
|
∼與南京大學(Nanjing University)
博士生交流活動 2008 ∼
【2008 第一屆臺灣大學—南京大學光學微結構與雷射技術博士生論壇】
(時間:97年5月19日至5月20日;地點:臺灣大學)
花絮整理:光電所博士班學生盧彥丞(代表團學生副隊長)
很榮幸能參加這次和南京大學物理系的交流活動,歷經數月的準備和幾經波折,來自南京大學的老師和同學們終於在五月十八日來到台灣,研討會如期地在十九日展開。這次的交流活動分成三大部分
:學術研討會、實驗室參觀和文化之旅。行程如下:
2008.05.18(週日) |
南京大學代表團抵達台灣 |
2008.05.19(週一) |
學術交流研討會 |
2008.05.20(週二) |
學術交流研討會 |
2008.05.21(週三) |
實驗室參觀 (光電所、凝態中心、校史館) |
2008.05.22(週四) |
文化之旅 (貓空纜車、自由廣場、101大樓、士林夜市) |
2008.05.23(週五) |
實驗室參觀 (國家同步幅射中心、友達光電) |
2008.05.24(週六) |
文化之旅 (故宮、淡水) |
2008.05.25(週日) |
南京大學代表團離開台灣 |
在黃升龍所長、林恭如副所長和林筱文小姐的協助,及各位隊員的努力下,這次的交流活動,對南大和台大的同學們而言,均是一個充滿知性及感性、寓鄉土人文於科學交流的旅程。這次參與的同學有:許森明(隊長)、趙家忻、鄭行傑、林昌廷、黃政傑、裴善莊、陳正彬、游政衛和我。李光立因家中突然有事,無法參加交流活動,但我們仍感謝他在籌備期間為我們設討的精美海報及名牌。最後,要感謝的是臨危受命,遞補光立的邱南福學長。
沒吃過豬肉,也看過豬走路。只看過豬走路,大概永遠也不會知道豬肉的味道。相信很多同學對於參加學術研討會已是習以為常的事了,但是,籌備研討會對大多數的同學而言卻是不曾有過的經驗。這次的研討會雖然只是個二十幾人的小型研討會,和國際會議動輒數百人的規模無法相提並論,但藉由籌備這次的研討會和整個交流活動的行程,讓我瞭解到要使整個會議圓滿順利,所要注意的事情是比我一開始預期來的多很多,整個籌備過程對我而言確實是受益良多。另外,看到我們不少同學可以如數家珍地把台北甚至是台灣的特色娓娓道來,讓平日只往返家裡和學校兩地的我感到十分地汗顏。昌廷對新竹一帶的地理位置及台灣產業均相當熟悉;善莊和正彬則對台灣著名的景點及美食瞭若指掌;行傑更是位傑出的導遊,在路上聽他為南大的同學介紹台灣的種種,讓我不得不對他肅然起敬。其實,這次的實驗室參觀和文化之旅,有許多地方是我不曾去過或是很久沒去的,真的要感謝安排行程的同學,安排這樣多彩多姿的行程。經過這次的交流活動,也讓我對生活有著另一層次的體會。
和大陸的學生交流,最敏感的話題莫過於兩岸問題。相信大家也都很好奇,關於這類的話題,我們是如何應對。基本上,不論是這次南大的同學來台參訪還是我們去年去大陸北京、清華大學參加研討會,大陸的同學幾乎都不太會主動提及這類的話題;當然,我們也不會主動去提。這大概就是最近很流行的「擱置爭議、共創雙贏」吧。和他們交流在這方面,真的是須特別小心,深怕一個不小心就擦槍走火,最後弄得不歡而散。有次,我和某位南大的同學在閒聊時,隨口問了他:「這是你第一次出國嗎?」他的表情突然顯得有點錯愕,似乎不知該如何回答,我也旋即發現我問錯話了,未待他回話,我立刻改問「這是你第一次離開大陸嗎?」他才笑著告訴我,這是他第一次離開大陸。除此之外,和大陸的同學相處起來,並不覺得有什麼特別之處,就好像我們平常的同學相處那樣。大家比較好奇的話題,還是目前兩岸的差異,大陸那邊的學制是碩士班三年、博士班三年,和我們這邊碩士班二年,博士班則端看個人造化三到七年的都大有人在。另外,我聽說他們碩士班畢業去業界,好一點的工作的起薪大約有五千人民幣左右;然而,博士畢業留在學校當講師(相當於我們的助理教授),薪水還不到四千人民幣。這是令我比較驚訝的,在台灣幾乎很難找到碩士起薪比博士高的工作。
最後,就好像昌廷告訴南大的同學,我們在和他們交流的同時,我們自己也在交流。在參加這次的交流活動之前,我們有很多同學其實是本來不認識或是不熟的,因透過這次的交流活動大家才有相聚一起的機會。近年來,光電所和國外大學的交流活動已日益頻繁,然而,本所內的交流活動卻顯得相對匱乏。在繼續和外國大學交流的同時,是否也可以舉辦一些所內的交流活動,或許可以營造所內更優質的氣氛。
|
參與交流活動雙方學生於研討會後交換禮物並合影留念 |
|
SiGe Based Optoelectronics
T. -H. Cheng and C. W. Liu
Graduate Institute of Photonics and Optoelectronics, National Taiwan University
(e-mail) chee@cc.ee.ntu.edu.tw
臺灣大學光電所劉致為教授
It is a long sought goal to integrate ultra-large scale integrated (ULSI)
circuits with the electro-optics to possibly overcome the speed
limitation of electrical interconnects and to add extra
functionalities on Si chip. LEDs and detectors are essential devices
to achieve this goal. The most common material systems used for the
light-emitting technology are III-V-based systems. However, due to
the compatibility with Si electronics, the Si-based light emitters
are the holy grails for the full integration of electrical and
optical devices. Besides the 1.1μm
infrared emission from Si band edge, the addition of Ge into Si can
tune the optical characteristic to longer wavelength infrared
emission from ~1.1μm
to ~2.2μm
wavelength (Fig. 1). The wavelengths of ~1.3μm,
~1.5μm,
and ~2.2μm
infrared have been achieved by Si0.8Ge0.2
Quantum well, Si0.45Ge0.55 quantum dot, Si/Si0.2Ge0.8
type-II heterojunction MIS LED, respectively. Moreover, due to the
high carrier mobility, strong photon absorption, and possible
integration with Si, the Ge based optoelectronic device also
attracts great interest for scientific research and practical
applications, recently. We develop successfully a Ge MIS tunneling
diodes to serve both as a light emitter (~1.8μm)
and a photodetector at a suitable bias.
Figure 2 shows the EL spectra from an Al/SiO2/Ge MIS
tunneling diode under continuous-wave operation at room temperature.
The simulated line shapes from the electron-hole-plasma
recombination model have a good agreement with the experimental
data. The extracted band gaps using the electron hole plasma model
are 40 meV lower than the Ge band gap obtained from Varshni’s
equation from 65 to 345 K. This energy reduction is due to the
longitudinal acoustic (LA) and longitudinal optic (LO) phonon
replica in the radiative recombination, the band gap
renormalization, and a significant amount of heat locally at the
high injection current density. The inset of Fig. 2 depicts the
schematic band diagram of the MIS tunneling diode for EL operation.
At the positive gate voltage, an accumulation layer of majority
electrons is formed at the Ge/SiO2 interface. The
tunneling holes from the Al electrode recombine with electrons
accumulated at the Ge/SiO2 interface and result in the
radiative EL. Interface roughness, phonons, and the spread in
k-space due to localized electrons can provide the extra momentum
during the electron-hole-radiative recombination to emit the
photons.
When the MIS tunneling diode is biased at the inversion region, it
can serve as a photodetector. The typical dark and photo currents of
a bulk Ge (100) MIS photodetector under different wavelength
exposures are shown in Fig. 3. The Ge MIS photodetector has a
responsivity of 0.180 and 0.053 A/W at the wavelengths of 1310 and
1550 nm, respectively. The main reasons for the smaller responsivity
in the MIS structure are the interface states at the
insulator/semiconductor interface and the larger light reflectance
of Al electrode. The inset of Fig. 3 shows the band diagram of the
Ge MIS photodetector at inversion bias. The inversion bias can cause
the deep depletion region to collect the photo-generated carriers.
The minority carriers are generated in the deep depletion region,
tunnel from the active absorption layer to the Al gate electrode via
the trap-assisted tunneling of the LPD oxide, and form the
photocurrent.
Data communication between the Ge MIS LED and the Ge MIS
photodetector is also demonstrated up to 15 Mbit/sec. The speed
limitation is mainly due to the function generator in our system and
can be enhanced with the high-speed modulator or detector. Other photodetectors such as GOG (Ge-on-Glass), GOI (Ge-on-Insulator), and
GOP (Ge-on-Polyimide) are also developed to reach the goal of low
cost, high speed, and extra functionality by wafer bonding and
smart-cut technique. Figure 4 shows the photograph of flexible GOP
structure. The thickness of the transferred Ge layer is about 1.6
μm,
and the surface roughness of the GOP structure is ~ 11 nm after the
smart-cut process. Al with a ring area was evaporated on Ge. Since
the Al ohmic contact has a large area (>0.1 cm2) and the
barrier height between Al and Ge is small (0.1 eV), the effect of
contact resistance is small.
The low-temperature (50℃)
liquid phase deposited (LPD) oxide and Pt gate were used as the gate
stack inside the Al ring.
As time goes by, the solar cell plays a more and more important role
to solve the energy crisis problem. In order to enhance the
efficiency of solar cell, the simulation of optimized grid space,
nano-texture structure, and external strain are added on the solar
cell. The optical methods such as photoluminescence (PL) and
electroluminescence (EL) are used to analysis the characteristics of
solar cell. Minority carrier lifetime and diffusion length can be
measured by temporal response of the electroluminescence and laser
beam induced current (LBIC) to analysis the solar cell
characteristics.
FIG. 1 The emission spectra of SiGe based
LEDs ( bulk Si, bulk Ge, SiGe QD, SiGe QW, and Si/SiGe
heterojunction). It covers the wavelength from 1.1 μm to
2.2 μm. |
FIG. 2. The measured electroluminescence
spectra of a bulk Ge
MIS LED at different temperatures. The
inset shows the schematic band diagram at the
accumulation positive bias for the Ge MIS LED. |
FIG. 3. The dark and photo
I-V
characteristics of a Ge
_100_
MIS photodetector under 1310 and 1550 nm
lightwave exposure. The inset shows the schematic band
diagram at the inversion
(negative)
bias for the Ge MIS photodetector. |
Fig. 4 The photograph of flexible Ge-on-polyimide
structure after bending. The inset shows the device
structure. |
|