Download the full version of the textbook now at textbookfull.

com

Investigation on SiGe Selective Epitaxy for

Source and Drain Engineering in 22 nm CMOS
Technology Node and Beyond Guilei Wang

https://textbookfull.com/product/investigation-on-
sige-selective-epitaxy-for-source-and-drain-
engineering-in-22-nm-cmos-technology-node-and-
beyond-guilei-wang/

Explore and download more textbook at https://textbookfull.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

CMOS-Compatible Key Engineering Devices for High-Speed

Silicon-Based Optical Interconnections Jing Wang

https://textbookfull.com/product/cmos-compatible-key-engineering-
devices-for-high-speed-silicon-based-optical-interconnections-jing-
wang/
textbookfull.com

The Source Drain Engineering of Nanoscale Germanium based

MOS Devices 1st Edition Zhiqiang Li (Auth.)

https://textbookfull.com/product/the-source-drain-engineering-of-
nanoscale-germanium-based-mos-devices-1st-edition-zhiqiang-li-auth/

textbookfull.com

Bioactive Factors and Processing Technology for Cereal

Foods Jing Wang

https://textbookfull.com/product/bioactive-factors-and-processing-
technology-for-cereal-foods-jing-wang/

textbookfull.com

The Runner s Guide to Healthy Feet and Ankles Simple Steps

to Prevent Injury and Run Stronger 1st Edition Dr. Brian
W. Fullem
https://textbookfull.com/product/the-runner-s-guide-to-healthy-feet-
and-ankles-simple-steps-to-prevent-injury-and-run-stronger-1st-
edition-dr-brian-w-fullem/
textbookfull.com
The republic for which it stands : the United States
during Reconstruction and the Gilded Age, 1865-1896 1st
Edition Richard White
https://textbookfull.com/product/the-republic-for-which-it-stands-the-
united-states-during-reconstruction-and-the-gilded-age-1865-1896-1st-
edition-richard-white/
textbookfull.com

Deep Learning for Vision Systems 1st Edition Mohamed

Elgendy

https://textbookfull.com/product/deep-learning-for-vision-systems-1st-
edition-mohamed-elgendy-2/

textbookfull.com

Sets Models and Proofs Ieke Moerdijk

https://textbookfull.com/product/sets-models-and-proofs-ieke-moerdijk/

textbookfull.com

Becoming Confederates: Paths to a New National Loyalty

(Mercer University Lamar Memorial Lectures Ser., 54)
Gallagher
https://textbookfull.com/product/becoming-confederates-paths-to-a-new-
national-loyalty-mercer-university-lamar-memorial-lectures-
ser-54-gallagher/
textbookfull.com

Betrayal Academy The Boys of Westview Academy Crawford

Academy 4 1st Edition Mina Thorne [Thorne

https://textbookfull.com/product/betrayal-academy-the-boys-of-
westview-academy-crawford-academy-4-1st-edition-mina-thorne-thorne/

textbookfull.com
Make It Henna Designs Level 2 Georgia Beth

https://textbookfull.com/product/make-it-henna-designs-
level-2-georgia-beth/

textbookfull.com
Springer Theses
Recognizing Outstanding Ph.D. Research

Guilei Wang

Investigation on SiGe
Selective Epitaxy for
Source and Drain
Engineering in 22 nm
CMOS Technology
Node and Beyond
Springer Theses

Recognizing Outstanding Ph.D. Research

Aims and Scope

The series “Springer Theses” brings together a selection of the very best Ph.D.
theses from around the world and across the physical sciences. Nominated and
endorsed by two recognized specialists, each published volume has been selected
for its scientific excellence and the high impact of its contents for the pertinent field
of research. For greater accessibility to non-specialists, the published versions
include an extended introduction, as well as a foreword by the student’s supervisor
explaining the special relevance of the work for the field. As a whole, the series will
provide a valuable resource both for newcomers to the research fields described,
and for other scientists seeking detailed background information on special
questions. Finally, it provides an accredited documentation of the valuable
contributions made by today’s younger generation of scientists.

Theses are accepted into the series by invited nomination only

and must fulfill all of the following criteria
• They must be written in good English.
• The topic should fall within the confines of Chemistry, Physics, Earth Sciences,
Engineering and related interdisciplinary fields such as Materials, Nanoscience,
Chemical Engineering, Complex Systems and Biophysics.
• The work reported in the thesis must represent a significant scientific advance.
• If the thesis includes previously published material, permission to reproduce this
must be gained from the respective copyright holder.
• They must have been examined and passed during the 12 months prior to
nomination.
• Each thesis should include a foreword by the supervisor outlining the signifi-
cance of its content.
• The theses should have a clearly defined structure including an introduction
accessible to scientists not expert in that particular field.

More information about this series at http://www.springer.com/series/8790

Guilei Wang

Investigation on SiGe
Selective Epitaxy for Source
and Drain Engineering
in 22 nm CMOS Technology
Node and Beyond
Doctoral Thesis accepted by
Chinese Academy of Sciences, Beijing, China

123
Author Supervisors
Dr. Guilei Wang Prof. Henry H. Radamson
Key Laboratory of Microelectronics Institute of Microelectronics
Devices and Integrated Technology Chinese Academy of Sciences
Institute of Microelectronics Beijing, China
Chinese Academy of Sciences
Beijing, China Prof. Chao Zhao
Institute of Microelectronics
University of Chinese Academy of Sciences
Chinese Academy of Sciences
Beijing, China
Beijing, China

ISSN 2190-5053 ISSN 2190-5061 (electronic)

Springer Theses
ISBN 978-981-15-0045-9 ISBN 978-981-15-0046-6 (eBook)
https://doi.org/10.1007/978-981-15-0046-6
© Springer Nature Singapore Pte Ltd. 2019
This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission
or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this
publication does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this
book are believed to be true and accurate at the date of publication. Neither the publisher nor the
authors or the editors give a warranty, expressed or implied, with respect to the material contained
herein or for any errors or omissions that may have been made. The publisher remains neutral with regard
to jurisdictional claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04 Gateway East, Singapore 189721,
Singapore
Supervisors’ Foreword

After five decades, CMOS has been constantly downscaled following the technology
road map, and by crossing to sub-10-nm node, we further approach to the end of road
map in few years. During this period of time, the IC manufacturer has solved many
problems and technical problems, and today’s transistor structure is totally different
than original one manufactured in the Bell Labs. The most important key points for
improving the transistor performance are strain engineering, new high-k & metal
gate, junction formations, and transition from 2D to 3D structure. By considering
these designs, the carrier mobility increased significantly, and short-channel effect
could be decreased in the ultra-small transistor sizes.
Many efforts have been spent to introduce new processing, specially the selec-
tive deposition of SiGe layers in the source/drain regions. The SiGe material has
been used as stressor material to create uniaxial strain in the channel. One major
problem for such design is pattern dependency of the growth where the profile of
SiGe layer varies due to the variation of the chip layout.
This thesis presents research in field of microelectronics which carried out
during four years in Chinese Academy of Science in Institute of Microelectronics.
This work presents integration, growth, and strain engineering of SiGe in source/
drain in 22- and 16-nm node CMOS. The transistor processing has been paid
attention, the steps such as in situ and ex situ cleaning of recess in S/D regions and
surface of the Si fins have been studied, and their impact on the shape of the recess
in S/D and Si fins and quality of SiGe epitaxy have been investigated. This research
provides valuable information about how to measure strain by XRD and TEM
techniques. Another focus has been spent on establishing a kinetic model to predict
the layer profile of SiGe and the pattern dependency behavior of the growth. The
quality of research is excellent, and it contains very important scientific content;
many parts of work have been even awarded in important conferences (EMRS).

Beijing, China Prof. Henry H. Radamson

May 2019 Prof. Chao Zhao

v
Abstract

As CMOS technology is continuously downscaled for decades, the traditional 2D

transistor design has ended in 22-nm technology node, and 3D transistors were
invented to control the short-channel effect, parasitic resistances, and capacitances.
Different methods have been also proposed to engineer the strain in the channel
region and boost the channel mobility. During these technology developments,
selective epitaxial growth (SEG) method has been used to deposit SiGe as stressor
material in source and drain (S/D) region to induce uniaxial strain in channel region.
This thesis is in the field of nanoelectronics and presents device processing and
epitaxy modeling. It presents the key parameters of high-quality SiGe selective
epitaxial growth with a focus on its pattern dependency behavior and the key
integration issues in the transistor structures. The main research outcome of this
thesis is as follows:
(1) It is found that in situ cleaning has large impact on dopant diffusion and to
shape of the S/D recess as well as the Si fins. For 22-nm planar devices,
pre-baking at 800 °C is applicable, whereas pre-baking at lower temperature
as 780–800 °C is necessary to remove native oxide but to avoid any damage to
Si fins which is essential for high-quality SiGe SEG. The S/D Si loss at high
pre-baking temperature is attributed to HCl etching because of residual Cl
atoms in H2 atmosphere. At lower pre-baking pressure, the morphology change
occurs, and this is caused by the migration of Si atoms originating from the
thermal mismatch between SiO2 and Si.
(2) The ex situ cleaning has also important role on the SiGe epitaxial layer. It is
observed that rinsing time in diluted buffer oxide etch (DBOE) and the amount
of HCl affect SiGe SEG significantly. These two parameters need to be opti-
mized in order to avoid mushroom-like growth on both dummy gate and Si3N4
isolation spacers.
(3) Selective growth Si1−xGex (0.25  x  0.35) with boron concentration of
1−3  1020 cm−3 was investigated and optimized for S/D regions of 22-nm
node planar transistors and 14-nm node FinFETs. A three-layer structure of
Si1−xGex was designed in S/D areas. A strain-relaxed buffer (SRB) layer as

vii
viii Abstract

bottom layer, the middle layer with high Ge content (Si0.65Ge0.35 or Si0.60
Ge0.40), was intended as stressor material to fill the recess and a cap layer with
low Ge content (Si0.75Ge0.25 or Si0.80Ge0.20) as a sacrificial material for the
Ni-silicidation process. The purpose of the cap layer was to avoid strain
reduction in the channel region when the layer is consumed. The characteristics
of transistors were measured, and the results were explained in terms of growth
conditions.
(4) A kinetic gas model was developed to evaluate the pattern dependency of the
SiGe SEG process and to predict the layer profile within transistor arrays in a
chip and over different locations on a wafer. The input parameters include
growth temperature, partial pressures of reactant gases, and chip layout. By
using this model, the number of test wafers for epitaxy experiments can be
decreased significantly when the chips are located with a distance from the edge
of the 200-mm wafer. SiGe layers with poor epi-quality were obtained when the
coverage of exposed Si of the chip was below 1%. In such chips, high Ge
content with layer thicknesses above the critical thickness was observed. When
the epitaxy process parameters can be predicted by the model for a desired
epi-profile in an advanced chip design, fast and cost-effective process devel-
opment can be achieved in the mass production.

Keywords CMOS Planar transistors FinFET Strain SiGe






Selective epitaxial growth RPCVD Pattern dependency
Acknowledgements

As time flies, my time of doctoral study is drawing to a close in the twinkling of an

eye. I still remember that I felt in tension when I just passed my Ph.D. entrance
examination many years ago, but for now I still feel uneasy at the end of my Ph.D.
study. This kind of nervous mood is like holding a newborn baby, happy and
bewildered. I once went astray on the way of learning, and I was not the
“Outstanding Student” in everyone’s eyes. I am so grateful that I could continue my
study after leaving the college for many years. Hereby, I would like to extend my
most sincere gratitude to all the teachers, classmates, relatives, friends, and col-
leagues who have cared, supported, and helped me.
First of all, I want to give my special thanks to Prof. Henry H. Radamson. He has
given me a lot of selfless help and specific guidance in my study and work. Thanks
to him for bringing me into the world of extension and all he has done for me;
I have gained a lot.
I also would like to express particular thanks to Prof. Chao Zhao. With his
careful guidance, I finally achieve this. I am always infected by his profound
knowledge, broad vision, and rigorous scientific research attitude. For example,
when I met difficulties during my work and study, I was enlightened by him with
the direction and motivation to move forward. When I made some achievements,
I was pointed out my shortcomings needed to improve. From him, I appreciate the
style of the master and learn the experience and method of being a man, doing a job,
and engaging in scholarship that will benefit me all my life.
I would also like to express my special thanks to some professors, Jun Luo,
Junfeng Li, Huaxiang Yin, Wenwu Wang, Jiang Yan, Huilong Zhu, Qiuxia Xu, and
Tianchun Ye, for their support and help in my study and work.
In the meanwhile, I would like to thank my colleagues in the Process Integration
Department for their help; they are Changliang Qin, Yefeng Xu, Yanbo Zhang, Bo
Tang, Zhaoyun Tang, Jing Xu, Zhiguo Zhao, Yongkui Zhang, Hong Yang, Xiaolei
Wang, and Hongli Wang. Thanks for their hard and serious work in integration, the
experimental result of SiGe source and drain in this paper can be verified in the
devices.

ix
x Acknowledgements

Thanks to my colleagues in the process modules, Tao Yang, Jinbiao Liu,

Jianghao Han, Hushan, Cui, Qiang Xu, Jinjuan Xiang, Xiaobin He, Wenjuan
Xiong, Lingkuan Meng, Jianfeng Gao, Guobin Bai, Tingting Li, Jiahan Yu, Junjie
Li, Chunlong Li, Peizhen Hong, ChengJia, Peng Zhang, Hao Zhang, Yihong Lu,
Haojie Jiang, Ying Luo, Yuyin Zhao, Mingzheng Ding, and others, for giving the
help in the development of process, and thanks to them and all the stall in the
process line for their hard work, which makes sure that the experimental work in
this paper can be carried out smoothly.
Thanks to the hardware engineers, Qihui Zhang, Guanghui Tian, Lin Li, Chen
Li, Yunling Ding, Shi Liu, and others, for guaranteeing the normal operation of the
equipment and supporting, which makes sure that the experimental work in this
paper can be carried out smoothly.
I would like to thank Profs. Renrong Liang, Jing Wang, and Jun Xu from
Tsinghua University for their help in the suggestion of SiGe S/D integration, and
thank Profession Ping Yang from the Synchronized Light Source Laboratory in the
University of Singapore for their guidance and help in the HRXRD test of SiGe thin
film on the Si Fin.
Finally, I would also like to thank my parents, parents-in-law, wife, and
daughter, and appreciate them for their expectation, commitment, selfless dedica-
tion, and support. Especially for my wife, I thank her for taking care of all the
family affairs alone while we are living in two different cities, and for supporting
and encouraging me to devote myself to my work and study to complete my Ph.D.
Life is so short, like a white pony’s shadow crossing a crevice. Also, our ancient
said “Having heard the way in the morning, one may die content in the evening.” In
my lifetime, the end of the Ph.D. career does not mean the end of “Way”
(Achievements reached), but as indicated in the beginning of my acknowledge that
I will still cherish the desire for knowledge and vision of a better future, like
holding a newborn baby, and continue to do more meaningful things in my limited
lifetime.

May 2016
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
Contents

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Development Status and Challenges of Integrated Circuit . . . . . . . 1
1.2 Major Research Works in This Thesis . . . . . . . . . . . . . . . . . . . . . 5
1.3 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2 Strained Silicon Technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.2 Physics of Strained Silicon Technology . . . . . . . . . . . . . . . . . . . . 10
2.2.1 Effect of Strain on Electron Mobility . . . . . . . . . . . . . . . . 10
2.2.2 Effect of Strain on Hole Mobility . . . . . . . . . . . . . . . . . . . 11
2.2.3 Effect of Strain Direction on Mobility Improvement . . . . . 13
2.3 Classification of Strain Techniques . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.1 Substrate-Induced Strain . . . . . . . . . . . . . . . . . . . . . . . . . . 14
2.3.2 Process-Induced Strain . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3 Epitaxial Growth of SiGe Thin Films . . . . . . . . . . . . . . . . . . . . . . . . 23
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.1.1 SiGe Crystal Structure and Strain . . . . . . . . . . . . . . . . . . . 24
3.1.2 Critical Thickness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.1.3 Epitaxy and Its Main Growth Techniques . . . . . . . . . . . . . 26
3.1.4 Non-selective Epitaxy and Selective Epitaxy . . . . . . . . . . . 28
3.2 SiGe Selective Epitaxy Using RPCVD Technology . . . . . . . . . . . 28
3.2.1 RPCVD Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.2.2 Epitaxial Growth of SiGe Films . . . . . . . . . . . . . . . . . . . . 29

xi
xii Contents

3.3 Characterization of SiGe Films . . . . . . . . . . . . . . . . . . . . . . . . . . 31

3.3.1 Analysis of SiGe/Si Interface . . . . . . . . . . . . . . . . . . . . . . 32
3.3.2 HRXRD Analysis of SiGe Films . . . . . . . . . . . . . . . . . . . 33
3.3.3 Surface Roughness of Strained SiGe Films . . . . . . . . . . . . 35
3.3.4 TEM Analysis of SiGe Films . . . . . . . . . . . . . . . . . . . . . . 37
3.4 Main Factors Affecting the Growth and Strain of SiGe Films . . . . 38
3.4.1 Experimental Process of SiGe Growth . . . . . . . . . . . . . . . 38
3.4.2 Influence of Reaction Temperature on the Strain
of SiGe Films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 38
3.4.3 Influence of Reaction Pressure on the Growth
of SiGe Thin Films . . . . . . . . . . . . . . . . . . . . . . . . . .... 40
3.4.4 Influence of HCl Gas on the Growth of SiGe Films . . .... 41
3.4.5 Influence of Doping Concentration on the Strain
of SiGe Films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.4.6 Effect of Ge Concent on the Strain of SiGe Films . . . . . . . 45
3.5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
4 SiGe S/D Integration and Device Verification . . . . . . . . . . . . . . . . . . 49
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
4.1.1 Development of SiGe S/D Strain Technology . . . . . . . . . . 49
4.1.2 Summary of SiGe S/D Technology . . . . . . . . . . . . . . . . . . 50
4.1.3 Challenges of SiGe S/D Technology . . . . . . . . . . . . . . . . . 52
4.2 Integration of SiGe in S/D in 22 nm Planar MOSFETs . . . . . . . . . 53
4.2.1 Device Integration Process, Experiment Details . . . . . . . . . 53
4.2.2 S/D “Ʃ” Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.2.3 S/D SiGe Selective Epitaxial Growth . . . . . . . . . . . . . . . . 56
4.3 Integration of SiGe S/D in 16 nm FinFET Device . . . . . . . . . . . . 56
4.3.1 Device Preparation and Experimental Details . . . . . . . . . . 56
4.3.2 Selective Epitaxial Growth of SiGe/Si in 16 nm FinFET
Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 57
4.4 Selective Epitaxial Process of SiGe for 22 and 16 nm
MOSFETs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 60
4.4.1 The Influence of Ex-situ Cleaning on SiGe Epitaxy . . . . .. 60
4.4.2 The Influence of Pre-baking Process on SiGe Integration
in S/D . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 64
4.4.3 SiGe S/D Epitaxial Selectivity . . . . . . . . . . . . . . . . . . . .. 73
4.5 Strain Analysis of Integrated SiGe S/D in Nano Scaled
Transistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 77
4.5.1 TCAD Simulation of Strain in Integrated SiGe S/D
Transistors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 77
4.5.2 HRXRD Analysis of Strain of Integrated SiGe S/D
Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 79
Contents xiii

4.6 TEM Analysis of Integrated SiGe S/D Devices . . . . . . . . . . . . . . 83

4.6.1 TEM Analysis of SiGe S/D in 22 nm Planar Devices . . . . 83
4.6.2 TEM Analysis of 16 nm FinFET SiGe S/D Devices . . . . . 84
4.7 SiGe S/D Electrical Performance Verification . . . . . . . . . . . . . . . . 87
4.7.1 Electrical Performance Verification of 22 nm Planar SiGe
S/D Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 87
4.7.2 Electrical Performance of 16 nm FinFET SiGe S/D
Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 88
4.8 Summary of This Chapter . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 89
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 90
5 Pattern Dependency of SiGe Layers Selective Epitaxy Growth . . . . . 93
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
5.1.1 The Generation of Pattern Dependency . . . . . . . . . . . . . . . 93
5.1.2 Recent Progress of Pattern Dependency . . . . . . . . . . . . . . 94
5.2 Evaluation and Calculation Model of Pattern Dependency . . . . . . 96
5.2.1 Model Calculation of 22 nm Planar Device . . . . . . . . . . . . 96
5.2.2 Model Calculation of 16 nm FinFET . . . . . . . . . . . . . . . . 99
5.3 Impact of Pattern Dependency on the SiGe Selective Epitaxy . . . . 101
5.3.1 Experimental Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
5.3.2 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . 103
5.4 Impact of Pattern Dependency on the Electrical Performance
of Integrated Devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.4.1 Experimental Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
5.4.2 Results and Discussions . . . . . . . . . . . . . . . . . . . . . . . . . . 106
5.5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
6 Conclusions and Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
6.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
6.2 Prospect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Abbreviations

AFM Atomic force microscopy

APCVD Atmospheric pressure chemical vapor deposition
APM NH4OH + H2O2 + H2O
B2H6 Diborane
C Carbon
CESL Contact etch stop layer
CMOS Complementary metal-oxide semiconductor
CMP Chemical-mechanical polishing
CVD Chemical vapor deposition
DBOE Diluted buffer oxide etch
DFM Design for manufacturability
DG Double gate
DHF Diluted hydrofluoric acid
DIBL Drain-induced barrier lowering
DSL Dual stress liner
DTMAH Diluted tetramethylammonium hydroxide
EBL Electron beam lithography
EDX Energy-dispersive X-ray detector
EOT Equivalent oxide thickness
ESD Electrostatic discharge
FIB Focused ion beam
FinFET Fin field-effect transistor
FWHM Full-width half-maximum
Ge Germanium
GeH4 Germane
HARP High-aspect-ratio process
HBT Heterojunction bipolar transistor
HCI Hot-carrier injection effect
HH Heavy-hole
HKMG High-k metal gate

xv
xvi Abbreviations

HOI Heterostructure on insulator

HRRLM High-resolution reciprocal lattice mapping
HRTEM High-resolution transmission electron microscopy
HRXRD High-resolution X-ray diffraction
IC Integrated circuit
LDD Lightly doped drain
LH Light-hole
LPCVD Low-pressure chemical vapor deposition
LPE Liquid-phase epitaxy
MBE Molecular beam epitaxy
MFC Mass flow controller
MGIS Metal gate-induced strain
MOS Metal-oxide semiconductor
NMOS N-type metal-oxide semiconductor
NSEG Non-selective epitaxial growth
PMOS P-type metal-oxide semiconductor
R&D Research and Development
RMS Root mean square
RPCVD Reduced pressure chemical vapor deposition
S/D Source/Drain
SCE Short-channel effect
SEG Selective epitaxial growth
Si Silicon
SiC Si1−yCy Silicon carbon alloy
SiGe Si1−xGex Silicon germanium alloy
SiH2Cl2 Dichlorosilane, DCS
SiH4 Silane
SIMS Secondary ion mass spectrometry
SMT Stress memorization technique
SOI Silicon on insulator
SPE Solid-phase epitaxy
SPM H2SO4 + H2O2
SRB Strain-relaxed buffer
SSDOI Strained silicon directly on insulator
SSLS Singapore synchrotron light source
SSRF Shanghai synchrotron radiation facility
STI Shallow trench isolation
TCAD Technology computer-aided design
TEM Transmission electron microscopy
TFET Tunnel field-effect transistor
UHVCVD Ultra-high vacuum chemical vapor deposition
VPE Vapor-phase epitaxy
XRD X-ray diffraction
Chapter 1
Introduction

Currently, we are living in the modern time and era of information technology. As
the foundation of information industry, the development of integrate circuit (IC) is
inseparable from our daily life. The emergence of new electronic applications are
commonly in used smart phones, computers, drones, and unmanned vehicle which is
all attributed to great advances and innovations of IC technology. The development
history of IC in recent decades is not only the result of the combination of technology
promotion and market demand, but also the history of the transformation of techno-
logical innovation into productivity [1]. In addition, the progressive development of
integrated circuit has always been considered as the key to promote the development
of information technology, and the technology node and production scale. Therefore,
the level of IC development are important metrics for evaluating the overall national
strength and level of economic growth.

1.1 Development Status and Challenges of Integrated

Circuit

On December 16, 1947, the first transistor was invented at Bell Laboratories by
William Shockley, John Bardeen and Walter Brattain. This was perhaps the most
important electronics event of the 20th century, as it later made possible the inte-
grated circuit and microprocessor that are the basis of modern electronics [2]. On
September 12, 1958, Jack Kilby of Texas Instruments built the first monolithic inte-
grated circuit using germanium mesa. Although the proposed IC comprises just one
transistor, it still marks the arrival of IC era for the world. In 1960, the first planar
integrated circuit is fabricated at Fairchild Semiconductor in the United States. Later
in 1963, Frank Wanlass of the Fairchild R&D Laboratory showed that logic circuits
combining p-channel and n-channel MOS transistors in a complementary symmetry
circuit configuration CMOS, which has still been used today owing to its superior
performance. In 1965, Gorden Moore proposed the famous “Moore’s Law”, which is
© Springer Nature Singapore Pte Ltd. 2019 1
G. Wang, Investigation on SiGe Selective Epitaxy for Source and Drain Engineering
in 22 nm CMOS Technology Node and Beyond, Springer Theses,
https://doi.org/10.1007/978-981-15-0046-6_1
2 1 Introduction

the observation that the number of transistors per unit area the degree of integration
would double every 18 months. In each technology generation, transistor dimension
is scaled by 30% (0.7x) and the chip performance is doubled [3]. In 1968, Intel was
founded by Robert Noyce, Gorden Moore, and Andrew Grove, indicating the start
of large-scale commercialization of LSI CMOS technology.
In the past decades, semiconductor technology has developed continuously in
accordance with the device scaling theory proposed by Dennard et al. [4, 5], which
has stimulated the increment of integration scale and improvement of performance
for integrated circuit. With the development of CMOS IC technology over the past
decade, it gradually becomes difficult to improve IC performance just through con-
tinuous scaling-down of device dimensions. Some of the fabrications techniques
that have been used for a long time become hard to maintain the scaling-down trend.
Thus, new process modules, new materials, and new device structures should be
developed in time to overcome these upcoming challenges and to continue Moore’s
Law. Figure 1.1 shows the key technology roadmap for CMOS integrated circuit.
The last 20 years have seen an aggressive scaling-down of silicon CMOS technol-
ogy from 1 μm planar structure to advanced 20 nm 3D FinFETs. Especially, when
entering the 45 nm technology node and below, the reduction of the physical gate

Fig. 1.1 Key technology roadmap for CMOS integrated circuit

1.1 Development Status and Challenges of Integrated Circuit 3

length brings more technical challenges to nanoscale CMOS devices [6, 7]. The main
technical challenges are listed as follows:
1. Gate electrode: The polysilicon depletion effect (PDE) induced by the traditional
polysilicon gate results in an increase of equivalent gate oxide thickness, reduced
gate control and occurrence of short channel effects (SCE) particularly in small-
size MOSFETs. Therefore, traditional polysilicon can’t meet the requirements
of gate control for small-size MOSFETs.
2. Gate dielectric: As the device size continues shrinking, the gate oxide thickness
needs to be continuously thinned down to improve short channel effects immu-
nity and gate controllability. However, using ultra-thin gate oxide exponentially
increases the tunneling leakage current, followed by increasing of power con-
sumption. Thus, the device reliability issues become more prominent. The dis-
tributed defects and traps in interfacial oxide cause significant interfacial and
Coulomb’s scattering, resulting in serious degradation of carrier mobility.
3. Channel/substrate: the reduction of gate length leads to serious short channel
effects, drain-induced barrier lowering (DIBL), deterioration of the sub-threshold
characteristics, etc. Due to the existence of parasitic effects, device drive current
no longer increases linearly. Hence, carrier mobility of the channel needs to be
further improved to boost device performance.
4. Source/drain (S/D) regions: In order to enhance the channel carrier mobility,
strain engineering by selective epitaxial growth of heterojunction materials in
S/D regions has been widely employed to induce uniaxial strain in channel region
and to boost carrier mobility. Besides, parasitic S/D series resistance is difficult
to scale down proportionally with the shrinkage of device size, which becomes
a more important contributor to the total resistance of a transistor and inevitably
impacts the drive current. Thus, suitable silicide processes corresponding to dif-
ferent S/D materials should be implemented to reduce the parasitic resistance.
In an effort to overcome the above-mentioned technical challenges, new tech-
nologies have been employed into each product generation to continue the trend of
Moore’s Law. For example, strained silicon technology has been utilized to improve
the current drive capability of transistors through enhancing the carrier mobility of
channel. To mitigate the impact of S/D parasitic series resistance and capacitance
on device performance, Schottky-barrier silicide S/D has been proposed to replace
the ion-implanted S/D in conventional CMOS process. In order to improve the short
channel effects, ultra-shallow and halo implantation techniques are applied in S/D
regions [8]. And in order to restrain hot-carrier injection effect (HCI), lightly-doped-
drain (LDD) implantation and other techniques are also suggested [9].
Later, dozens of new semiconductor materials are introduced to resolve the process
problems and improve the performance of the device. In 2007, Intel introduced high-
κ dielectric and metal gate in 45 nm planar technology node to reduce the device’s
equivalent oxide thickness (EOT) and to enhance gate control capability [10]. In
2009, Intel unveiled the second generation of HKMG process in its 32 nm technology
node, for the purpose of further EOT scaling and device performance improvement
[11]. When CMOS process enters 10 nm node and below, new channel materials
4 1 Introduction

with higher carrier mobility than silicon, such as Ge, III–V compounds and so on,
are likely to be introduced to further promote the development of CMOS integrated
circuit technology [12].
With respect to novel device structures, several new types of MOSFETs have been
reported, such as silicon on insulator (SOI) MOS devices [13–15], planar double-gate
(DG) MOSFETs [16], fin field-effect transistor (FinFETs) [17],  gate and multi-gate
MOSFETs [18], tunnel field-effect transistor (TFET) [19] and so on. Among these
new device structures, the 22 nm FinFET product proposed by Intel in 2011 is the most
attractive one. FinFET devices differ from planar structure counterparts by the three-
dimensional Fin structure, which leads to further improvement of device performance
and integration density of IC [20]. In 2013, Intel developed FinFET technology to
16 nm node [21], and optimized the Fin structure through further reduction of Fin
pitch to increase integration density. At present, research and development (R&D)
of Intel’s 10 nm technology is in progress.
In brief, the development of CMOS IC technology and the extension of Moore’s
Law are inseparable from the breakthrough of new process modules, new materials
and new device structures. In each technology node, technical schemes always follow
the previous technology roadmap with minor changes. If technical changes are too
significant, mass production would become more challenging and the manufacturing
cost would rise simultaneously. Thus, industries always give priority to technology
solutions adapted for large-scale production instead of those with major changes.
For example, strained silicon technology is one of the technical solutions suitable for
large-scale production [22]. As shown in Fig. 1.2, strain has had tremendous impact
in advancing the transistor scaling roadmap for generations after the 130 nm node
(90, 65, 45, 32 nm) [23]. One way to carry out strain engineering in the transistor
structure is using SiGe alloy as stressor material in S/D areas. Epitaxial growth
of SiGe in S/D can generate compressive strain in PMOS channel region which can
enhance to hole mobility, leading to performance enhancement of CMOS transistors.
Integration of highly strained SiGe in S/D presents excellent compatibility with
current CMOS technology and plays an increasingly important role in advanced
manufacturing processes.

Fig. 1.2 Importance of

strain in modern CMOS
scaling [23]
1.2 Major Research Works in This Thesis 5

1.2 Major Research Works in This Thesis

In this paper, integration of strained SiGe in S/D for 22 nm planar and 16 nm FinFET
devices has been systematically investigated to meet the requirements of the “Na-
tional S&T Major Project 02”. This research focuses on selective epitaxial growth
of high-quality strained SiGe thin films and its process integration and industrial-
relevant applications. The research content in the work is closely relevant to practical
issues in industry. The major research works are listed as follows:
1. Epitaxial growth of high-quality, defect-free strained SiGe thin films on an 8-inch
process platform using reduced pressure chemical vapor deposition (RPCVD)
and its process optimization for high epitaxial quality. Key factors that affects
the high-quality selective epitaxial growth of SiGe films are analyzed to promote
the applications of strained SiGe technology.
2. Key technical issues for selective epitaxial growth of SiGe film on S/D are studied.
The research contents include the influence of cleaning process prior to epitaxy
and prebake during epitaxial growth on the surface morphology of SiGe film in
S/D.
3. The influence of the DBOE rinse time and the amount of HCl vapor on the
selectivity of SiGe epitaxial growth are studied.
4. Electrical characteristics for 22 nm planar and 16 nm FinFET devices with SiGe
S/D are systematically analyzed. High-resolution X-ray diffraction (HRXRD)
detection technique and Technology Computer Aided Design (TCAD) simu-
lations are performed to analyze strain variation during SiGe epitaxial growth
on S/D. During the source-drain epitaxial process, SiGe films with several Ge
concentration gradients are experimentally grown as the strained S/D.
5. The reaction kinetics and growth mechanisms of SiGe selective epitaxy are stud-
ied. Meanwhile, the microscopic and macroscopic effects of the pattern density
during selective epitaxial growth are explained in detail. By studying the diffu-
sion kinetics of the reaction gas and the established calculation model, the causes
for the dissimilarity of the properties of selective epitaxial grown SiGe films in
different regions are also explained. The results of the calculation model are
verified by the film growth rate and Ge composition distribution obtained from
experimental test. This could establish a design for manufacturability (DFM)
reference for SiGe selective epitaxial growth integration and layout design to
resolve some practical issues for the application of SiGe technology.

1.3 Thesis Organization

According to the main research contents, the chapters in this thesis are arranged as
follows:
This chapter presents the background of the thesis, the significance of the work,
the main research contents and the organization of the thesis.
Visit https://textbookfull.com
now to explore a rich
collection of eBooks, textbook
and enjoy exciting offers!
6 1 Introduction

Chapter 2 presents the development and classifications of strained silicon tech-

nology, mechanisms of carrier mobility enhancement via strained silicon technology
and main strain engineering techniques.
Chapter 3 focuses on epitaxial growth of high-quality SiGe films on S/D regions.
The mechanisms of selective epitaxial growth of thin films are studied, and key
process parameters affecting the growth of high quality SiGe films are also discussed.
In Chap. 4, the influence of cleaning process prior to epitaxy and pre-baking during
epitaxial growth are systematically studied. In order to introduce the maximum stress
on channel, SiGe film in S/D regions is sequentially grown in the sequence of buffer
layer, core layer and sacrificial layer. The distribution and variation of strain during
SiGe epitaxial growth on S/D were studied by HRXRD test technology and TCAD
simulations. Integration of strained SiGe in S/D for 22 nm planar and 16 nm FinFET
devices and electrical characterization have been performed and discussed.
In Chap. 5, based on the reaction kinetics and growth mechanism of SiGe selective
epitaxy, a calculation model is established to study the influence of microscopic and
macroscopic effects of the pattern density on the epitaxy results during the selective
epitaxy process. And the theoretical calculation model is also verified through the
test results of HRXRD and EDX.
In Chap. 6, key contributions of this research are summarized and future prospects
for the applications of SiGe selective epitaxy technology are offered.

1.4 Summary

In this chapter, the development history of integrated circuits is briefly reviewed,

and a series of CMOS scaling challenges as well as the corresponding technical
solutions carried out in large-scale manufacturing are illustrated. Strained silicon
technology adopted in 90 nm technology node is one of the key technologies for
device driving capability improvement. Among the family of strained silicon tech-
nologies, SiGe source/drain is compatible with the CMOS process, and it plays an
increasingly important role in advanced manufacturing processes. Therefore, this
thesis focuses on the source/drain engineering. Finally, the main research work and
the thesis organization are given.

References

1. 王阳元 Wang Y, 张兴Zhang X, 刘晓彦 Liu X et al (2008). 32 nm 及其以下技术节点 CMOS

技术中的新工艺及新结构器件. Sci China Ser F-Inf Sci 38(6):921–932 (in Chinese)
2. Shockley W, Bardeen J, Brattain W (1947) The first transistor. Bell Laboratories, 16 Dec 1947
3. Moore G (1965) Cramming more components onto integrated circuits. Electronics 38(8)
4. Dennard RH, Rideout V, Bassous E, LeBlanc A (1974) Design of ion-implanted MOSFET’s
with very small physical dimensions. IEEE J Solid-State Circuits 9:256–268
References 7

5. Frank DJ, Dennard RH, Nowak E, Solomon PM, Taur Y, Wong H-SP (2001) Device scaling
limits of Si MOSFETs and their application dependencies. Proc IEEE 89:259–288
6. Taur Y (1999) CMOS scaling beyond 0.1 μm: how far can it go? In: International symposium
on VLSI technology, systems, and applications, 1999, pp 6–9
7. Kubicek S, de Meyer K (2002) CMOS scaling to 25 nm gate lengths. In: The fourth international
conference on advanced semiconductor devices and microsystems, 2002, pp 259–270
8. Ogura S, Codella CF, Rovedo N, Shepard JF, Riseman J (1982) A half micron MOSFET using
double implanted LDD. In: 1982 international electron devices meeting, 1982, pp 718–721
9. Ogura S, Tsang PJ, Walker WW, Critchlow DL, Shepard JF (1980) Design and characteristics
of the lightly doped drain-source (LDD) insulated gate field-effect transistor. IEEE Trans Electr
Dev 27:1359–1367
10. Auth C, Cappellani A, Chun JS, Dalis A, Davis A, Ghani T et al (2008) 45 nm high-k+ metal
gate strain-enhanced transistors. In: 2008 symposium on VLSI technology, pp 128–129
11. Packan P, Akbar S, Armstrong M, Bergstrom D, Brazier M, Deshpande H et al (2009) High
performance 32 nm logic technology featuring 2nd generation high-k+ metal gate transistors.
In: 2009 IEEE international electron devices meeting (IEDM), pp 1–4
12. Del Alamo JA (2011) Nanometre-scale electronics with III-V compound semiconductors.
Nature 479:317–323
13. Lam H, Tasch A Jr, Holloway T (1980) Characteristics of MOSFETs fabricated in laser-
recrystallized polysilicon islands with a retaining wall structure on an insulating substrate.
IEEE Electron Dev Lett 1:206–208
14. Chen CL, Chen CK, Vitale SA (2011) SOI circuits powered by embedded solar cell. In: 2011
IEEE international SOI conference (SOI), 2011, pp 1-2
15. Chen JH, Helmi SR, Mohammadi S (2011) Millimeter-wave power amplifiers in 45 nm CMOS
SOI technology. In: 2011 IEEE international SOI conference (SOI), 2011, pp 1–2
16. Shruti K, Thomas DM, Samuel PC, Kumar V (2011) Analysis of single halo double gate MOS-
FETs using high-k dielectrics. In: 2011 3rd international conference on electronics computer
technology (ICECT), 2011, pp 26–30
17. Hisamoto D, Lee W-C, Kedzierski J, Anderson E, Takeuchi H, Asano K et al (1998) A folded-
channel MOSFET for deep-sub-tenth micron era. IEDM Tech Dig 1998:1032–1034
18. Abraham D, Poehler T (1965) Double-gate thin-film transistor. Electron Lett 1:49
19. Ionescu AM, Riel H (2011) Tunnel field-effect transistors as energy-efficient electronic
switches. Nature 479:329–337
20. Auth C, Allen C, Blattner A, Bergstrom D, Brazier M, Bost M et al (2012) A 22 nm high
performance and low-power CMOS technology featuring fully-depleted tri-gate transistors,
self-aligned contacts and high density MIM capacitors. In: 2012 symposium on VLSI technol-
ogy (VLSIT), pp 131–132
21. Natarajan S, Agostinelli M, Akbar S, Bost M, Bowonder A, Chikarmane V et al (2014) A 14 nm
logic technology featuring 2nd-generation FinFET, air-gapped interconnects, self-aligned dou-
ble patterning and a 0.0588 μm 2 SRAM cell size. In: 2014 IEEE international electron devices
meeting (IEDM), pp 3.7. 1–3.7. 3
22. Mohta N, Thompson SE (2005) Mobility enhancement. IEEE Circuits Dev Mag 21:18–23
23. Kuhn KJ, Liu MY, Kennel H (2010) Technology options for 22 nm and beyond. In: Proceeding
of the 10th international workshop on junction technology (IWJT-2010), 2010, pp 10–11
Chapter 2
Strained Silicon Technology

2.1 Introduction

With transistor dimension shrinking into nanoscale regime as described in the pre-
vious chapter, conventional gate oxide thickness has been scaled near the thickness
limit of 1 nm. This scaling trend brings issues about power consumption, transistor
density and off-leakage current, together with carrier mobility degradation. Further-
more, the continuous scaling-down of device dimensions and further performance
enhancement are facing more and more severe challenges. Therefore, to overcome
these challenges occurred during technology development, various technical solu-
tions can be implemented to improve device performance. One of the important and
simple technical solutions is strained silicon technology. For device driving capability
improvement, strained silicon technology is applied to enhance the carrier mobility
on channel to compensate the mobility degradation caused by the scaling-down of
device dimensions [1–4].
In 2002, Intel unveiled its microprocessors using strained silicon technology at the
90 nm process node [5]. This is the first process in the industry to implement strained
silicon in production. Since then, each process node has adopted different strained
silicon technologies to improve device performance. For example, since 65 nm tech-
nology node, semiconductor industry has applied strained silicon technology into
CMOS production [5], including strained SiGe on source/drain and silicon nitride
capping layer. In addition, metal gate-induced strain is introduced in 45 nm process
node [6]. Among the current approaches for transistor performance improvement,
the mobility enhancement via strained silicon technology has had tremendous impact
in device performance enhancement and has become one of the important techniques
to extend Moore’s Law [7].

© Springer Nature Singapore Pte Ltd. 2019 9

G. Wang, Investigation on SiGe Selective Epitaxy for Source and Drain Engineering
in 22 nm CMOS Technology Node and Beyond, Springer Theses,
https://doi.org/10.1007/978-981-15-0046-6_2
10 2 Strained Silicon Technology

2.2 Physics of Strained Silicon Technology

As we all know, the relationship between the drive current and the mobility of MOS-
FETs is shown in the following equation [8]:

W μeff Cox W μeff Kox ε0

ID = (VG − VT )2 = (VG − VT )2
2L 2L
where I D is the drive current, W is the channel gate width, L is the channel gate
length, μeff is the effective carrier mobility, C ox is the gate oxide capacitance per unit
area, V G is the gate voltage and V T is the threshold voltage. With the down-scaling
of CMOS transistors, especially the aggressive shrinkage of channel length, channel
doping concentration is increased to suppress the source-drain punch-through and
short-channel effect. However, this results in the degradation of μeff and threshold
voltage roll-up. Therefore, in order to further enhance device performance, strained
silicon technology, which can induce additional strain in channel region, should be
performed to increase carrier mobility. This approach could compensate the mobility
degradation due to several contributing factors, such as significant Coulomb’s scat-
tering which is originated from high channel doping concentration, and increased
effective gate electric field and enhanced interface scattering caused by the down-
scaling of gate dielectrics [1, 2, 4].
Based on the study of the impact of strain on carrier mobility [9–11], the carrier
mobility can be expressed by Eq. (2.2):
qτ
μ= (2.2)
m∗
where q is the electron charge, 1/τ is the scattering rate and m* is the conductivity
effective mass. Strain enhances the mobility by reducing the conductivity effective
mass and/or the scattering rate. For electrons, both mass and scattering changes are
generally accepted as important for mobility enhancement [12]. However, for holes,
mainly the mass change due to band warping and repopulation plays a significant role
in (<1 GPa) stress level since strain induces smaller valence band splitting compared
to the conduction band [13].

2.2.1 Effect of Strain on Electron Mobility

For electron transport in bulk Si at room temperature, the conduction band is com-
prised of six degenerate valleys, as shown in Fig. 2.1a. These valleys are of equal
energy, as shown by 6 in Fig. 2.1c. The degeneracy reflects the cubic symmetry of
the Si lattice. The effective mass for any direction is the reciprocal of the curvature
of the electron energy function in that direction. Consequently, the effective mass
of each ellipsoid is anisotropic, with the transverse mass (perpendicular to the axis)
2.2 Physics of Strained Silicon Technology 11

Fig. 2.1 Electron conduction band valleys for a relaxed silicon, b strained silicon with biaxial
tensile strain. c Energy level at the bottom of the six conduction band valleys. Application of
advantageous strain splits the energy level as shown, removing the degeneracy (i.e., the equivalence
in energy) between the 2 and 4 valleys

given by mt = 0.19 mo being significantly smaller than the longitudinal mass (paral-
lel to the axis) given by ml = 0.98 mo , where mo is the free electron mass [14]. For
unstressed bulk Si, the total electron conductivity mass, m*, is obtained by adding
the contributions of the six degenerate valleys and is given by
  −1
1 2 4
m∗ = +
6 ml mt

For MOSFETs on a (001) wafer, advantageous strain removes the degeneracy

between the four in-plane valleys (4 ) and the two out-of-plane valleys (2 ) by
splitting them in energy, as shown in Fig. 2.1b. The lower energy of the 2 valleys
means that they are preferentially occupied by electrons. The electron mobility partly
improves via a reduced in-plane and increased out-of-plane m* due to the favorable
mass of the 2 valleys, which results in more electrons with an in-plane transverse
effective mass (mt = 0.19 mo ) and out-of-plane longitudinal mass (ml = 0.98 mo ).
For a given strain, quantifying the effective mass reduction and comparing it to
the enhanced mobility reveals that mass reduction alone explains only part of the
mobility enhancement [15]. Hence, electron scattering must also be reduced due to
the conduction valleys splitting into two sets of energy levels, which lowers the rate
of intervalley phonon scattering between the 2 and 4 valleys [16].

2.2.2 Effect of Strain on Hole Mobility

For holes, the valence-band structure of Si is more complex than the conduction-band.
Random documents with unrelated
content Scribd suggests to you:
religion. They had usually been caricatured in fiction and art. Even
Scott had not drawn them as they were. Macaulay, who was always
one of my chief inspirations, had alone made them comprehensible
—the sombre fighters, with their Bibles and their broadswords.
There is a great passage of his—I cannot quote it verbally—in which
he says that after the Restoration if ever you saw a carter more
intelligent than his fellows, or a peasant who tilled his land better,
you would be likely to find that it was an old pikeman of Cromwell’s.
This, then, was my inspiration in “Micah Clarke,” where I fairly let
myself go upon the broad highway of adventure. I was well up in
history, but I spent some months over details and then wrote the
book very rapidly. There are bits of it, the picture of the Puritan
household, and the sketch of Judge Jeffreys, which I have never
bettered. When it was finished early in 1888 my hopes ran high and
out it went on its travels.
But, alas! although my Holmes booklet was out, and had attracted
some little favourable comment, the door still seemed to be barred.
James Payn had first peep, and he began his letter of rejection with
the sentence “How can you, can you, waste your time and your wits
writing historical novels!” This was depressing after a year of work.
Then came Bentley’s verdict: “It lacks in our opinion the one great
necessary point for fiction, i.e. interest; and this being the case we
do not think it could ever become popular with libraries and the
general public.” Then Blackwood had its say: “There are
imperfections which would militate against success. The chances of
the book proving a popular success do not seem to be strong
enough to warrant us in publishing it.” There were others even more
depressing. I was on the point of putting the worn manuscript into
hospital with its mangled brother “Girdlestone” when as a last
resource, I sent it to Longmans, whose reader, Andrew Lang, liked it
and advised its acceptance. It was to “Andrew of the brindled hair,”
as Stevenson called him, that I owe my first real opening, and I
have never forgotten it. The book duly appeared in February, 1889,
and though it was not a boom book it had extraordinarily good
reviews, including one special one all to itself by Mr. Prothero in the
“Nineteenth Century,” and it has sold without intermission from that
day to this. It was the first solid corner-stone laid for some sort of
literary reputation.
British literature had a considerable vogue in the United States at
this time for the simple reason that there was no copyright and they
had not to pay for it. It was hard on British authors, but far harder
on American ones, since they were exposed to this devastating
competition. Like all national sins it brought its own punishment not
only to American authors, who were guiltless, but to the publishers
themselves, for what belongs to everyone belongs practically to no
one, and they could not bring out a decent edition without being at
once undersold. I have seen some of my early American editions
which might have been printed on the paper that shopmen use for
parcels. One good result, however, from my point of view was that a
British author, if he had anything in him, soon won recognition in
America, and afterwards, when the Copyright Act was passed, he
had his audience all ready for him. My Holmes book had met with
some American success and presently I learned that an agent of
Lippincott’s was in London and that he wished to see me, to arrange
for a book. Needless to say that I gave my patients a rest for a day
and eagerly kept the appointment.
Once only before had I touched the edge of literary society. That
was when “Cornhill” was turned into a fully illustrated journal, an
experiment which failed for it was quickly abandoned. The change
was celebrated by a dinner at the Ship, at Greenwich, to which I was
invited on the strength of my short contributions. All the authors and
artists were there, and I remember the reverence with which I
approached James Payn, who was to me the warden of the sacred
gate. I was among the first arrivals, and was greeted by Mr. Smith,
the head of the firm, who introduced me to Payn. I loved much of
his work and waited in awe for the first weighty remark which should
fall from his lips. It was that there was a crack in the window and he
wondered how the devil it had got there. Let me add, however, that
my future experience was to show that there was no wittier or more
delightful companion in the world. I sat next to Anstey that night,
who had just made a most deserved hit with his “Vice Versa,” and I
was introduced to other celebrities, so that I came back walking on
air.
Now for the second time I was in London on literary business.
Stoddart, the American, proved to be an excellent fellow, and had
two others to dinner. They were Gill, a very entertaining Irish M.P.,
and Oscar Wilde, who was already famous as the champion of
æstheticism. It was indeed a golden evening for me. Wilde to my
surprise had read “Micah Clarke” and was enthusiastic about it, so
that I did not feel a complete outsider. His conversation left an
indelible impression upon my mind. He towered above us all, and yet
had the art of seeming to be interested in all that we could say. He
had delicacy of feeling and tact, for the monologue man, however
clever, can never be a gentleman at heart. He took as well as gave,
but what he gave was unique. He had a curious precision of
statement, a delicate flavour of humour, and a trick of small gestures
to illustrate his meaning, which were peculiar to himself. The effect
cannot be reproduced, but I remember how in discussing the wars
of the future he said: “A chemist on each side will approach the
frontier with a bottle”—his upraised hand and precise face conjuring
up a vivid and grotesque picture. His anecdotes, too, were happy
and curious. We were discussing the cynical maxim that the good
fortune of our friends made us discontented. “The devil,” said Wilde,
“was once crossing the Libyan Desert, and he came upon a spot
where a number of small fiends were tormenting a holy hermit. The
sainted man easily shook off their evil suggestions. The devil
watched their failure and then he stepped forward to give them a
lesson. ‘What you do is too crude,’ said he. ‘Permit me for one
moment.’ With that he whispered to the holy man, ‘Your brother has
just been made Bishop of Alexandria.’ A scowl of malignant jealousy
at once clouded the serene face of the hermit. ‘That,’ said the devil
to his imps, ‘is the sort of thing which I should recommend.’”
The result of the evening was that both Wilde and I promised to
write books for “Lippincott’s Magazine”—Wilde’s contribution was
“The Picture of Dorian Grey,” a book which is surely upon a high
moral plane, while I wrote “The Sign of Four,” in which Holmes made
his second appearance. I should add that never in Wilde’s
conversation did I observe one trace of coarseness of thought, nor
could one at that time associate him with such an idea. Only once
again did I see him, many years afterwards, and then he gave me
the impression of being mad. He asked me, I remember, if I had
seen some play of his which was running. I answered that I had not.
He said: “Ah, you must go. It is wonderful. It is genius!” All this with
the gravest face. Nothing could have been more different from his
early gentlemanly instincts. I thought at the time, and still think, that
the monstrous development which ruined him was pathological, and
that a hospital rather than a police court was the place for its
consideration.
When his little book came out I wrote to say what I thought of it. His
letter is worth reproducing, as showing the true Wilde. I omit the
early part in which he comments on my own work in too generous
terms.
“Between me and life there is a mist of words always. I throw
probability out of the window for the sake of a phrase, and the
chance of an epigram makes me desert truth. Still I do aim at
making a work of art, and I am really delighted that you think my
treatment subtle and artistically good. The newspapers seem to me
to be written by the prurient for the Philistine. I cannot understand
how they can treat ‘Dorian Grey’ as immoral. My difficulty was to
keep the inherent moral subordinate to the artistic and dramatic
effect, and it still seems to me that the moral is too obvious.”
Encouraged by the kind reception which “Micah Clarke” had received
from the critics, I now determined upon an even bolder and more
ambitious flight. It seemed to me that the days of Edward III
constituted the greatest epoch in English History—an epoch when
both the French and the Scottish Kings were prisoners in London.
This result had been brought about mainly by the powers of a body
of men who were renowned through Europe but who had never
been drawn in British literature, for though Scott treated in his
inimitable way the English archer, it was as an outlaw rather than as
a soldier that he drew him. I had some views of my own, too, about
the Middle Ages which I was anxious to set forth. I was familiar with
Froissart and Chaucer and I was aware that the famous knights of
old were by no means the athletic heroes of Scott, but were often of
a very different type. Hence came my two books “The White
Company,” written in 1889, and “Sir Nigel,” written fourteen years
later. Of the two I consider the latter the better book, but I have no
hesitation in saying that the two of them taken together did
thoroughly achieve my purpose, that they made an accurate picture
of that great age, and that as a single piece of work they form the
most complete, satisfying and ambitious thing that I have ever done.
All things find their level, but I believe that if I had never touched
Holmes, who has tended to obscure my higher work, my position in
literature would at the present moment be a more commanding one.
The work needed much research and I have still got my notebooks
full of all sorts of lore. I cultivate a simple style and avoid long words
so far as possible, and it may be that this surface of ease has
sometimes caused the reader to underrate the amount of real
research which lies in all my historical novels. It is not a matter
which troubles me, however, for I have always felt that justice is
done in the end, and that the real merit of any work is never
permanently lost.
I remember that as I wrote the last words of “The White Company” I
felt a wave of exultation and with a cry of “That’s done it!” I hurled
my inky pen across the room, where it left a black smudge upon the
duck’s-egg wall-paper. I knew in my heart that the book would live
and that it would illuminate our national traditions. Now that it has
passed through fifty editions I suppose I may say with all modesty
that my forecast has proved to be correct. This was the last book
which I wrote in my days of doctoring at Southsea, and marks an
epoch in my life, so I can now hark back to some other phases of
my last years at Bush Villa before I broke away into a new existence.
I will only add that “The White Company” was accepted by
“Cornhill,” in spite of James Payn’s opinion of historical novels, and
that I fulfilled another ambition by having a serial in that famous
magazine.
A new phase of medical experience came to me about this time, for
I suddenly found myself a unit in the British Army. The operations in
the East had drained the Medical Service, and it had therefore been
determined that local civilian doctors should be enrolled for
temporary duty of some hours a day. The terms were a guinea a
day, and a number of us were tempted to volunteer where there
were only a few vacancies. When I was called before the Board of
Selection a savage-looking old army doctor who presided barked out,
“And you, sir—what are you prepared to do?” To which I answered,
“Anything.” It seems that the others had all been making bargains
and reservations, so my wholehearted reply won the job.
It brought me into closer contact with the savage-looking medico,
who proved to be Sir Anthony Home, V.C.—an honour which he had
won in the Indian Mutiny. He was in supreme charge, and as he was
as fierce in speech and in act as in appearance, everyone was
terrified of him. On one occasion I had told the orderly to draw a
man’s tooth, knowing that he was a very much more skilful dentist
than I. I was on my way home when I was overtaken by an excited
soldier who told me that Sergeant Jones was being court-martialled
and would certainly lose his stripes because he had done a minor
operation. I hurried back and on entering the room found Sir
Anthony glaring at the unhappy man, while several other orderlies
stood round awaiting their own turn. Sir Anthony’s glare was
transferred to me when I said that whatever the Sergeant had done
was by my express order. He grunted, banged the book he was
holding, and broke up the meeting. He seemed a most disagreeable
old man, and yet when I was married shortly afterwards he sent me
a most charming message wishing me good fortune. Up to then I
had never had anything from him save a scowl from his thick
eyebrows, so I was most agreeably surprised. Soon afterwards the
pressure ceased and we civilians were all dismissed.
 CHAPTER IX
PULLING UP THE ANCHOR
Psychic Studies—Experiments in Telepathy—My First Séances—A Curious Test
—General Drayson—Opinion on Theosophy—A. P. Sinnett—W. T. Stead—
Journey to Berlin—Koch’s Treatment—Brutality of Bergmann—Malcolm Morris
—Literary Society—Political Work—Arthur Balfour—Our Departure.

It was in these years after my marriage and before leaving Southsea

that I planted the first seeds of those psychic studies which were
destined to revolutionize my views and to absorb finally all the
energies of my life. I had at that time the usual contempt which the
young educated man feels towards the whole subject which has
been covered by the clumsy name of Spiritualism. I had read of
mediums being convicted of fraud, I had heard of phenomena which
were opposed to every known scientific law, and I had deplored the
simplicity and credulity which could deceive good, earnest people
into believing that such bogus happenings were signs of intelligence
outside our own existence. Educated as I had been during my most
plastic years in the school of medical materialism, and soaked in the
negative views of all my great teachers, I had no room in my brain
for theories which cut right across every fixed conclusion that I had
formed. I was wrong and my great teachers were wrong, but still I
hold that they wrought well and that their Victorian agnosticism was
in the interests of the human race, for it shook the old iron-clad
unreasoning Evangelical position which was so universal before their
days. For all rebuilding a site must be cleared. There were two
separate Victorian movements towards change, the one an attempt
to improve the old building and make it good enough to carry on—as
shown in the Oxford and High Church development, the other a
knocking down of ruins which could only end in some fresh erection
springing up. As I have shown my own position was that of a
respectful materialist who entirely admitted a great central intelligent
cause, without being able to distinguish what that cause was, or why
it should work in so mysterious and terrible a way in bringing its
designs to fulfilment.
From my point of view the mind (and so far as I could see the soul,
which was the total effect of all the hereditary or personal
functionings of the mind) was an emanation from the brain and
entirely physical in its nature. I saw, as a medical man, how a
spicule of bone or a tumour pressing on the brain would cause what
seemed an alteration in the soul. I saw also how drugs or alcohol
would turn on fleeting phases of virtue or vice. The physical
argument seemed an overpowering one. It had never struck me that
the current of events might really flow in the opposite direction, and
that the higher faculties could only manifest themselves imperfectly
through an imperfect instrument. The broken fiddle is silent and yet
the musician is the same as ever.
The first thing which steadied me and made me reconsider my
position was the question of telepathy, which was already being
discussed by William Barrett and others, even before the appearance
of Myers’ monumental work on “Human Personality”—the first book
which devoted to these psychic subjects the deep study and
sustained brain power which they demand. It may, in my opinion,
take a permanent place in human literature like the “Novum
Organum” or “The Descent of Man” or any other great root-book
which has marked a date in human thought. Having read some of
the evidence I began to experiment in thought transference, and I
found a fellow-researcher in Mr. Ball, a well-known architect in the
town. Again and again, sitting behind him, I have drawn diagrams,
and he in turn has made approximately the same figure. I showed
beyond any doubt whatever that I could convey my thought without
words.
But if I could verify such conclusions up to six feet I could not well
doubt them when they gave me the evidence that the same results
could be obtained at a distance. With an appropriate subject, and
some undefined sympathy between the two individuals, it was
independent of space. So the evidence seemed to show. I had
always sworn by science and by the need of fearless following
wherever truth might lie. It was clear now that my position had been
too rigid. I had compared the thought-excretion of the brain to the
bile-excretion of the liver. Clearly this was untenable. If thought
could go a thousand miles and produce a perceptible effect then it
differed entirely not only in degree but in kind from any purely
physical material. That seemed certain, and it must involve some
modification of my old views.
About this time (1886) the family of a General whom I attended
professionally became interested in table turning and asked me to
come and check their results. They sat round a dining-room table
which after a time, their hands being upon it, began to sway and
finally got sufficient motion to tap with one leg. They then asked
questions and received answers, more or less wise and more or less
to the point. They were got by the tedious process of reciting the
alphabet and writing down the letter which the tap indicated. It
seemed to me that we were collectively pushing the table, and that
our wills were concerned in bringing down the leg at the right
moment. I was interested but very sceptical. Some of these
messages were not vague platitudes but were definite and from
dead friends of the family, which naturally impressed them greatly,
though it had not the same effect upon me, since I did not know
them. I have the old records before me as I write. “Don’t tell the
girls when you see them, but they will talk about me. Kiss my baby
for me. I watch her always. Francie.” This was the style of message,
mixed up with a good many platitudes. We held twenty or more of
such meetings, but I never received anything evidential to my own
address, and I was very critical as to the whole proceedings.
None the less there was a problem to be solved and I went on with
its solution, reading the pros and the cons, and asking advice from
those who had experience, especially from General Drayson, a very
distinguished thinker and a pioneer of psychic knowledge, who lived
at that time at Southsea. I had known Drayson first as an
astronomer, for he had worked out a revolutionary idea by which
there is a fatal mistake in our present idea as to the circle which is
described in the heavens by the prolonged axis of the earth. It is
really a wider circle round a different centre, and this correction
enables us to explain several things now inexplicable, and to make
astronomy a more exact science, with certain very important
reactions upon geology and the recurrent glacial epochs, the exact
date of which could be fixed. His views impressed me much at the
time, and several books upholding them have appeared since his
death, notably “Draysoniana” by Admiral de Horsey. If he makes
good, as I think he will, Drayson will make a great permanent name.
His opinion therefore was not negligible upon any subject, and when
he told me his views and experiences on Spiritualism I could not fail
to be impressed, though my own philosophy was far too solid to be
easily destroyed. I was too poor to employ professional mediums,
and to work on such subjects without a medium is as if one worked
at astronomy without a telescope. Once only an old man with some
reputed psychic power came for a small fee and gave us a
demonstration. He went into a loud-breathing trance to the alarm of
his audience, and then gave each of us a test. Mine was certainly a
very remarkable one, for it was “Do not read Leigh Hunt’s book.” I
was hesitating at the time whether I should read his “Comic
Dramatists of the Restoration” or not, for on the one hand it is
literature and on the other the treatment repelled me. This then was
a very final and excellent test so far as telepathy went, but I would
not fully grant that it was more. I was so impressed, however, that I
wrote an account of it to “Light,” the psychic weekly paper, and so in
the year 1887 I actually put myself on public record as a student of
these matters. That was thirty-seven years ago, as I write, so I am a
very senior student now. From that time onwards I read and thought
a great deal, though it was not until the later phase of my life that I
realized whither all this was tending. This question I will treat in a
final section by itself, so that those to whom it is of less interest can
avoid it.
I was deeply interested and attracted for a year or two by
Theosophy, because while Spiritualism seemed at that time to be
chaos so far as philosophy went, Theosophy presented a very well
thought-out and reasonable scheme, parts of which, notably
reincarnation and Karma, seemed to offer an explanation for some
of the anomalies of life. I read Sinnett’s “Occult World” and
afterwards with even greater admiration I read his fine exposition of
Theosophy in “Esoteric Buddhism,” a most notable book. I also met
him, for he was an old friend of General Drayson’s, and I was
impressed by his conversation. Shortly afterwards, however, there
appeared Dr. Hodgson’s report upon his investigation into Madame
Blavatsky’s proceedings at Adyar, which shook my confidence very
much. It is true that Mrs. Besant has since then published a powerful
defence which tends to show that Hodgson may have been
deceived, but the subsequent book “A Priestess of Isis” which
contains many of her own letters leaves an unpleasant impression,
and Sinnett’s posthumous work seems to show that he also had lost
confidence. On the other hand Colonel Olcott shows that the woman
undoubtedly had real psychic powers, whatever their source. As to
Spiritualism it seems to have only interested her in its lower
phenomenal aspect. Her books show extraordinary erudition and
capacity for hard work, even if they represent the transfer of other
people’s conclusions, as they frequently do. It would be unjust,
however, to condemn the old wisdom simply because it was
introduced by this extraordinary and volcanic person. We have also
had in our branch of the occult many dishonest mediums, but we
have hastened to unveil them where we could do so, and Theosophy
will be in a stronger position when it shakes off Madame Blavatsky
altogether. In any case it could never have met my needs for I ask
for severe proof, and if I have to go back to unquestioning faith I
should find myself in the fold from which I wandered.
My life had been a pleasant one with my steadily-increasing literary
success, my practice, which was enough to keep me pleasantly
occupied, and my sport, which I treat in a later chapter. Suddenly,
however, there came a development which shook me out of my rut,
and caused an absolute change in my life and plans. One daughter,
Mary, had been born to us, our household was a happy one, and as
I have never had personal ambitions, since the simple things of life
have always been the most pleasant to me, it is possible that I
should have remained in Southsea permanently but for this new
episode in my life. It arose when in 1890 Koch announced that he
had discovered a sure cure for consumption and that he would
demonstrate it upon a certain date in Berlin.
A great urge came upon me suddenly that I should go to Berlin and
see him do so. I could give no clear reason for this but it was an
irresistible impulse and I at once determined to go. Had I been a
well-known doctor or a specialist in consumption it would have been
more intelligible, but I had, as a matter of fact, no great interest in
the more recent developments of my own profession, and a very
strong belief that much of the so-called progress was illusory.
However, at a few hours’ notice I packed up a bag and started off
alone upon this curious adventure. I had had an interchange of
letters with Mr. W. T. Stead over some matter and I called upon him
at the “Review of Reviews” office as I passed through London to ask
him if he could give me an introduction to Koch or to Dr. Bergmann,
who was to give the demonstration. Mr. Stead was very amiable to
this big unknown provincial doctor, and he gave me a letter for the
British Ambassador—Sir Edward Malet, if I remember right—and for
Mr. Lowe, “The Times” correspondent. He also asked me to do a
character sketch of Koch for him, adding that he would have Count
Mattei as a feature of his magazine this month and Koch the next. I
said, “Then you will have the greatest man of science and the
greatest quack in Europe following each other.” Stead glared at me
angrily, for it seems that the Mattei treatment with its blue electricity
and the rest of it was at that moment his particular fad. However, we
parted amiably and all through his life we kept in distant touch,
though we came into sharp collision at the time of the Boer war. He
was a brave and honest man, and if he was impulsive at times it was
only the sudden outflame of that fire which made him the great
force for good that he was. In psychic knowledge he was a
generation before his time, though his mode of expressing it may
sometimes have been injudicious.
I went on to Berlin that night and found myself in the Continental
express with a very handsome and courteous London physician
bound upon the same errand as myself. We passed most of the night
talking and I learned that his name was Malcolm Morris and that he
also had been a provincial doctor, but that he had come to London
and had made a considerable hit as a skin specialist in Harley Street.
It was the beginning of a friendship which endured.
Having arrived at Berlin the great thing was to be present at
Bergmann’s demonstration, which was to be next day at twelve. I
went to our Ambassador, was kept long waiting, had a chilly
reception and was dismissed without help or consolation. Then I
tried “The Times” correspondent, but he could not help me either.
He and his amiable wife showed me every courtesy and invited me
to dinner that night. Tickets were simply not to be had and neither
money nor interest could procure them. I conceived the wild idea of
getting one from Koch himself and made my way to his house. While
there I had the curious experience of seeing his mail arrive—a large
sack full of letters, which was emptied out on the floor of the hall,
and exhibited every sort of stamp in Europe. It was a sign of all the
sad broken lives and wearied hearts which were turning in hope to
Berlin. Koch remained a veiled prophet, however, and would see
neither me nor any one else. I was fairly at my wit’s ends and could
not imagine how I could attain my end.
Next day I went down to the great building where the address was
to be given and managed by bribing the porter to get into the outer
Hall. The huge audience was assembling in a room beyond. I tried
further bribing that I might be slipped in, but the official became
abusive. People streamed past me, but I was always the waiter at
the gate. Finally every one had gone in and then a group of men
came bustling across, Bergmann, bearded and formidable, in the
van, with a tail of house surgeons and satellites behind him. I threw
myself across his path. “I have come a thousand miles,” said I. “May
I not come in?” He halted and glared at me through his spectacles.
“Perhaps you would like to take my place,” he roared, working
himself up into that strange folly of excitement which seems so
strange in the heavy German nature. “That is the only place left.
Yes, yes, take my place by all means. My classes are filled with
Englishmen already.” He fairly spat out the word “Englishmen” and I
learned afterwards that some recent quarrel with Morel MacKenzie
over the illness of the Emperor Frederick had greatly incensed him. I
am glad to say that I kept my temper and my polite manner, which
is always the best shield when one is met by brutal rudeness. “Not
at all,” I said. “I would not intrude, if there was really no room.” He
glared at me again, all beard and spectacles, and rushed on with his
court all grinning at the snub which the presumptuous Englishman
had received. One of them lingered, however—a kindly American.
“That was bad behaviour,” said he. “See here! If you meet me at
four this afternoon I will show you my full notes of the lecture, and I
know the cases he is about to show, so we can see them together
to-morrow.” Then he followed on.
So it came about that I attained my end after all, but in a
roundabout way. I studied the lecture and the cases, and I had the
temerity to disagree with every one and to come to the conclusion
that the whole thing was experimental and premature. A wave of
madness had seized the world and from all parts, notably from
England, poor afflicted people were rushing to Berlin for a cure,
some of them in such advanced stages of disease that they died in
the train. I felt so sure of my ground and so strongly about it that I
wrote a letter of warning to “The Daily Telegraph,” and I rather think
that this letter was the very first which appeared upon the side of
doubt and caution. I need not say that the event proved the truth of
my forecast.
Two days later I was back in Southsea, but I came back a changed
man. I had spread my wings and had felt something of the powers
within me. Especially I had been influenced by a long talk with
Malcolm Morris, in which he assured me that I was wasting my life in
the provinces and had too small a field for my activities. He insisted
that I should leave general practice and go to London. I answered
that I was by no means sure of my literary success as yet, and that I
could not so easily abandon the medical career which had cost my
mother such sacrifices and myself so many years of study. He asked
me if there was any special branch of the profession on which I
could concentrate so as to get away from general practice. I said
that of late years I had been interested in eye work and had amused
myself by correcting refractions and ordering glasses in the
Portsmouth Eye Hospital under Mr. Vernon Ford. “Well,” said Morris,
“why not specialize upon the eye? Go to Vienna, put in six months’
work, come back and start in London. Thus you will have a nice
clean life with plenty of leisure for your literature.” I came home with
this great suggestion buzzing in my head and as my wife was quite
willing and Mary, my little girl, was old enough now to be left with
her grandmother, there seemed to be no obstacle in the way. There
were no difficulties about disposing of the practice, for it was so
small and so purely personal that it could not be sold to another and
simply had to dissolve.
The Portsmouth Literary and Scientific Society gave me a God-speed
banquet. I have many pleasant and some comic reminiscences of
this Society, of which I had been secretary for several years. We kept
the sacred flame burning in the old city with our weekly papers and
discussions during the long winters. It was there I learned to face an
audience, which proved to be of the first importance for my life’s
work. I was naturally of a very nervous, backward, self-distrustful
disposition in such things and I have been told that the signal that I
was about to join in the discussion was that the whole long bench
on which I sat, with every one on it, used to shake with my emotion.
But once up I learned to speak out, to conceal my trepidations, and
to choose my phrases. I gave three papers, one on the Arctic seas,
one on Carlyle and one on Gibbon. The former gave me a quite
unmerited reputation as a sportsman, for I borrowed from a local
taxidermist every bird and beast that he possessed which could
conceivably find its way into the Arctic Circle. These I piled upon the
lecture table, and the audience, concluding that I had shot them all,
looked upon me with great respect. Next morning they were back
with the taxidermist once more. We had some weird people and
incidents at these debates. I remember one very learned discussion
on fossils and the age of the strata, which was ended by a
cadaverous major-general of the Evangelical persuasion who rose
and said in a hollow voice that all this speculation was vain, and
indeed incomprehensible, since we knew on an authority which
could not possibly be questioned that the world was made exactly
five thousand eight hundred and ninety years ago. This put the lid
on the debate and we all crept home to bed.
My political work also caused me to learn to speak. I was what was
called a Liberal-Unionist, that is, a man whose general position was
Liberal, but who could not see his way to support Gladstone’s Irish
Policy. Perhaps we were wrong. However, that was my view at the
time. I had a dreadful first experience of platform speaking on a
large scale, for at a huge meeting at the Amphitheatre the
candidate, Sir William Crossman, was delayed, and to prevent a
fiasco I was pushed on at a moment’s notice to face an audience of
3,000 people. It was one of the tight corners of my life. I hardly
knew myself what I said, but the Irish part of me came to my aid
and supplied me with a torrent of more or less incoherent words and
similes which roused the audience greatly, though it read to me
afterwards more like a comic stump speech than a serious political
effort. But it was what they wanted and they were mostly on their
feet before I finished. I was amazed when I read it next day, and
especially the last crowning sentence which was: “England and
Ireland are wedded together with the sapphire wedding ring of the
sea, and what God has placed together let no man pluck asunder.” It
was not very good logic but whether it was eloquence or
rodomontade I could not even now determine.
I was acting Secretary when Mr. Balfour came down to address a
great meeting and, as such, when the Hall was full, I waited on the
curb outside to receive him. Presently his carriage drove up and out
he stepped, tall, thin and aristocratic. There were two notorious
artisans of the other side waiting for him and I warned them not to
make trouble. However, the moment Balfour appeared one of them
opened a huge mouth with the intention of emitting a howl of
execration. But it never got out, for I clapped my hand pretty forcibly
over the orifice while I held him by the neck with the other hand. His
companion hit me on the head with a stick, and was promptly
knocked down by one of my companions. Meanwhile Balfour got
safely in, and we two secretaries followed, rather dishevelled after
our adventure. I met Lord Balfour several times in after life but I
never told him how I once had my hat smashed in his defence.
What with the Literary Society and the politicians I left a gap behind
me in Portsmouth and so did my wife, who was universally popular
for her amiable and generous character. It was a wrench to us to
leave so many good friends. However, towards the end of 1890 the
die was cast, and we closed the door of Bush Villa behind us for the
last time. I had days of privation there, and days of growing success
during the eight long years that I had spent in Portsmouth. Now it
was with a sense of wonderful freedom and exhilarating adventure
that we set forth upon the next phase of our lives.
 CHAPTER X
THE GREAT BREAK
Vienna—A Specialist in Wimpole Street—The Great Decision—Norwood—“The
Refugees”—Reported Death of Holmes.

We set forth upon a bitter winter day at the close of 1890 with every
chance of being snowed up on our long trek. We got through all
right, however, and found ourselves in Vienna, arriving on a deadly
cold night, with deep snow under foot and a cutting blizzard in the
air. As we looked from the station the electric lights threw out the
shining silver drift of snow flakes against the absolute darkness of
the sky. It was a gloomy, ominous reception, but half an hour
afterwards when we were in the warm cosy crowded tobacco-laden
restaurant attached to our hotel we took a more cheerful view of our
surroundings.
We found a modest pension which was within our means, and we
put in a very pleasant four months, during which I attended eye
lectures at the Krankenhaus, but could certainly have learned far
more in London, for even if one has a fair knowledge of
conversational German it is very different from following accurately a
rapid lecture filled with technical terms. No doubt “has studied in
Vienna” sounds well in a specialist’s record, but it is usually taken for
granted that he has exhausted his own country before going abroad,
which was by no means the case with me. Therefore, so far as eye
work goes, my winter was wasted, nor can I trace any particular
spiritual or intellectual advance. On the other hand I saw a little of
gay Viennese society. I received kind and welcome hospitality from
Brinsley Richards, “The Times” correspondent, and his wife, and I
had some excellent skating. I also wrote one short book, “The
Doings of Raffles Haw,” not a very notable achievement, by which I
was able to pay my current expenses without encroaching on the
very few hundred pounds which were absolutely all that I had in the
world. This money was invested on the advice of a friend, and as it
was almost all lost—like so much more that I have earned—it is just
as well that I was never driven back upon it.
With the spring my work at Vienna had finished, if it can be said to
have ever begun, and we returned via Paris, putting in a few days
there with Landolt, who was the most famous French oculist of his
time. It was great to find ourselves back in London once more with
the feeling that we were now on the real field of battle, where we
must conquer or perish, for our boats were burned behind us. It is
easy now to look back and think that the issue was clear, but it was
by no means so at the time, for I had earned little, though my
reputation was growing. It was only my own inward conviction of
the permanent merits of “The White Company,” still appearing
month by month in “Cornhill,” which sustained my confidence. I had
come through so much in the early days at Southsea that nothing
could alarm me personally, but I had a wife and child now, and the
stern simplicity of life which was possible and even pleasant in early
days was now no longer to be thought of.
We took rooms in Montague Place, and I went forth to search for
some place where I could put up my plate as an oculist. I was aware
that many of the big men do not find time to work out refractions,
which in some cases of astigmatism take a long time to adjust when
done by retinoscopy. I was capable in this work and liked it, so I
hoped that some of it might drift my way. But to get it, it was clearly
necessary that I should live among the big men so that the patient
could be easily referred to me. I searched the doctors’ quarters and
at last found suitable accommodation at 2 Devonshire Place, which is
at the top of Wimpole Street and close to the classical Harley Street.
There for £120 a year I got the use of a front room with part use of
a waiting room. I was soon to find that they were both waiting
rooms, and now I know that it was better so.
Every morning I walked from the lodgings at Montague Place,
reached my consulting room at ten and sat there until three or four,
with never a ring to disturb my serenity. Could better conditions for
reflection and work be found? It was ideal, and so long as I was
thoroughly unsuccessful in my professional venture there was every
chance of improvement in my literary prospects. Therefore when I
returned to the lodgings at tea-time I bore my little sheaves with
me, the first fruits of a considerable harvest.
A number of monthly magazines were coming out at that time,
notable among which was “The Strand,” then as now under the
editorship of Greenhough Smith. Considering these various journals
with their disconnected stories it had struck me that a single
character running through a series, if it only engaged the attention
of the reader, would bind that reader to that particular magazine. On
the other hand, it had long seemed to me that the ordinary serial
might be an impediment rather than a help to a magazine, since,
sooner or later, one missed one number and afterwards it had lost all
interest. Clearly the ideal compromise was a character which carried
through, and yet instalments which were each complete in
themselves, so that the purchaser was always sure that he could
relish the whole contents of the magazine. I believe that I was the
first to realize this and “The Strand Magazine” the first to put it into
practice.
Looking round for my central character I felt that Sherlock Holmes,
whom I had already handled in two little books, would easily lend
himself to a succession of short stories. These I began in the long
hours of waiting in my consulting room. Greenhough Smith liked
them from the first, and encouraged me to go ahead with them. My
literary affairs had been taken up by that king of agents, A. P. Watt,
who relieved me of all the hateful bargaining, and handled things so
well that any immediate anxiety for money soon disappeared. It was
as well, for not one single patient had ever crossed the threshold of
my room.
I was now once more at a crossroads of my life, and Providence,
which I recognize at every step, made me realize it in a very
energetic and unpleasant way. I was starting off for my usual trudge
one morning from our lodgings when icy shivers passed over me,
and I only got back in time to avoid a total collapse. It was a virulent
attack of influenza, at a time when influenza was in its deadly prime.
Only three years before my dear sister Annette, after spending her
whole life on the family needs, had died of it at Lisbon at the very
moment when my success would have enabled me to recall her from
her long servitude. Now it was my turn, and I very nearly followed
her. I can remember no pain or extreme discomfort, and no psychic
experiences, but for a week I was in great danger, and then found
myself as weak as a child and as emotional, but with a mind as clear
as crystal. It was then, as I surveyed my own life, that I saw how
foolish I was to waste my literary earnings in keeping up an oculist’s
room in Wimpole Street, and I determined with a wild rush of joy to
cut the painter and to trust for ever to my power of writing. I
remember in my delight taking the handkerchief which lay upon the
coverlet in my enfeebled hand, and tossing it up to the ceiling in my
exultation. I should at last be my own master. No longer would I
have to conform to professional dress or try to please any one else. I
would be free to live how I liked and where I liked. It was one of the
great moments of exultation of my life. The date was in August,
1891.
Presently I was about, hobbling on a stick and reflecting that if I
lived to be eighty I knew already exactly how it would feel. I
haunted house-agents, got lists of suburban villas, and spent some
weeks, as my strength returned, in searching for a new home.
Finally I found a suitable house, modest but comfortable, isolated
and yet one of a row. It was 12 Tennison Road, South Norwood.
There we settled down, and there I made my first effort to live
entirely by my pen. It soon became evident that I had been playing
the game well within my powers and that I should have no difficulty
in providing a sufficient income. It seemed as if I had settled into a
life which might be continuous, and I little foresaw that an
unexpected blow was about to fall upon us, and that we were not at
the end, but really at the beginning, of our wanderings.
I could not know this, however, and I settled down with a stout
heart to do some literary work worthy of the name. The difficulty of
the Holmes work was that every story really needed as clear-cut and
original a plot as a longish book would do. One cannot without effort
spin plots at such a rate. They are apt to become thin or to break. I
was determined, now that I had no longer the excuse of absolute
pecuniary pressure, never again to write anything which was not as
good as I could possibly make it, and therefore I would not write a
Holmes story without a worthy plot and without a problem which
interested my own mind, for that is the first requisite before you can
interest any one else. If I have been able to sustain this character
for a long time and if the public find, as they will find, that the last
story is as good as the first, it is entirely due to the fact that I never,
or hardly ever, forced a story. Some have thought there was a falling
off in the stories, and the criticism was neatly expressed by a
Cornish boatman who said to me, “I think, sir, when Holmes fell over
that cliff, he may not have killed himself, but all the same he was
never quite the same man afterwards.” I think, however, that if the
reader began the series backwards, so that he brought a fresh mind
to the last stories, he would agree with me that, though the general
average may not be conspicuously high, still the last one is as good
as the first.
I was weary, however, of inventing plots and I set myself now to do
some work which would certainly be less remunerative but would be
more ambitious from a literary point of view. I had long been
attracted by the epoch of Louis XIV and by those Huguenots who
were the French equivalents of our Puritans. I had a good
knowledge of the memoirs of that date, and many notes already
prepared, so that it did not take me long to write “The Refugees.” It
has stood the acid test of time very well, so I may say that it was a
success. Soon after its appearance it was translated into French, and
my mother, herself a great French scholar, had the joy when she
visited Fontainebleau to hear the official guide tell the drove of
tourists that if they really wanted to know about the Court of the
great monarch, they would find the clearest and most accurate
account in an Englishman’s book, “The Refugees.” I expect the guide
would have been considerably astonished had he then and there
been kissed by an elderly English lady, but it was an experience
which he must have narrowly missed. I used in this book, also, a
great deal which was drawn from Parkman, that great but neglected
historian, who was in my opinion the greatest serious writer that
America has produced.
There was an amusing episode connected with “The Refugees,”
when it was read aloud in some strict Irish convent, the innocent
Reverend Mother having mistaken my name and imagined that I was
a canon, and therefore of course a holy man. I am told that the
reading was a tremendous success and that the good sisters rejoiced
that the mistake was not found out until the story was completed.
My first name has several times led to mistakes, as when, at a big
dinner at Chicago, I was asked to say Grace, as being the only
ecclesiastic present. I remember that at the same dinner one of the
speakers remarked that it was a most sinister fact that though I was
a doctor no living patient of mine had ever yet been seen.
During this Norwood interval, I was certainly working hard, for
besides “The Refugees” I wrote “The Great Shadow,” a booklet
which I should put near the front of my work for merit, and two
other little books on a very inferior plane—“The Parasite” and
“Beyond the City.” The latter was of a domestic type unusual for me.
It was pirated in New York just before the new Copyright Act came
into force, and the rascal publisher thinking that a portrait—any sort
of portrait—of the author would look well upon the cover, and being
quite ignorant of my identity, put a very pretty and over-dressed
young woman as my presentment. I still preserve a copy of this
most flattering representation. All these books had some decent
success, though none of it was remarkable. It was still the Sherlock
Holmes stories for which the public clamoured, and these from time
to time I endeavoured to supply. At last, after I had done two series
of them I saw that I was in danger of having my hand forced, and of
being entirely identified with what I regarded as a lower stratum of
literary achievement. Therefore as a sign of my resolution I
determined to end the life of my hero. The idea was in my mind
when I went with my wife for a short holiday in Switzerland, in the
course of which we saw there the wonderful falls of Reichenbach, a
terrible place, and one that I thought would make a worthy tomb for
poor Sherlock, even if I buried my banking account along with him.
So there I laid him, fully determined that he should stay there—as
indeed for some years he did. I was amazed at the concern
expressed by the public. They say that a man is never properly
appreciated until he is dead, and the general protest against my
summary execution of Holmes taught me how many and how
numerous were his friends. “You Brute” was the beginning of the
letter of remonstrance which one lady sent me, and I expect she
spoke for others besides herself. I heard of many who wept. I fear I
was utterly callous myself, and only glad to have a chance of
opening out into new fields of imagination, for the temptation of
high prices made it difficult to get one’s thoughts away from Holmes.
That Sherlock Holmes was anything but mythical to many is shown
by the fact that I have had many letters addressed to him with
requests that I forward them. Watson has also had a number of
letters in which he has been asked for the address or for the
autograph of his more brilliant confrère. A press-cutting agency
wrote to Watson asking whether Holmes would not wish to
subscribe. When Holmes retired several elderly ladies were ready to
keep house for him and one sought to ingratiate herself by assuring
me that she knew all about bee-keeping and could “segregate the
queen.” I had considerable offers also for Holmes if he would
examine and solve various family mysteries. Once the offer—from
Poland—was that I should myself go, and my reward was practically
left to my own judgment. I had judgment enough, however, to avoid
it altogether.
I have often been asked whether I had myself the qualities which I
depicted, or whether I was merely the Watson that I look. Of course
I am well aware that it is one thing to grapple with a practical
problem and quite another thing when you are allowed to solve it
under your own conditions. I have no delusions about that. At the
same time a man cannot spin a character out of his own inner
consciousness and make it really life-like unless he has some
possibilities of that character within him—which is a dangerous
admission for one who has drawn so many villains as I. In my poem
“The Inner Room,” describing our multiplex personality, I say:

“There are others who are sitting,

Grim as doom,
In the dim ill-boding shadow
Of my room.
Darkling figures, stern or quaint,
Now a savage, now a saint,
Showing fitfully and faint
In the gloom.”

Among those figures there may perhaps be an astute detective also,

but I find that in real life in order to find him I have to inhibit all the
others and get into a mood when there is no one in the room but
he. Then I get results and have several times solved problems by
Holmes’ methods after the police have been baffled. Yet I must
admit that in ordinary life I am by no means observant and that I
have to throw myself into an artificial frame of mind before I can
weigh evidence and anticipate the sequence of events.
 CHAPTER XI
SIDELIGHTS ON SHERLOCK HOLMES
“The Speckled Band”—Barrie’s Parody on Holmes—Holmes on the Films—
Methods of Construction—Problems—Curious Letters—Some Personal Cases—
Strange Happenings.

I may as well interrupt my narrative here in order to say what may

interest my readers about my most notorious character.
The impression that Holmes was a real person of flesh and blood
may have been intensified by his frequent appearance upon the
stage. After the withdrawal of my dramatization of “Rodney Stone”
from a theatre upon which I held a six months’ lease, I determined
to play a bold and energetic game, for an empty theatre spells ruin.
When I saw the course that things were taking I shut myself up and
devoted my whole mind to making a sensational Sherlock Holmes
drama. I wrote it in a week and called it “The Speckled Band” after
the short story of that name. I do not think that I exaggerate if I say
that within a fortnight of the one play shutting down I had a
company working upon the rehearsals of a second one, which had
been written in the interval. It was a considerable success. Lyn
Harding, as the half epileptic and wholly formidable Doctor Grimesby
Rylott, was most masterful, while Saintsbury as Sherlock Holmes was
also very good. Before the end of the run I had cleared off all that I
had lost upon the other play, and I had created a permanent
property of some value. It became a stock piece and is even now
touring the country. We had a fine rock boa to play the title-rôle, a
snake which was the pride of my heart, so one can imagine my
disgust when I saw that one critic ended his disparaging review by
the words “The crisis of the play was produced by the appearance of
a palpably artificial serpent.” I was inclined to offer him a goodly
sum if he would undertake to go to bed with it. We had several
snakes at different times, but they were none of them born actors
and they were all inclined either to hang down from the hole in the
wall like inanimate bell-pulls, or else to turn back through the hole
and get even with the stage carpenter who pinched their tails in
order to make them more lively. Finally we used artificial snakes, and
every one, including the stage carpenter, agreed that it was more
satisfactory.
This was the second Sherlock Holmes play. I should have spoken
about the first, which was produced very much earlier, in fact at the
time of the African war. It was written and most wonderfully acted
by William Gillette, the famous American. Since he used my
characters and to some extent my plots, he naturally gave me a
share in the undertaking, which proved to be very successful. “May I
marry Holmes?” was one cable which I received from him when in
the throes of composition. “You may marry or murder or do what
you like with him,” was my heartless reply. I was charmed both with
the play, the acting and the pecuniary result. I think that every man
with a drop of artistic blood in his veins would agree that the latter
consideration, though very welcome when it does arrive, is still the
last of which he thinks.
Sir James Barrie paid his respects to Sherlock Holmes in a rollicking
parody. It was really a gay gesture of resignation over the failure
which we had encountered with a comic opera for which he
undertook to write the libretto. I collaborated with him on this, but in
spite of our joint efforts, the piece fell flat. Whereupon Barrie sent
me a parody on Holmes, written on the fly leaves of one of his
books. It ran thus:—

The Adventure of the Two Collaborators

In bringing to a close the adventures of my friend Sherlock Holmes I

am perforce reminded that he never, save on the occasion which, as
you will now hear, brought his singular career to an end, consented
to act in any mystery which was concerned with persons who made
a livelihood by their pen. “I am not particular about the people I mix
among for business purposes,” he would say, “but at literary
characters I draw the line.”
We were in our rooms in Baker Street one evening. I was (I
remember) by the centre table writing out “The Adventure of the
Man without a Cork Leg” (which had so puzzled the Royal Society
and all the other scientific bodies of Europe), and Holmes was
amusing himself with a little revolver practice. It was his custom of a
summer evening to fire round my head, just shaving my face, until
he had made a photograph of me on the opposite wall, and it is a
slight proof of his skill that many of these portraits in pistol shots are
considered admirable likenesses.
I happened to look out of the window, and perceiving two
gentlemen advancing rapidly along Baker Street asked him who they
were. He immediately lit his pipe, and, twisting himself on a chair
into the figure 8, replied:
“They are two collaborators in comic opera, and their play has not
been a triumph.”
I sprang from my chair to the ceiling in amazement, and he then
explained:
“My dear Watson, they are obviously men who follow some low
calling. That much even you should be able to read in their faces.
Those little pieces of blue paper which they fling angrily from them
are Durrant’s Press Notices. Of these they have obviously hundreds
about their person (see how their pockets bulge). They would not
dance on them if they were pleasant reading.”
I again sprang to the ceiling (which is much dented), and shouted:
“Amazing! but they may be mere authors.”
“No,” said Holmes, “for mere authors only get one press notice a
week. Only criminals, dramatists and actors get them by the
hundred.”
“Then they may be actors.”
“No, actors would come in a carriage.”
“Can you tell me anything else about them?”
“A great deal. From the mud on the boots of the tall one I perceive
that he comes from South Norwood. The other is as obviously a
Scotch author.”
“How can you tell that?”
“He is carrying in his pocket a book called (I clearly see) ‘Auld Licht
Something.’ Would any one but the author be likely to carry about a
book with such a title?”
I had to confess that this was improbable.
It was now evident that the two men (if such they can be called)
were seeking our lodgings. I have said (often) that my friend Holmes
seldom gave way to emotion of any kind, but he now turned livid
with passion. Presently this gave place to a strange look of triumph.
“Watson,” he said, “that big fellow has for years taken the credit for
my most remarkable doings, but at last I have him—at last!”
Up I went to the ceiling, and when I returned the strangers were in
the room.
“I perceive, gentlemen,” said Mr. Sherlock Holmes, “that you are at
present afflicted by an extraordinary novelty.”
The handsomer of our visitors asked in amazement how he knew
this, but the big one only scowled.
“You forget that you wear a ring on your fourth finger,” replied Mr.
Holmes calmly.
I was about to jump to the ceiling when the big brute interposed.
“That Tommy-rot is all very well for the public, Holmes,” said he,
“but you can drop it before me. And, Watson, if you go up to the
ceiling again I shall make you stay there.”
Here I observed a curious phenomenon. My friend Sherlock Holmes
shrank. He became small before my eyes. I looked longingly at the
ceiling, but dared not.
“Let us cut the first four pages,” said the big man, “and proceed to
business. I want to know why——”
“Allow me,” said Mr. Holmes, with some of his old courage. “You
want to know why the public does not go to your opera.”
“Exactly,” said the other ironically, “as you perceive by my shirt stud.”
He added more gravely, “And as you can only find out in one way I
must insist on your witnessing an entire performance of the piece.”
It was an anxious moment for me. I shuddered, for I knew that if
Holmes went I should have to go with him. But my friend had a
heart of gold. “Never,” he cried fiercely, “I will do anything for you
save that.”
“Your continued existence depends on it,” said the big man
menacingly.
“I would rather melt into air,” replied Holmes, proudly taking another
chair. “But I can tell you why the public don’t go to your piece
without sitting the thing out myself.”
“Why?”
“Because,” replied Holmes calmly, “they prefer to stay away.”
A dead silence followed that extraordinary remark. For a moment the
two intruders gazed with awe upon the man who had unravelled
their mystery so wonderfully. Then drawing their knives——
Holmes grew less and less, until nothing was left save a ring of
smoke which slowly circled to the ceiling.
The last words of great men are often noteworthy. These were the
last words of Sherlock Holmes: “Fool, fool! I have kept you in luxury
for years. By my help you have ridden extensively in cabs, where no
author was ever seen before. Henceforth you will ride in buses!”
The brute sunk into a chair aghast.
The other author did not turn a hair.