Silicon Photonics An Introduction 1st Edition

Graham T. Reed download

https://ebookname.com/product/silicon-photonics-an-
introduction-1st-edition-graham-t-reed/

Get Instant Ebook Downloads – Browse at https://ebookname.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

Nanostructured Silicon for Photonics 1st Edition Z.

Gaburro

https://ebookname.com/product/nanostructured-silicon-for-
photonics-1st-edition-z-gaburro/

Photonics and lasers An introduction Richard S. Quimby

https://ebookname.com/product/photonics-and-lasers-an-
introduction-richard-s-quimby/

Silicon Photonics Design From Devices to Systems 1st

Edition Lukas Chrostowski

https://ebookname.com/product/silicon-photonics-design-from-
devices-to-systems-1st-edition-lukas-chrostowski/

Urban Sustainability in Theory and Practice Circles of

sustainability 1st Edition Paul James

https://ebookname.com/product/urban-sustainability-in-theory-and-
practice-circles-of-sustainability-1st-edition-paul-james/
Global Nuclear Energy Partnership 1st Edition Alan N.
Bernstein

https://ebookname.com/product/global-nuclear-energy-
partnership-1st-edition-alan-n-bernstein/

Making Families Through Adoption Contemporary Family

Perspectives CFP 1st Edition Nancy E. Riley

https://ebookname.com/product/making-families-through-adoption-
contemporary-family-perspectives-cfp-1st-edition-nancy-e-riley/

Japan s Postwar Economic Recovery and Anglo Japanese

Relations 1948 62 Noriko Yokoi

https://ebookname.com/product/japan-s-postwar-economic-recovery-
and-anglo-japanese-relations-1948-62-noriko-yokoi/

Systemic Homeostasis and Poikilostasis in Sleep Is REM

Sleep a Physiological Paradox 1st Edition Pier Luigi
Parmeggiani

https://ebookname.com/product/systemic-homeostasis-and-
poikilostasis-in-sleep-is-rem-sleep-a-physiological-paradox-1st-
edition-pier-luigi-parmeggiani/

Applied equine nutrition and training Equine NUtrition

and TRAining COnference ENUTRACO 2015 1st Edition
Lindner

https://ebookname.com/product/applied-equine-nutrition-and-
training-equine-nutrition-and-training-conference-
enutraco-2015-1st-edition-lindner/
Lung Cancer Vol 1 Molecular Pathology Methods and
Reviews Methods in Molecular Medicine 1st Edition
Barbara Driscoll

https://ebookname.com/product/lung-cancer-vol-1-molecular-
pathology-methods-and-reviews-methods-in-molecular-medicine-1st-
edition-barbara-driscoll/
Silicon Photonics
 Silicon Photonics
An Introduction

Graham T. Reed
Advanced Technology Institute,
University of Surrey, Guildford, UK

Andrew P. Knights
McMaster University, Hamilton, Ontario, Canada
Copyright  2004 John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester,
West Sussex PO19 8SQ, England

Telephone (+44) 1243 779777

Email (for orders and customer service enquiries): [email protected]

Visit our Home Page on www.wileyeurope.com or www.wiley.com

All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval
system or transmitted in any form or by any means, electronic, mechanical, photocopying,
recording, scanning or otherwise, except under the terms of the Copyright, Designs and
Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency
Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of
the Publisher. Requests to the Publisher should be addressed to the Permissions Department,
John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ,
England, or emailed to [email protected], or faxed to (+44) 1243 770620.

This publication is designed to provide accurate and authoritative information in regard to

the subject matter covered. It is sold on the understanding that the Publisher is not engaged in
rendering professional services. If professional advice or other expert assistance is required,
the services of a competent professional should be sought.

Other Wiley Editorial Offices

John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA

Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA

Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany

John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia

John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore
129809

John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W 1L1

Wiley also publishes its books in a variety of electronic formats. Some content that appears
in print may not be available in electronic books.

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library

ISBN 0-470-87034-6

Typeset in 10.5/13pt Sabon by Laserwords Private Limited, Chennai, India

Printed and bound in Great Britain by TJ International, Padstow, Cornwall
This book is printed on acid-free paper responsibly manufactured from sustainable forestry
in which at least two trees are planted for each one used for paper production.
GTR dedicates this book to the following people:

To Alison, Hannah and Matthew for love and inspiration;

To my parents Colleen and John for a lifetime of support;
To Jackie for sibling rivalry!

APK dedicates this book to Melanie

Contents

About the Authors xiii

Foreword xv

Acknowledgements xix

1 Fundamentals 1
1.1 What is Phase? 1
1.2 What is Polarisation? 4
1.3 What is Interference? 6

2 The Basics of Guided Waves 11

2.1 The Ray Optics Approach to Describing Planar
Waveguides 11
2.2 Reflection Coefficients 12
2.3 Phase of a Propagating Wave and its Wavevector 15
2.4 Modes of a Planar Waveguide 16
2.4.1 The Symmetrical Planar Waveguide 18
2.4.2 The Asymmetrical Planar Waveguide 20
2.4.3 Solving the Eigenvalue Equations for
Symmetrical and Asymmetrical Waveguides 21
2.4.4 Monomode Conditions 22
2.4.5 Effective Index of a Mode 24
2.5 A Taste of Electromagnetic Theory 25
2.6 Simplifying and Solving the Wave Equation 27
2.7 Another Look at Propagation Constants 32
viii CONTENTS

2.8 Mode Profiles 33

2.9 Confinement Factor 34
2.10 The Goos–Hänchen Shift 35

3 Characteristics of Optical Fibres for Communications 37

3.1 The Structure of Optical Fibres 37
3.2 Modes of an Optical Fibre 39
3.2.1 Modes of a Step-index Fibre 40
3.2.2 Modes of a Graded-index Fibre 42
3.3 Numerical Aperture and Acceptance Angle 43
3.4 Dispersion in Optical Fibres 46
3.4.1 Intermodal Dispersion 46
3.4.2 Intramodal Dispersion 49
3.5 Single-mode Fibres: Mode Profile, Mode-field Diameter,
and Spot Size 52
3.6 Normalised Frequency, Normalised Propagation
Constant, and Cutoff Wavelength 54
References 56

4 Silicon-on-Insulator (SOI) Photonics 57

4.1 Introduction 57
4.2 Silicon-on-Insulator Waveguides 58
4.2.1 Modes of Two-dimensional Waveguides 60
4.3 The Effective Index Method of Analysis 60
4.4 Large Single-mode Rib Waveguides 64
4.5 Refractive Index and Loss Coefficient in Optical
Waveguides 69
4.6 Contributions to Loss in an Optical Waveguide 70
4.6.1 Scattering 70
4.6.2 Absorption 73
4.6.3 Radiation 74
4.7 Coupling to the Optical Circuit 76
4.7.1 Grating Couplers 78
4.7.2 Butt Coupling and End-fire Coupling 81
4.7.3 Robust Coupling to Waveguides for
Commercial Applications 87
4.7.4 Measurement of Propagation Loss in Integrated
Optical Waveguides 91
4.8 Optical Modulation Mechanisms in Silicon 97
4.8.1 Electric Field Effects 98
CONTENTS ix

4.8.2 Carrier Injection or Depletion 101

4.8.3 The Thermo-optic Effect 103
4.9 Other Advantages and Disadvantages of Silicon
Photonics 103
References 108

5 Fabrication of Silicon Waveguide Devices 111

5.1 Silicon-on-Insulator (SOI) 111
5.1.1 Separation by IMplanted OXygen (SIMOX) 112
5.1.2 Bond and Etch-back SOI (BESOI) 114
5.1.3 Wafer Splitting (SmartCut Process to Produce
Unibond Wafers) 116
5.1.4 Silicon Epitaxial Growth 117
5.1.5 Deciding on the SOI 118
5.2 Fabrication of Surface Etched Features 119
5.2.1 Photolithography 119
5.2.2 Silicon Etching 123
5.2.3 Critical Dimension Control 125
5.3 Oxidation 127
5.4 Formation of Submicron Silicon Waveguides 129
5.4.1 Silicon Dioxide Thickness 129
5.4.2 Surface and Interface Roughness 130
5.4.3 Sidewall Roughness 131
5.5 Silicon Doping 132
5.5.1 Ion Implantation 133
5.5.2 The Implantation System 134
5.5.3 Implantation Parameters 136
5.5.4 Dopant Activation and Drive-in 137
5.6 Metallisation 138
5.6.1 Via Formation 138
5.6.2 Metal Deposition 138
5.6.3 Material Choice 140
5.6.4 Sintering and Barrier Materials 140
5.7 Summary 141
References 141

6 A Selection of Photonic Devices 145

6.1 Optical Phase Modulators and Variable Optical
Attenuators 145
6.1.1 The Optical Phase Modulator 146
6.1.2 Modelling of Semiconductor Devices 146
x CONTENTS

6.1.3 Basic Device Geometry, and the Aim of

Modelling 147
6.1.4 Effect of Parametric Variation on the DC
Efficiency of an Optical Modulator 150
6.1.5 Predicted Device Operation as a Phase
Modulator 157
6.1.6 Fabrication and Experimental Results 159
6.1.7 Influence of the Thermo-optic Effect on
Experimental Devices 162
6.1.8 Switching Characteristics of the Optical Phase
Modulator 163
6.2 The Mach–Zehnder Interferometer 165
6.3 The Waveguide Bend 167
6.4 The Waveguide-to-Waveguide Coupler 172
6.4.1 Applications of the Waveguide-to-Waveguide
Coupler 174
6.5 The Arrayed Waveguide Grating (AWG) 176
6.5.1 Interference of N Coherent Light Sources 176
6.5.2 Operation of the AWG 180
6.6 Waveguide Couplers for Small-dimension Waveguides 186
References 189

7 Polarisation-dependent Losses: Issues for Consideration 191

7.1 The Effect of Waveguide Thickness 191
7.2 Surface Scattering Loss for Different Waveguide
Thickness and Polarisation 197
7.3 Polarisation-dependent Coupling Loss 201
7.4 Birefringence 204
7.4.1 Birefringence in Planar Silicon Waveguides 205
7.4.2 Birefringence in Silicon Rib Waveguides 207
7.5 The Effect of Stress 213
7.6 Discussion 214
7.6.1 The Effect of Polarisation and Multimode
Sections on the AWG 215
7.6.2 The Effect of PDL on Other Devices 219
7.7 Conclusion 219
References 220

8 Prospects for Silicon Light-emitting Devices 223

8.1 Erbium Doping 223
8.1.1 Erbium Ion Implantation 224
CONTENTS xi

8.1.2 Optical Efficiency of Er-implanted Si 226

8.1.3 Optical Intensity Quenching 228
8.1.4 Electroluminescent (EL) Devices 229
8.2 Low-dimensional Structures 232
8.2.1 Porous Silicon 232
8.2.2 Nano-crystals 235
8.2.3 Nano-crystals with Erbium 239
8.3 Dislocation-engineered Emitters 242
8.4 Raman Excitation 244
8.4.1 Spontaneous Raman Effect 245
8.4.2 Stimulated Raman Effect 246
8.4.3 Raman Emission from Silicon Waveguides
at 1.54 µm 246
8.5 Summary 248
References 248

Index 251
About the Authors

Graham T Reed BSc (Hons), PhD, FIEE, CEng

Silicon Photonics Research Group, Advanced Technology Institute,
University of Surrey, UK
Graham Reed is Professor of Optoelectronics at the University of Surrey
in the UK. He graduated in 1983 with a First Class Honours degree in
Electronic and Electrical Engineering. Subsequently he obtained a PhD
in Integrated Optics in 1987. After a brief period as leader of the Electro-
Optics Systems Group at ERA Technology Ltd, he joined the University
of Surrey in 1989, where he established the Silicon Photonics Group.
As such this was one of the pioneering groups in silicon photonics,
and has made a significant impact on the state of the art. The group
is currently the leading group in the UK in this field, and Professor
Reed is acknowledged as the individual who initiated research on silicon
photonic circuits and devices in the UK. The aim of the silicon work has
been to develop a technology that would have a variety of applications,
although telecommunications remains the dominant application area.
The work has been carried out with collaborators from all around the
world, both from academic and industrial institutions. Professor Reed
has published extensively in the international scientific literature, has
contributed presentations to numerous international conferences both
as a submitting and an invited speaker, and has served on a variety of
international committees.
xiv ABOUT THE AUTHORS

Andrew P Knights BSc (Hons), Ph.D

Department of Engineering Physics, McMaster University, Hamilton,
Canada
Andy Knights received his Ph.D. in 1994 in the area of surface and sub-
surface material characterisation with beams of low energy positrons
and electrons. His subsequent research took him first to the University
of Western Ontario where he performed ground-breaking work on the
generation and evolution of implant induced defects in silicon, and then
to the University of Surrey as part of the EPSRC Ion Beam Centre,
researching novel fabrication processes for micro and opto-electronic
materials. In 2000 he joined Bookham Technology and worked on a
range of silicon-based, highly integrated photonic devices. He played a
critical role in the development of the solid-state electronic optical vari-
able attenuator (VOA); the multiplexer + VOA and the monolithically
integrated optical detector. In 2003 he moved to McMaster University in
Hamilton, Canada, where he holds a faculty position in the Department
of Engineering Physics. He currently leads a research group working
on the interaction of optical and electrical functionality in silicon-based
structures. Dr. Knights has published extensively on semiconductor
device design, fabrication and characterization and his work has been
presented at international conferences on numerous occasions.
Foreword

‘Siliconizing Photonics’

The observation by Gordon Moore in 1965 (now universally referred

to as Moore’s law) that the number of transistors on an integrated
circuit would double every couple of years has become a beacon that
continues to drive the electronics industry. Integrated circuits have
grown exponentially from the 30-transistor devices of 1965 to today’s
high-end microprocessors exceeding 500 million transistors integrated
on a silicon chip the size of your fingernail. Moore’s law will continue,
with over one billion transistors per chip expected by 2010. Decades
of research and manufacturing investment to drive Moore’s law has
resulted in significant performance gains while simultaneously bringing
about significant cost reductions. As an example, in 1968 the cost
of a transistor was around one dollar. By 1995, one dollar bought
about 3000 transistors. Today, one dollar purchases about five million
transistors.
The Internet explosion has changed how we go about our everyday
lives. The thirst for information and the need to ‘always be connected’
is spawning a new era of communications. This new era will continue
to spur the need for higher bandwidth technologies to keep pace with
processor performance. Because of Moore’s law, computing today is
limited less by the computer’s performance than by the rate at which
data can travel between the processor and the outside world. Fiber-optic
solutions are replacing copper-based solutions, which can no longer
meet the bandwidth and distance requirements needed for worldwide
data communications.
xvi FOREWORD

Over the last decade, optical communication technologies have

migrated steadily from long-haul backbones to the network edge,
invading metropolitan area networks (MANs) and campus-level local
area networks (LANs). A key inflection point will come with the ability
to economically connect central offices to diverse network access points.
One of the most important consequences of this migration has been
the need to develop more efficient and lower cost optical solutions.
The future of optical networking rests on the ability to bring optical
technology from the MAN or LAN, into the data center, to the curb, to
the home and, if possible, maybe some day directly to the microprocessor.
Today, optical devices are large, bulky and mostly not fabricated
from silicon. Most optical components are made from III–V-based
compounds such as indium phosphide (InP), gallium arsenide (GaAs) or
the electro-optic crystal lithium niobate (LiNbO3 ). These optical devices
are often custom-made and assembled from discrete components. They
typically are assembled by hand with very little automation. We are
just now beginning to see some standardization occurring in the optical
arena. The net result of all this is that these optical devices are relatively
expensive. Optical technology to the mass market may happen only
if one can bring data-com economics, high-volume manufacturing and
assembly to the optical world.
This raises several questions. Can we ‘siliconize’ photonics? Can we call
on the decades of research and manufacturing experience gained from the
microelectronics industry and apply it to photonics? Could silicon be used
as an alternative to more exotic materials (such as InP or LiNbO3 ) typi-
cally used to produce optical devices? Could one monolithically integrate
multiple photonic circuits on a single silicon chip to increase performance
while simultaneously reducing cost? Could one implement standardiza-
tion and high-volume manufacturing techniques to reduce cost? Could
we combine electronics with photonics to bring new levels of integration,
and possibly a derivative form of Moore’s law to photonics?
These are very good questions. Although it is well known that silicon
is the optimal material for electronics, only recently has silicon been
considered as a practical option for optics. Silicon in fact has many
properties conducive to fiber optics. The band gap of silicon (∼1.1 eV) is
such that the material is transparent to wavelengths commonly used for
optical transport (around 1.3–1.6 µm). One can use standard CMOS-
processing techniques to sculpt optical waveguides onto the silicon
surface. Similar to an optical fiber, these optical waveguides can be used
to confine and direct light as it passes through the silicon.
FOREWORD xvii

Due to the wavelengths typically used for optical transport and

silicon’s high index of refraction, the feature sizes needed for processing
these silicon waveguides are on the order of 0.5–1 µm. The lithography
requirements needed to process waveguides with these sizes exist today.
If we push forward to leading-edge research currently under way in the
area of photonic band-gap devices (PBGs), today’s state-of-the-art 90-nm
fabrication facilities should meet the technical requirements needed for
processing PBGs. What this says is that we may already have all or most
of the processing technologies needed to produce silicon-based photonic
devices for the next decade.
In addition, the same carriers used for the basic functionality of
the transistor (i.e. electrons and holes) can be used to modulate the
phase of light passing through silicon waveguides and thus produce
‘active’ rather than passive photonic devices. Finally, if all this remains
CMOS-compatible, it could be possible to process transistors alongside
photonic devices, the combination of which could bring new levels of
performance, functionality, power and size reduction, all at a lower cost.
So the answer is, yes, it is possible to ‘siliconize’ photonics. Will it
happen? The Internet growth engine is alive and well and the optical
communications industry will need to move from custom low-volume
technologies to high-volume, standardized building blocks. This may
happen only if silicon is the material on which we build this photonic
technology in the future.
So what of this book? I believe this to be the first comprehen-
sive book on silicon photonics. I found it to be easy to read and
believe it will be useful to a wide range of readers. For those study-
ing at a university, at either an undergraduate or graduate level,
this book will provide the foundation you need to enter the field.
For the corporate researcher working in this new and exciting field,
this book will provide an insight into the practical issues and chal-
lenges involved with fabricating and producing silicon-based pho-
tonic devices.
The authors have put together a comprehensive repository of informa-
tion, starting from the theoretical fundamentals to outlining the technical
and practical issues in producing optical devices in silicon. The book
starts with a solid theoretical analysis on the properties of guided waves,
optical modes and optical dispersion. The text takes this theory and
describes how to translate it into designing waveguides in silicon using a
basic rib structure and the parameters required to make these waveguides
single-mode. The reader is walked through the processing steps that are
needed to produce optical waveguides and photonic devices in silicon,
xviii FOREWORD

and is exposed to some of the processing tolerances and techniques that

effect optical device performance.
A very thorough and detailed analysis for producing an active device
(specifically a phase modulator based on current injection) is presented
and this gives the reader a good idea of the various parameters that can
be varied to improve device performance.
In addition, other devices such as evanescent couplers, AWGs and
structures designed for coupling light from small waveguides into an
optical fiber are discussed. Finally, there is a technical review of the
relatively hotly debated and heavily researched topic ongoing in the
world, namely, that of silicon-based light emitters.
This book complements and builds upon over a decade of research that
Professor Graham Reed has spent in silicon photonics, and two decades
in the wider field of opto-electronics. Professor Reed has been a pioneer in
the field of silicon photonics from his initial research on the fundamental
properties of waveguiding in silicon to his internationally recognized
leading work on devices such as optical modulators, grating couplers and
even silicon sensing applications. In addition, Dr Andy Knights brings
to the book hands-on fabrication experience in an industrial setting.
This experience has given Dr Knights insight into the practical issues
associated with the production of silicon-based photonic devices. The
combination of experience from Professor Reed and Dr Knights brings
together a complementary set of know-how to produce a well-written,
comprehensive, introductory book on silicon photonics.
We are at the beginning of a new era in communications, one in
which silicon photonics may play a significant role. This book should
give readers the foundation on which to participate in this exciting field.
When you have read this book I look forward to seeing you some day
on this wonderful journey to ‘siliconize’ photonics.

Dr Mario Paniccia
Dr Paniccia is Director of Optical Technology Development
within the Communications & Interconnect Technology Lab
at Intel Corporation
Acknowledgements

The experience of producing a technical text is a time-consuming

business. In particular, the chances of producing an error-free text
appear to us to be almost negligible. Therefore we have enlisted help
from a number of people to try to minimise the errors in our text, by
asking them to read and comment on both the content and accuracy.
This has proved invaluable to us, and has identified several omissions
and errors. In particular we would like to express our sincere gratitude
to the following people:

Dr Mario Paniccia for suggestions and advice on the content of the text,
and for his attitude of support and encouragement, which has gone a
long way towards getting this book published
Goran Masanovic for meticulous proof reading and diagram production
C. E. Jason Png for additional device simulation, and redrafting some of
the diagrams
William Headley, Goran Masanovic, C. E. Jason Png, Seong Phun Chan,
Dr Richard Jones, Soon Thor Lim, and Dr Haisheng Rong for finding
typographical errors and/or technical errors

All those who allowed us to use their diagrams or other material.

1
Fundamentals

This chapter is intended to reinforce some fundamental concepts before

we embark upon a discussion of silicon photonics. When producing
such a text, one always faces the philosophical question of ‘How much
knowledge to assume?’ We have chosen to reinforce some fundamental
concepts in this chapter, in order to make the text suitable for a
wide range of readers. These preliminary discussions will simplify the
parameters discussed to some extent to enable the reader to make some
rapid progress. However, the more experienced reader should not be
put of by the very fundamental nature of this chapter, as it is essential
to understand the fundamentals before full use can be made of more
advanced material.

1.1 WHAT IS PHASE?

The understanding of phase of an optical wave is fundamental to under-

standing how waveguides and optical circuits operate. In its simplest
terms the concept of phase is very straightforward. If we are dealing
with a function of time, the meaning of phase is simply the proportion
of a periodic waveform that passes some reference point, after a time t,
typically with respect to time t = 0.
Let us begin by plotting the function sin θ, as in Figure 1.1. The phase
of the waveform is simply the angle θ. It is implicit that the reference
value is 0. Things are a little more complicated if we plot a sinusoidal
function that varies with time. For example a plot of sin ωt is shown in
Figure 1.2, with a time axis. In this case the parameter ω is a constant

Silicon Photonics: An Introduction Graham T. Reed and Andrew P. Knights

 2004 John Wiley & Sons, Ltd ISBN: 0-470-87034-6
2 FUNDAMENTALS

1.0

0.5

3p/2 2p
sin q

0.0 q
0 p/2 p 5p/2 6p

−0.5

−1.0

Figure 1.1 Plot of sin θ

1.0

0.5

3 p/2w
sin wt

2p/w
0.0 t
0 p/2w p/w 5p/2w 3p/w

−0.5

−1.0

Figure 1.2 Plot of sin ωt

called the angular frequency. It is related to the frequency, f , of the

waveform by the relationship:

ω = 2πf (1.1)

Once again the phase of the function is the angle, in this case ωt, but
note that because the period of the function is always 2π radians, and
the plot is on a time axis and not an ωt axis, the axis crossings now
occur every π/ω seconds rather than every π radians.
Let us now consider a propagating optical wave. Such a wave is
expressed as:
E = E0 exp[j(kz ± ωt)] (1.2)
WHAT IS PHASE? 3

The exponential function is used for generality, but under some circum-
stances this can be simplified to a sinusoidal function. For example, let
us use a sinusoidal function to help to understand phase a little more:

E = E0 sin(kz ± ωt) (1.3)

Following from our previous discussion, the phase is the angle of the
sinusoidal function. Thus the phase is:

(kz ± ωt) (1.4)

In this expression, ωt has the same meaning as before, and the term kz
is used to describe the progress of the wave in the z direction. k is a
constant, similar in concept to ω, which determines the rate of progress
of the wave with distance, z.
We can think of the propagating wave, then, as moving in a direction
z, but also varying with time. This is very like a wave moving in sea
water. The peak of the wave moves through the water at a certain rate
(or speed), but at a fixed position in the sea the size of the wave also
varies with time.
Therefore equation 1.4 tells us that the wave varies with distance,
at a rate determined by the constant k, and varies with time at a rate
determined by the constant ω. If we stay at a fixed position and consider
the variation of the wave with time, we obtain the variation shown in
Figure 1.2.
If we now consider the variation of the wave with distance at a fixed
time, we can plot the function sin kz, as in Figure 1.3. Notice that we
are now plotting the sinusoidal function against a distance on the z axis.
The period of the waveform must still be 2π radians, and therefore the
axis crossings now occur every π/k metres rather than every π radians.
So k has units of radians per metre, and determines how much of a phase
change is experienced per metre of distance that the wave propagates.
Therefore k is known as the propagation constant. The propagation
constant will be discussed in more detail later in this text.
We can also note in passing that the wavelength of the function can be
measured directly from Figure 1.3, because the axis has units of metres.
The wavelength is the distance between two peaks of the function,
which from Figure 1.3 can be seen to be 2π/k. Therefore we can write
an expression for wavelength, λ, as:

2π
λ= (1.5)
k
4 FUNDAMENTALS

1.0

0.5

3 p/2k 2p/k
sin kz

0.0 z
0 p/2k p/k 5p/2k 3p/k

−0.5

−1.0

Figure 1.3 Plot of sin kz

Obviously, then, k must be given by:

2π
k= (1.6)
λ

If we refer to the phase change over a distance z as φ, we can evaluate

this phase change as the product of the propagation constant and the
distance:
φ = kz (1.7)

In many applications in photonic circuits, we need to measure the

phase of one wave with respect to another, perhaps in an interferometer.
In such cases the time variation of both waves will be the same, so the
time variations are usually neglected. Hence the relative phase difference
between the waves is a function of the propagation constant k and the
distance of propagation z only. We will use this fact later in the text to
explain the operation of several photonic devices.

1.2 WHAT IS POLARISATION?

In order to understand optical waveguides, we must understand what

is meant by the polarisation of light. In order to gain such an under-
standing it is necessary to consider the nature of a lightwave. Light is
an electromagnetic wave, which has the characteristic of both electric
 Another Random Document on
Scribd Without Any Related Topics
In 1822 there were thus four great companies, having in all 47 gasometers at work, capable of
containing 917,940 cubic feet of gas, supplied by 1315 retorts, which generated per annum upwards of
397,000,000 cubic feet of gas, by which 61,203 private lamps, and 7268 public or street lamps, were
lighted in the metropolis. Besides these public companies, there were likewise several private ones.
1. Of the generation of illuminating gases.—Pure hydrogen gas burns with too feeble a flame to be
employed for illumination. But carburetted hydrogen having the property of precipitating its carbon in
the act of burning, its solid particles become incandescent, and diffuse a vivid light. The more carbon it
contains, the more brightly does it burn. This gas exists in two distinct states of combination. In the
first, two measures of hydrogen gas are combined with one measure of the vapour of carbon, forming
together one measure whose specific gravity is of course the sum of the weights of the constituents, or
0·559; atmospherical air being 1·000. This is the gas which is found in mines, and is also evolved in
ditches from decomposing vegetable matter. In the second, two measures of hydrogen gas are
combined with two of gaseous carbon, forming also one volume or measure whose weight or specific
gravity is 0·985. This was at one time called the olefiant gas, because when mixed with chlorine an oily
looking compound was produced. It may be called as well oil gas, because it is generated in
considerable quantities by the igneous decomposition of oil. Thus the olefiant gas contains in the same
volume double the quantity of carbon of common carburetted hydrogen, and it burns with a
proportionably brighter flame. The gaseous oxide of carbon, as well as sulphuretted hydrogen gas,
burns with a feeble blue light, but the latter produces in combustion sulphurous acid, an offensive and
noxious gas.
By dry distillation or carbonization in close vessels, all bodies of vegetable and animal origin disengage
carburetted hydrogen gas; even charcoal when placed in ignition in contact with steam, by
decomposing the water, produces abundance of carbonic acid, carburetted hydrogen, hydrogen, and
carbonic oxide. After separating the carbonic acid with lime water, that mixed gas contains in 100
measures, 20 of carburetted hydrogen; the rest being hydrogen and carbonic oxide, so that the gaseous
mixture cannot be used for illumination. The best substances for furnishing a gas rich in luminifereous
materials are, pitcoal, especially the cannel coal, resin, oil, fats of all kinds, tar, wax, &c. In some cases
the gases evolved during the igneous decomposition of bones and other animal matters for the
production of ammonia, may be employed for procuring light, but they are apt to emit a fetid odour.
When coals are heated in a cast-iron retort to ignition, the progress of decomposition is as follows. First,
and before the retort becomes red hot, steam issues along with the atmospheric air. When the retort
begins to redden, tar distils in considerable quantity with some combustible gas, of which hydrogen
mixed with ammoniacal gas forms a part. The evolution of gas increases as the retort becomes hotter,
with a continual production of tar and ammoniacal liquor as well as sulphurous acid from the pyrites of
the coal, which unites with the ammonia. When the retort has come to a bright cherry red heat, the
disengagement of gas is most active. By and bye the gaseous production diminishes, and eventually
ceases entirely, although the heat be increased. In the retort a quantity of carbonized coal or coke
remains, while tar is found at the bottom of the receiver, covered with the ammoniacal liquor, and
combined with carbonic and sulphurous acids, and sulphuretted hydrogen.
If during this distillation, the combustible gas be collected and examined at the several stages of the
process, it is found to differ extremely in its luminiferous powers. That which comes off before the retort
has acquired its proper temperature, gives a feeble light, and resembles the gas obtained by the ignition
of moist charcoal, consisting chiefly of hydrogen. That evolved when the retort has just acquired
throughout a vivid red heat, is the best of all, consisting chiefly of bicarburetted hydrogen or olefiant
gas. From good coal, it consists, for example in 100 measures, of 13 of olefiant gas, 82·5 of carburetted
hydrogen, 3·2 carbonic oxide, 1·3 azote; the mixture having a specific gravity of 0·650. At a later
period, as after 5 hours, it contains 7 measures of olefiant gas, 56 of carburetted hydrogen, 11 of
carbonic oxide, 21·3 of hydrogen, 4·7 of azote; the specific gravity of the whole being 0·500. Towards
the end of the operation, as after 10 hours, it contains twenty measures of carburetted hydrogen, 10 of
carbonic oxide, 60 of hydrogen, 10 of azote, with a specific gravity of only 0·345. The hydrogen
becomes sulphuretted hydrogen, if there be much pyritous matter in the coal. The larger proportion of
the gas is disengaged during the first hour, amounting to about one fifth of the whole; in the three
following hours the disengagement is tolerably uniform, constituting in all fifty-four hundredths; in the
sixth hour, it is one tenth; in the seventh and eighth hours, sixteen hundredths.
From these observations are derived the rules for the production of a good light gas from coals. They
show that the distillation should commence with a retort previously heated to a cherry red, since
thereby good gas is immediately produced, and a portion of the tar is also converted into gas, instead
of being simply distilled over into the condenser pit; that this heat should be steadily continued during
the whole operation, from 5 to 8 hours; that it should not be increased, especially towards the end, for
fear of generating carbonic oxide and hydrogen gases, as well as of injuring the retort when the cooling
agency of gasefication has become feeble; and that the operation should be stopped some time before
gas ceases to come over, lest gases with feeble illuminating power should impoverish the contents of
the gasometer. Upon the average, a pound of good coal affords four cubic feet of gas, or a chaldron =
26 cwt. London measure, affords from 12,000 to 15,000 cubic feet, according to the form of the retort,
and the manner of firing it.
When oil, fats, rosin, tar, &c. are employed for the production of a light gas, it is not sufficient to
introduce these substances into the retorts, and to heat them, as is done with coals. In this case, the
greater part of them would distil over in the state of volatile oils, and very little gas be generated, only
as much as corresponded to the quantity of fat, &c. in immediate contact with the retort. It becomes
therefore necessary to fill the retorts with pieces of brick or coke; and to keep them in ignition, while
the oil, &c. is slowly introduced into their interior. The fats instantly assume the vaporous state, and
thus coming into contact upon an extensive surface with the ignited bricks, are decomposed into
combustible gases. A small portion of carbonaceous matter remains in the retort, while much olefiant
gas is formed, possessing a superior illuminating power to common coal gas, and entirely free from
sulphureous impregnation. The best oil gas is generated at a dull red, a heat much below what is
requisite for the decomposition of coal. A more intense heat would indeed produce a greater volume of
gas, but of a poorer quality, because the olefiant gas thereby deposits one half of its carbon, and is
converted into common carburetted hydrogen. Oil affords at a lively red heat, gases which contain in
100 measures, 19 of olefiant gas, 32·4 of carburetted hydrogen, 12·2 of carbonic oxide gas, 32·4 of
hydrogen, and 4 of azote; the mean specific gravity being only 0·590. At a more moderate temperature
it yields 22·5 of the olefiant, 50·3 carburetted hydrogen, 15·5 carbonic oxide, 7·7 hydrogen, and 4
azote, with a specific gravity of 0·758. It contains when generated by dull ignition, as is usual in works
on the manufacturing scale, in 100 parts, from 38 to 40 of olefiant gas, and besides the carburetted
hydrogen, a few per cents. of carbonic oxide and azote, with a specific gravity of 0·900, and even
upwards. One pound of oil or fluid fat affords 15 cubic feet of gas; of tar affords about 12 cubic feet; of
rosin or pitch, 10 cubic feet.
When the oil gas is compressed by a force of from 15 to 20 atmospheres, as was the practice of the
Portable Gas Company, about one fifth of the volume of the gas becomes liquefied into an oily, very
volatile fluid, having the specific gravity 0·821. It is a mixture of three fluids (consisting of carburetted
hydrogen), of different degrees of volatility. The most volatile of these boils even under 32° F. Some of
the vapour of this gas-oil is mixed with the olefiant gas in the general products of decomposition; in
consequence of which they are sometimes richer in carbon than even olefiant gas, and have a higher
illuminating power. Oil gas contains about 22 per cent. and coal gas about 31⁄4 per cent. of this oily
vapour. In the estimations of the composition of the gases given above, this vapour is included under
olefiant gas. This vapour combines readily with sulphuric acid, and is thus precipitated from the gaseous
mixture. The amount of olefiant gas is shown, by adding to the gas, contained over water, one half of
its volume of chlorine, which, in the course of an hour or two, condenses the olefiant gas into an oily
looking liquid (chloride of hydrocarbon.) After the mixture, the gases must be screened from the light,
otherwise the common carburetted hydrogen would also combine with the chlorine, while water and
carbonic acid would make their appearance.
The oil employed for affording gas is the crudest and cheapest that can be bought; even the blubber
and sediment of whale oil are employed with advantage. After all, however, coal is so much cheaper,
and the gas produced from it is now so well purified, that oil and rosin are very little used in gas
apparatus.
Apparatus for Coal Gas.—Coal gas, as it issues from the retort, cannot be directly employed for
illumination; for it contains vapours of tar and coal oil, as also steam impregnated with the carbonate,
sulphite, and hydrosulphuret of ammonia. These vapours would readily condense in the pipes through
which the gas must be distributed, and would produce obstructions; they must therefore be so far
removed by previous cooling, as to be liable to occasion no troublesome condensation at ordinary
temperatures. The crude coal gas contains moreover sulphuretted hydrogen, whose combustion for light
would exhale an offensive sulphureous odour, that ought to be got rid of as much as possible. Carbonic
acid and carbonic oxide gases, generated at first from the decomposition of the steam by the ignited
coal, enfeeble the illuminating power of the gas, and should be removed. The disengagement of gas in
the retorts is never uniform, but varies with the degree of heat to which they are exposed; for which
reason the gas must be received in a gasometer, where it may experience uniform pressure, and be
discharged uniformly into the pipes of distribution, in order to ensure a steady discharge of gas, and
uniform intensity of light in the burners. A coal gas apparatus ought therefore to be so constructed as
not only to generate the gas itself, but to fulfil the above conditions.
In fig. 482., such an apparatus is represented, where the various parts are shown connected with each
other, in section.

Fig. 482 enlarged (201 kB)

A is the furnace with its set of cylindrical or elliptical retorts, five in number. From each of these retorts,
a tube b proceeds perpendicularly upwards, and then by a curve or saddle-tube, it turns downwards,
where it enters a long horizontal cylinder under B, shut at each end with a screw cap, and descends to
beneath its middle, so as to dip about an inch into the water contained in it. From one end of this
cylinder the tube d passes downwards, to connect itself with a horizontal tube which enters into the tar
pit or cistern C, by means of the vertical branch f. This branch reaches to near the bottom of the
cylindrical vessel, which sits on the sole of the tar cistern. From the other side of the vertical branch f,
the main pipe proceeds to the condenser D, and thence by the pipe l, into the purifier E; from which the
gas is immediately transmitted by the pipe p into the gasometer F.
The operation proceeds in the following way:—As soon as gas begins to be disengaged from the ignited
retort, tar and ammoniacal liquor are deposited in the cylindrical receiver B, and fill it up till the
superfluity runs over by the pipe d, the level being constantly preserved at the line shown in the figure.
By the same tarry liquid, the orifices of the several pipes b, issuing from the retorts, are closed;
whereby the gas in the pipe d has its communication cut off with the gas in the retorts. Hence if one of
the retorts be opened and emptied, it remains shut off from the rest of the apparatus. This insulation of
the several retorts is the function of the pipe under B, and therefore the recurved tube b must be
dipped as far under the surface of the tarry liquid, as to be in equilibrio with the pressure of the gas
upon the water in the purifier. The tube b is closed at top with a screw cap, which can be taken off at
pleasure, to permit the interior to be cleansed.
Both by the overflow from the receiver-pipe B, and by subsequent condensation in the tube d, tar and
ammoniacal liquor collect progressively in the cistern or pit under C, by which mingled liquids the lower
orifice of the vertical tube f is closed, so that the gas cannot escape into the empty space of this cistern.
These liquids flow over the edges of the inner vessel when it is full, and may, from time to time, be
drawn off by the stopcock at the bottom of the cistern.
Though the gas has, in its progress hitherto, deposited a good deal of its tarry and ammoniacal
vapours, yet, in consequence of its high temperature, it still retains a considerable portion of them,
which must be immediately abstracted, otherwise the tar would pollute the lime in the vessel E, and
interfere with its purification. On this account the gas should, at this period of the process, be cooled as
much as possible, in order to condense these vapours, and to favour the action of the lime in the
purifier E, upon the sulphuretted hydrogen, which is more energetic the lower the temperature of the
gas. The coal gas passes, therefore, from the tube f into the tube h of the condenser D, which is placed
in an iron chest g filled with water, and it deposits more tar and ammoniacal liquor in the under part of
the cistern at t, t. When these liquids have risen to a certain level, they overflow into the tar-pit, as
shown in the figure, to be drawn off by the stopcock as occasion may require.
The refrigerated gas is now conducted into the purifier E, which is filled with milk of lime, made by
mixing one part of slaked lime with 25 parts of water. The gas, as it enters by the pipe l, depresses the
water in the wide cylinder n, thence passes under the perforated disc in the under part of that cylinder,
and rising up through innumerable small holes is distributed throughout the lime liquid in the vessel m.
By contact with the lime on this extended surface, the gas is stripped of its sulphuretted hydrogen and
carbonic acid, which are condensed into the hydro-sulphuret and carbonate of lime; it now enters the
gasometer F in a purified state, through the pipe p t, and occupies the space q. The gasometer, pressing
with a small unbalanced force over the counterweight s, expels it through the main u u, in
communication with the pipes of distribution through the buildings or streets to be illuminated.
The parts A B C D E and F, of which this apparatus consists, are essential constituents of every good coal-
gas work. Their construction rests upon peculiar principles, is susceptible of certain modifications, and
therefore deserves to be considered in detail.
The Retorts.—These are generally made of cast iron, though they have occasionally been made of
baked clay, like common earthenware retorts. The original form was a cylinder, which was changed to
an ellipse, with the long axis in a horizontal direction, then into the shape of the letter D with the
straight line undermost, and lastly into a semi-cylinder, with its horizontal diameter 22 inches, and its
vertical varying from 9 to 12. The kidney form was at one time preferred, but it has been little used of
late.
The form of retort represented in fig. 483. has been found to yield the largest quantity of good gas in
the shortest time, and with the least quantity of firing. The length is 71⁄2, and the transverse area, from
one foot to a foot and a half square. The arrows show the direction of the flame and draught in this
excellent bench of retorts, as mounted by Messrs. Barlow.
The charge of coals is most conveniently introduced in a tray of sheet iron, made somewhat like a
grocer’s scoop, adapted to the size of the retort, which is pushed home to its further end, inverted so as
to turn out the contents, and then immediately withdrawn.
The duration of the process, or the time of completing a distillation, depends upon the nature of the
coal and the form of the retort. With cylindrical retorts it cannot be finished in less than 6 hours, but
with elliptical and semi-cylindrical retorts, it may be completed in 4 or 5 hours. If the distillation be
continued in the former for 8 hours, and in the latter for 6, gas will continue to be obtained, but during
the latter period of the operation, of indifferent quality.

The Receiver.—If the furnace contains only 2 or 3 retorts, a simple cylindrical vessel standing on the
ground half filled with water, may serve as a receiver; into which the tube from the retort may be
plunged. It should be provided with an overflow pipe for the tar and ammoniacal liquor. For a range of
several retorts, a long horizontal cylinder is preferable, like that represented at B in fig. 484. Its
diameter is from 10 to 15 inches. This cylinder may be so constructed as to separate the tar from the
ammoniacal liquor, by means of a syphon attached to one of its ends.
The Condenser.—The condenser, represented in fig. 482., consists of a square chest, g, made of
wrought iron plates open at top, but having its bottom pierced with a row of holes, to receive a series of
tubes. To these holes the upright four-inch tubes h h are secured by flanges and screws, and they are
connected in pairs at top by the curved or saddle tubes. The said bottom forms the cover of the chest t,
t, which is divided by vertical iron partitions, into half as many compartments as there are tubes.
These partition plates are left open at bottom, so as to place the liquids of each compartment in
communication. Thereby the gas passes up and down the series of tubes, in proceeding from one
compartment to another. The condensed liquids descend into the box t, t, and flow over into the tar
cistern, when they rise above the level t, t. The tar may be drawn off from time to time by the stopcock.
Through the tube k, cold water flows into the condenser chest, and the warm water passes away by a
pipe at its upper edge.
The extent of surface which the gas requires for its refrigeration before it is admitted into the washing-
lime apparatus, depends upon the temperature of the milk of lime, and the quantity of gas generated in
a certain time.
It may be assumed as a determination sufficiently exact, that 10 square feet of surface of the
condenser can cool a cubic foot of gas per minute to the temperature of the cooling water. For example,
suppose a furnace or arch with 5 retorts of 150 pounds of coal each, to produce in 5 hours 3000 cubic
feet of gas, or 10 cubic feet per minute, there would be required, for the cooling surface of the
condenser, 100 square feet = 10 × 10. Suppose 100,000 cubic feet of gas to be produced in 24 hours,
for which 8 or 9 such arches must be employed, the condensing surface must contain from 800 to 900
square feet.
The Purifier.—The apparatus represented in the preceding figure is composed of a cylindrical iron vessel,
with an air-tight cover screwed upon it, through which the cylinder n is also fixed air-tight. The bottom
of this cylinder spreads out like the brim of a hat, forming a horizontal circular partition, which is pierced
with holes. Through a stuffing box, in the cover of this interior cylinder, the vertical axis of the agitator
passes, which is turned by wheel and pinion work, in order to stir up the lime from the bottom of the
water in the purifier. The vessel o serves for introducing fresh milk of lime, as also for letting it off by a
stopcock when it has become too foul for further use.
The quantity of lime should be proportioned to the quantity of sulphuretted hydrogen and carbonic acid
contained in the gas. Supposing that in good coal gas there is 5 per cent. of these gases, about one
pound and a half of lime will be requisite for every hundred cubic feet of coal gas generated, which
amounts to nearly one-sixteenth of the weight of coal subjected to decomposition. This quantity of lime
mixed with the proper quantity of water will form about a cubic foot of milk of lime. Consequently, the
capacity of the purifier, that is, of the interior space filled with liquid, may be taken at four-sevenths of a
cubic foot for every hundred cubic feet of gas passing through it in one operation; or for 175 cubic feet
of gas, one cubic foot of liquor. After every operation, that is, after every five or six hours, the purifier
must be filled afresh. Suppose that in the course of one operation 20,000 cubic feet of gas pass through
the machine, this should be able to contain 20,000 = 114 cubic feet of milk of lime; whence its diameter
175
should be seven feet, and the height of the liquid three feet. If the capacity of the vessel be less, the
lime milk must be more frequently changed.
In some of the large gas works of London the purifier has the following construction, whereby an
uninterrupted influx and efflux of milk of lime takes place. Three single purifiers are so connected
together, that the second vessel stands higher than the first, and the third than the second; so that the
discharge tube of the superior vessel, placed somewhat below its cover, enters into the upper part of
the next lower vessel; consequently, should the milk of lime in the third and uppermost vessel rise
above its ordinary level, it will flow over into the second, and thence in the same way into the first; from
which it is let off by the eduction pipe. A tube introduces the gas from the condenser into the first
vessel, another tube does the same thing for the second vessel, &c., and the tube of the third vessel
conducts the gas into the gasometer. Into the third vessel, milk of lime is constantly made to flow from
a cistern upon a higher level. By this arrangement, the gas passing through the several vessels in
proportion as it is purified, comes progressively into contact with purer milk of lime, whereby its
purification becomes more complete. The agitator c, provided with two stirring paddles, is kept in
continual rotation. The pressure which the gas has here to overcome is naturally three times as great as
with a single purifier of like depth.
Fig. 485. is a simple form of purifier, which has been found to answer well in practice. Through the
cover of the vessel A B, the wide cylinder e d is inserted, having its lower end pierced with numerous
holes. Concentric with that cylinder is the narrower one s z, bound above with the flange a b, but open
at top and bottom. The under edge g h of this cylinder descends a few inches below the end c d of the
outer one. About the middle of the vessel the perforated shelf m n is placed. The shaft of the agitator l,
passes through a stuffing box upon the top of the vessel. The gas-pipe g, proceeding from the
condenser, enters through the flange a b in the outer cylinder, while the gas-pipe h goes from the cover
to the gasometer. A stopcock upon the side, whose orifice of discharge is somewhat higher than the
under edge of the outer cylinder, serves to draw off the milk of lime. As the gas enters through the pipe
g into the space between the two cylinders, it displaces the liquor till it arrives at the holes in the under
edge of the outer cylinder, through which, as well as under the edge, it flows, and then passes up
through the apertures of the shelf m n into the milk of lime chamber; the level of which is shown by the
dotted line. The stirrer, l, should be turned by wheel work, though it is here shown as put in motion by a
winch handle.
In order to judge of the degree of purity of the gas after its transmission through the lime machine, a
slender syphon tube provided with a stopcock may have the one end inserted in its cover, and the other
dipped into a vessel containing a solution of acetate of lead. Whenever the solution has been rendered
turbid by the precipitation of sulphuret of lead, it should be renewed. The saturated and fetid milk of
lime is evaporated in oblong cast-iron troughs placed in the ash-pit of the furnaces, and the dried lime is
partly employed for luting the apparatus, and partly disposed of for a mortar or manure.
By this purifier, and others of similar construction, the gas in the preceding parts of the apparatus, as in
the retorts and the condenser, suffers a pressure equal to a column of water about two feet high; and in
the last described purifier even a greater pressure. This pressure is not disadvantageous, but is of use in
two respects; 1. it shows by a brisk jet of gas when the apparatus is not air-tight, and it prevents
common air from entering into the retorts; 2. this compression of the gas favours the condensation of
the tar and ammoniacal liquor. The effect of such a degree of pressure in expanding the metal of the
ignited retorts is quite inconsiderable, and may be neglected. Two contrivances have, however, been
proposed for taking off this pressure in the purifier.
In fig. 486., m m are two similar vessels of a round or rectangular form, furnished at their upper border
with a groove filled with water, into which the under edge of the cover fits, so as to make the vessel air-
tight. The cover is suspended by a cord or chain, which goes over a pulley, and may be raised or
lowered at pleasure. The vessels themselves have perforated bottoms, r r′, covered with wetted moss or
hay sprinkled over with slaked and sifted quicklime. The gas passes through the loosely compacted
matter of the first vessel, by entering between its two bottoms, rises into the upper space t, thence it
proceeds to the second vessel, and, lastly, through the pipe u into the gasometer. This method,
however, requires twice as much lime as the former, without increasing the purity of the gas.

The second method consists in compressing the gas by the action of an Archimedes screw, to such a
degree, before it is admitted into the purifier, as that it may overcome the pressure of the column of
water in that vessel. Fig. 487. exhibits this apparatus in section. D D is the Archimedes worm, the axis of
which revolves at bottom upon the gudgeon e; it possesses a three-fold spiral, and is turned in the
opposite direction to that in which it scoops the water. The cistern which contains it has an air-tight
cover. The gas to be purified passes through the pipe C into the space D, over the water level d; the
upper cells of the worm, scoop in the gas at this point, and carry it downwards, where it enters at g into
the cavity E of a second cistern. In order that the gas, after it escapes from the bottom of the worm,
may not partially return through g into the cavity D, an annular plate g h is attached to its under edge,
so as to turn over it. The compressed gas is conducted from the cavity E through the pipe G into the
purifying machine; a is a manometer, to indicate the elastic tension of the gas in D. On the top of the
worm a mechanism is fitted for keeping it in constant rotation.
A perfect purification of light-gas from sulphuretted hydrogen, either by milk of lime or a solution of the
green sulphate of iron, is attended with some difficulty, when carried so far as to cause no precipitation
of sulphuret in acetate of lead, because such a degree of washing is required as is apt to diminish its
illuminating power, by abstracting the vapour of the rich oily hydrocarburet which it contains. Moreover,
the coal gas obtained towards the end of the distillation contains some sulphuret of carbon, which
affords sulphurous acid on being burned, and can be removed by no easy method hitherto known. The
lime in the purifier disengages from the carbonate and hydrosulphuret of ammonia carried over with the
gas, especially when it has been imperfectly cooled in the condenser, a portion of ammoniacal gas,
which, however, is not injurious to its illuminating power. The best agent for purifying gas would be the
pyrolignite of lead, were it not rather expensive, because it would save the trouble of stirring, and
require a smaller and simpler apparatus.
The Gasometer.—The gasometer serves not merely as a magazine for receiving the gas when it is
purified, and keeping it in store for use, but also for communicating to the gas in the act of burning
such an uniform pressure as may secure a steady unflickering flame. It consists of two essential parts;
1. of an under cistern, open at top and filled with water; and 2. of the upper floating cylinder or chest,
which is a similar cistern inverted, and of somewhat smaller dimensions, called the gas-holder: see F,
fig. 482. The best form of this vessel is the round or cylindrical; both because under equal capacity it
requires least surface of metal, and it is least liable to be warped by its own weight or accidents. Since a
cylindrical body has the greatest capacity with a given surface when its height is equal to its semi-
diameter, its dimensions ought to be such that when elevated to the highest point in the water, the
height may be equal to the radius of the base. For example, let the capacity of the gas-holder in cubic
feet be k, the semi-diameter of its base be x, the height out of the water be h; h is = x = ∛3·14
k . This

height may be increased by one or two feet, according to its magnitude, to prevent the chance of any
gas escaping beneath its under edge, when it is raised to its highest elevation in the water.
The size of the gasometer should be proportional to the quantity of gas to be consumed in a certain
time. If 120,000 cubic feet be required, for instance, in 10 hours for street illumination, and if the gas
retorts be charged four times in 24 hours, 30,000 cubic feet of gas will be generated in 6 hours. Hence
the gasometer should have a capacity of at least 70,000 cubic feet, supposing the remaining 50,000
cubic feet to be produced during the period of consumption. If the gasometer has a smaller capacity, it
must be supplied from a greater number of retorts during the lighting period, which is not
advantageous, as the first heating of the supernumerary retorts is wasteful of fuel. Some engineers
consider that a capacity of 30,000 cubic feet is the largest which can with propriety be given to a
gasometer; in which case, they make its diameter 42 feet, and its height 23. When the dimensions are
greater, the sheet iron must be thicker and more expensive; and the hollow cylinder must be fortified by
strong internal cross braces.
The water cistern is usually constructed in this country with cast-iron plates bolted together, and made
tight with rust-cement.
In cases where the weight of water required to fill such a cistern might be inconvenient to sustain, it
may be made in the form represented in fig. 488.; which, however, will cost nearly twice as much.
Parallel with the side of the cistern, a second cylinder C, of the same shape but somewhat smaller, is
fixed in an inverted position to the bottom of the first, so as to leave an annular space B B between
them, which is filled with water, and in which the floating gasometer A plays up and down. The water
must stand above the cover of the inverted cylinder. a and b are the pipes for leading the gas in and
out. Through an opening in the masonry upon which the gasometer apparatus rests, the space C may
be entered, in order to make any requisite repairs.
The water cistern may also be sunk in the ground, and the sides made tight with hydraulic mortar, as is
shown in fig. 489., and to make it answer with less water, a concentric cylindrical mass of masonry may
be built at a distance of 2 or 3 inches within it.
Every large gasometer must be strengthened interiorly with cross iron rods, to stiffen both its top and
bottom. The top is supported by rods stretching obliquely down to the sides, and to the under edge an
iron ring is attached, consisting of curved cast-iron bars bolted together; with which the oblique rods
are connected by perpendicular ones. Other vertical rods stretch directly from the top to the bottom
edge. Upon the periphery of the top, at the end of the rods, several rings are made fast, to which the
gas-holder is suspended, by means of a common chain which runs over a pulley at the centre. Upon the
other end of the chain there is a counterpoise, which takes off the greater part of the weight of the gas-
holder, leaving only so much as is requisite for the expulsion of the gas. The inner and outer surfaces of
the gas-holder should be a few times rubbed over with hot tar, at a few days’ interval between each
application. The pulley must be made fast to a strong frame.
If the water cistern be formed with masonry, the suspension of the gas-holder may be made in the
following way. A A, fig. 489., is a hollow cylinder of cast iron, standing up through the middle of the
gasometer, and which is provided at either end with another small hollow cylinder G, open at both ends
and passing through the top, with its axis placed in the axis of the gas-holder. In the hollow cylinder G,
the counterweight moves up and down, with its chain passing over the three pulleys B, B, B, as shown in
fig. 489. E F are the gas pipes made fast to a vertical iron rod. Should the gasometer be made to work
without a counterweight, as we shall presently see, the central cylinder A A, serves as a vertical guide.
In proportion as the gas-holder sinks in the water of the cistern, it loses so much of its weight, as is
equal to the weight of the water displaced by the sides of the sinking vessel; so that the gas-holder
when entirely immersed, exercises the least pressure upon the gas, and when entirely out of the water,
it exercises the greatest pressure. In order to counteract this inequality of pressure, which would
occasion an unequal velocity in the efflux of the gas, and of course an unequal intensity of light in its
flame, the weight of the chain upon which the gas-holder hangs is so adjusted as to be equal,
throughout the length of its motion, to one half of the weight which the gas-holder loses by immersion.
In this case, the weight which it loses by sinking into the water, is replaced by the portion of the chain
which passing the pulley, and hanging over, balances so much of the chain upon the side of the
counterweight; and the weight which it gains by rising out of the water, is counterpoised by the links of
the chain which passing over the pulley, add to the amount of the counterweight. The pressure which
the gas-holder exercises upon the gas, or that with which it forces it through the first main pipe, is
usually so regulated as to sustain a column of from one to two inches of water; so that the water will
stand in the cistern from one to two inches higher within, than without the gas-holder. The following
computation will place these particulars in a clear light.
Let the semi-diameter of the gas-holder, equal to the vertical extent of its motion into and out of the
water, = x; let the weight of a foot square of the side of the gas-holder, including that of the
strengthening bars and ring, which remain plunged under the water, be = p; then
1. the weight of the gas-holder in its highest position = 3 p π x2;
2. the weight of the sides of the gas-holder which play in the water = 2 p π x2;

3. the cubic contents of the immersed portion of the gas-holder = 2 p π x ;

2

400

4. its loss of weight in water = 112 p π x2;

400
 (
5. the weight of the gas-holder in its lowest position = p π x2 3 - 112 = 2·72 p π x2;
400
)
6. the weight of n inches, height of water = 56 n π x2;
12

(
7. the amount of the counterweight = π x2 3 p - 56 n ;
12
)
8. the weight of the chain for the length x = 112 p π x.
800

If we reduce the weight of the gas-holder in its highest and lowest positions to the height of a stratum
of water equal to the surface of its top, this height is that of the column of water which would press the
gas within the gasometer, were no counterweight employed; it consists as follows;—

9. for the highest position = 3 p;

56

10. for the lowest = 2·72 p;

50

For the case, when the height of the gas-holder is different from its semi-diameter, let this height = m
x; then the height of the water level is

(
11. for the highest position = p 1 + 2 m ;
56
)
(
12. for the lowest = p 1 + 1·72 m ;
6
)
(
13. the counterweight = π x2 p (1 + 2m) - 56 n ;
12
)
14. the weight of the equalizing chain = 112 p π m x2.
800

For example, let the diameter of the gas-holder be 30 feet, the height 15 (the contents in cubic feet will
be 10,597), p = 4 pounds; then the counterweight for a height of an inch and a half of water pressure
= 3532 pounds; the weight of the chain for a length of 15 feet = 395 pounds. Were no counterweight
employed, so that the gas-holder pressed with its whole weight upon the gas, then the height of the
equivalent column of water in its highest position = 2·56 inches; and in its lowest, 2·33. The
counterweight may hence be lessened at pleasure, if the height of the pressing water-column n be
increased. The weight of the equalising or compensating portion of the chain remains the same. When
n = 2 inches, for instance, the counterweight = 1886 pounds.
The velocity with which the gas passes along the mains for supplying the various jets of light, may be
further regulated by opening the main-cock or slide-valve in a greater or less degree.
Gasometers whose height is greater than their semi-diameter, are not only more costly in the
construction, but require heavier counterweights and equilibration chains.
The above estimate is made on the supposition of the gas in the gas-holder being of the same specific
gravity as the atmospherical air, which would be nearly true with regard to oil gas under the ordinary
pressure. But coal gas, whose specific gravity may be taken on an average at about 0·5, exercises a
buoyancy upon the top of the gas-holder, which of course diminishes its absolute weight. Supposing the
cubic foot of gas to be = 0·0364 pounds, the buoyancy will be = 0·0364 π x3 pounds; a quantity which
deserves to be taken into account for large gasometers. Hence,
15. the weight of the gas-holder in its highest position = 3 p π x2 - 0·1143 x3;
 ( )
16. the counterweight = π x2 3 p - 56 n - 0·1143 x2;
12

17. The weight of the chain for the length x, = 112 p π x2 0·1143 x ;
3

800 2

18. The height of the water pressure for the highest position, without the counterweight =
3 p π - 0·1143 x
;
56 π

19. the same for the lowest position = 2·72 p in feet.

56

The preceding values of p and x, are,

(16) = 3147; (17) = 203; (18) = 2·44 inches; (19) = 2·33 inches.
The water columns in the highest and lowest situations of the gas-holder here differ about 0·1 of an
inch, and this difference becomes still less when p has a smaller value, for example, 3 pounds, or when
the diameter of the gas-holder is still greater.
It would thus appear that for coal-gas gasometers, in which the height of the gas-holder does not
exceed its semi-diameter, and especially when it has a considerable size, neither a compensation chain
nor a counterweight is necessary. The only thing requisite, is to preserve the vertical motion of the gas-
holder by a sufficient number of guide rods or pillars, placed either within the water cistern, or round
about it. Should the pressure of the gas in the pipe proceeding from the gasometer, be less than in the
gasometer itself, this may be regulated by the main valve, or by water valves of various kinds. Or a
small intermediate regulating gasometer may be introduced between the great gas-holder, and the main
pipe of distribution. With a diameter of 61 feet in the gas-holder, the pressure in the highest and lowest
positions is the same.
The gasometers employed in storing up gas until required for use, occupy, upon the old plan, much
space, and are attended with considerable expense in erecting. The water tank, whether sunk in the
ground, or raised, must be of equal dimensions with the gasometer, both in breadth and depth. The
improved construction which we are about to describe, affords a means of reducing the depth of the
tank, dispensing with the bridge of suspension, and of increasing at pleasure the capacity of the
gasometer, upon a given base; thus rendering a small apparatus capable, if required, of holding a large
quantity of gas, the first cost of which will be considerably less than even a small gasometer
constructed upon the ordinary plan.
Mr. Tait, of Mile-End Road, the inventor, has, we believe, been for some years connected with gas
establishments, and is therefore fully aware of the practical defects or advantages of the different
constructions of gasometers now in use. Fig. 490. is a section of Mr. Tait’s improved contrivance; a a is
the tank, occupied with water, b b two iron columns, with pulley-wheels on the top, c c, chains attached
to a ring of iron, d d, extending round the gasometer, which chains pass over the pulley-wheels, and are
loaded at their extremities, for the purpose of balancing the weight of the materials of which the
gasometer is composed.
The gasometer is formed by 2 or 3 cylinders, sliding one within the other, like the tubes of a telescope;
e, e, e, is the first or outer cylinder, closed at the top, and having the ring of iron d, passing round it, by
which the whole is suspended; f f, is the second cylinder, sliding freely within the first, and there may be
a third and fourth within these if necessary.
When there is no gas in the apparatus, all the cylinders are slidden down, and remain one within the
other immersed in the tank of water; but when the gas rises through the water pressing against the top
of the gasometer, its buoyancy causes the cylinder e to ascend. Round the lower edge of this cylinder a
groove is formed by the turning in of the plate of iron, and as it rises, the edge takes hold of the top rim
of the cylinder f, which is overlapped for that purpose. The groove at the bottom of the cylinder fills
itself with water as it ascends, and by the rim of the second cylinder falling into it, an air-tight hydraulic
joint is produced.
Thus, several cylinders may be adapted to act in a small tank of water, by sliding one within the other,
with lapped edges forming hydraulic joints, and by supporting the apparatus in the way shown, the
centre of gravity will always be below the points of suspension. A gasometer may be made upon this
plan of any diameter, as there will be no need of frame work, or a bridge to support it; and the
increasing weight of the apparatus, as the cylinders are raised one after the other, may be
counterpoised by loading the ends of the chains c c.
The water in the gasometer need not be renewed; but merely so much of it as evaporates or leaks out,
is to be replaced. Indeed the surface of the water in the cistern gets covered with a stratum of coal oil,
a few inches deep, which prevents its evaporation, and allows the gas to be saturated with this volatile
substance, so as to increase its illuminating powers.

The gasometer may be separated from the purifier by an intermediate vessel, such as is represented
fig. 491., with which the two gas pipes are connected. A is the cylindrical vessel of cast iron, a, the end
of the gas pipe which comes from the purifier, immersed a few inches deep into the liquid with which
the vessel is about two-thirds filled; b is the gas-pipe which leads into the gasometer, c is a
perpendicular tube, placed over the bottom of the vessel, and reaching to within one-third of the top,
through which the liquid is introduced into the vessel, and through which it escapes when it overflows
the level d. In this tube the liquid stands towards the inner level higher, in proportion to the pressure of
the gas in the gasometer. The fluid which is condensed in the gas pipe, b, and in its prolongation from
the gasometer, runs off into the vessel A; and therefore the latter must be laid so low that the said tube
may have the requisite declivity. A straight stop-cock may also be attached to the side over the bottom,
to draw off any sediment.
II. Application of Light-Gas.
1. Distribution of the pipes.—The pressure by which the motion of the gas is maintained in the pipes,
corresponds to a certain height of water in the cistern of the gasometer. From the magnitude of this
pressure, and the quantity of gas which in a given time, as an hour, must be transmitted through a
certain length of pipes, depends the width or the diameter that they should have, in order that the
motion may not be retarded by the friction which the gas, like all other fluids, experiences in tubes, and
thereby the gas might be prevented from issuing with the velocity required for the jets of flame. The
velocity of the gas in the main pipe increases in the ratio of the square root of the pressing column of
water upon the gasometer, and therefore by increasing this pressure, the gas may be forced more
rapidly along the remoter and smaller ramifications of the pipes. Thus it happens, however, that the gas
will be discharged from the orifices near the gasometer, with superfluous velocity. It is therefore
advisable to lay the pipes in such a manner, that in every point of their length, the velocity of discharge
may be nearly equal. This may be nearly effected as follows;—
From experiment it appears that the magnitude of the friction, or the resistance which the air suffers in
moving along the pipes, under a like primary pressure, that is for equal initial velocity, varies with the
square root of the length. The volume of gas discharged from the end of a pipe, is directly proportional
to the square of its diameter, and inversely as the square root of its length; or, calling the length L, the
2
diameter D, the cubic feet of gas discharged in an hour k; then k = D . Experience likewise shows, that
√L
for a pipe 250 feet long, which transmits in an hour 200 cubic feet of gas, one inch is a sufficient
diameter.

: D ; and D = √k √L
1 2
Consequently, 200 : k ∷
144 √250 √L 455,000

From this formula the following table of proportions is calculated.

Number of Length of Diameter,

cubic feet pipe, in inches.
per hour. in feet.
50 100 0·40
250 200 1·00
500 600 1·97
700 1000 2·65
1000 1000 3·16
1500 1000 3·87
2000 1000 4·47
2000 2000 5·32
2000 4000 6·33
2000 6000 7·00
6000 1000 7·75
6000 2000 9·21
8000 1000 8·95
8000 2000 16·65
These dimensions are applicable to the case where the body of gas is transmitted through pipes without
being let off in its way by burners, that is, to the mains which conduct the gas to the places where it is
to be used. If the main sends off branches for burners, then for the same length the diameter may be
reduced, or for like diameter the length may be greater. For example, if a pipe of 5·32 inches, which
transmits 2000 cubic feet through a length of 2000 feet, gives off, in this space, 1000 cubic feet of gas;
then the remainder of the pipe, having the same diameter, can continue to transmit the gas through a
( )
length of 2450 feet = 450,000 2, with undiminished pressure for the purposes of lighting. Inversely, the
k
diameter should be progressively reduced in proportion to the number of jets sent off in the length of
the pipe.
Suppose for instance, the gasometer to discharge 2000 cubic feet per hour, and the last point of the jets
to be at a distance of 4000 feet. Suppose also that from the gasometer to the first point of lighting, the
gas proceeds through 1000 feet of close pipe, the diameter of the pipe will be here 4·47 inches; in the
second 1000 feet of length, suppose the pipe to give off, at equal distances, 1000 cubic feet of gas, the
diameter in this length (calculated at 1500 cubic feet for 1000 feet long) = 3·87 inches; in the third
extent of 1000 feet, 600 cubic feet of gas will be given off, and the diameter (reckoning 700 cubic feet
for 1000 feet long) will be 2·65 inches; in the fourth and last space (for 200 cubic feet in 1000 feet
long) the pipe has a diameter of only an inch and a half, for which, in practice, a two-inch cast iron pipe
is substituted; this being the smallest used in mains, into which branch pipes can be conveniently
inserted.
The same relations hold with regard to branch pipes through which the gas is transmitted into buildings
and other places to be illuminated. If such pipes make frequent angular turnings, whereby they retard
the motion of the gas, they must be a third or a half larger in diameter. The smallest tubes of
distribution are never less than one fourth of an inch in the bore.
Where, from one central gas work, a very great quantity of light is required in particular localities, there
ought to be placed near these spots gasometers of distribution, which, being filled during the slack
hours of the day, are ready to supply the burners at night, without making any considerable demand
upon the original main pipe. Suppose the first main be required to supply 8000 cubic feet in the hour,
for an illumination of 8 hours, at the distance of 2000 feet, a pipe 102⁄3 inches in diameter would be
necessary; but if two or three gasometers of distribution, or station gasometers be had recourse to, into
which the gas during the course of 24 hours would flow through the same distance continuously from
the central gas works, the quantity required per hour from them would be only one third of 8000, =
2666·6 cubic feet; consequently the diameter for such a pipe is only 6·15 inches.

All the principal as well as branch pipes, whose interior diameter exceeds an inch and a half, are made
of cast iron from 6 to 8 feet long, with elbow pipes cast in them where it is necessary. These pipe
lengths are shown in fig. 492., having at one end a wide socket a, and at the other a nozzle b, which
fits the former. After inserting the one in the other in their proper horizontal position, a coil of hemp
soaked with tar is driven home at the junction; then a luting of clay is applied at the mouth, within
which a ring of lead is cast into the socket, which is driven tight home with a mallet and blunt chisel.
The pipes should be proved by a force pump before being received into the gas works; two or three
lengths of them should be joined before laying them down, and they should be placed at least two feet
below the surface, to prevent their being affected by changes of temperature, which would loosen the
joints. The tubes for internal distribution, when of small size are made of lead, copper, wrought iron, or
tin.

Instead of a stopcock for letting off the gas in regulated quantities from the gasometer, a peculiarly
formed water or mercurial valve is usually employed. Fig. 493. shows the mode of construction for a
water trap or lute, and is, in fact, merely a gasometer in miniature. C D E F is a square cast iron vessel,
in the one side of which a pipe A is placed in communication with the gasometer, and in the other, one
with the main B. The movable cover or lid H G I K has a partition, L M, in its middle. If this cover be
raised by its counterweight, the gas can pass without impediment from A to B; but if the counterweight
be diminished so as to let the partition plate L M sink into the water, the communication of the two pipes
is thereby interrupted. In this case the water-level stands in the compartment A so much lower than
outside of it, and in the compartment B, as is equivalent to the pressure in the gasometer; therefore the
pipes A and B must project thus far above the water. In order to keep the water always at the same
height, and to prevent it from flowing into the mouths of these pipes, the rim C D of the outer vessel
stands somewhat lower than the orifices A B; and thence the vessel may be kept always full of water.

If a quicksilver valve be preferred, it may be constructed as shown in fig. 494. A B are the terminations
of the two gas pipes, which are made fast in the rectangular iron vessel M. E is an iron vessel of the
same form, which is filled with quicksilver up to the level a, and which, by means of the screw G, which
presses against its bottom, and works in the fixed female screw C C, may be moved up or down, so that
the vessel M may be immersed more or less into the quicksilver. The vessel M is furnished with a vertical
partition m; the passage of the gas from A to B is therefore obstructed when this partition dips into the
quicksilver, and from the gradual depression of the vessel E by its screw, the interval between the
quicksilver and the lower edge of the partition, through which the gas must enter, may be enlarged at
pleasure, whereby the pressure of the gas in B may be regulated to any degree. The transverse section
of that interval is equal to the area of the pipe or rather greater; the breadth of the vessel M from A to B
amounts to the double of that space, and its length to the mere diameter of A or B. The greatest height
to which the partition m can rise out of the quicksilver, is also equal to the above diameter, and in this
case the line a comes to the place of b. The vertical movement of the outer vessel E, is secured by a
rectangular rim or hoop which surrounds it, and is made fast to the upper part of the vessel M, within
which guide it moves up and down. Instead of the lever D D, an index with a graduated plate may be
employed to turn the screw, and to indicate exactly the magnitude in the opening of the valve.

In order to measure the quantity of gas which passes through a pipe for lighting a factory, theatre, &c.,
the gas-meter is employed, of whose construction a sufficiently precise idea may be formed from the
consideration of fig. 495., which shows the instrument in a section perpendicular to its axis.
Within the cylindrical case a, there is a shorter cylinder b b, shut at both ends, and movable round an
axis, which is divided into four compartments, that communicate by the opening d, with the interval
between this cylinder and the outer case. The mode in which this cylinder turns round its axis is as
follows:—The end of the tube c, which is made fast to the side of the case, and by which the gas
enters, carries a pivot or gudgeon, upon which the centre of its prop turns; the other end of the axis
runs in the cover, which here forms the side of a superior open vessel, in which, upon the same axis,
there is a toothed wheel. The vessel is so far filled with water, that the tube c just rises above it, which
position is secured by the level of the side vessel. When the gas enters through the tube c, by its
pressure upon the partition e, (fig. 495.) it turns the cylinder from right to left upon its axis, till the
exterior opening d rises above the water, and the gas expands itself in the exterior space, whence it
passes off through a tube at top. At every revolution a certain volume of gas thus goes through the
cylinder, proportional to its known capacity. The wheel on the axis works in other toothed wheels,
whence, by means of an index upon a graduated disc or dial, placed at top or in front of the gas-meter,
the number of cubic feet of gas, which pass through this apparatus in a given time, is registered.
B. Employment of the gas for lighting.—The illuminating power of different gases burned in the same
circumstances, is proportional, generally speaking, to their specific gravity, as this is to the quantity of
carbon they hold in combination. The following table exhibits the different qualities of gases in respect
to illumination.

Density or specific gravity. Proportion of light

afforded by
 Coal gas. Oil gas. coal gas to oil gas.
0·659 0·818 100 : 140
0·578 0·910 100 : 225
0·605 1·110 100 : 250
0·407 0·940 100 : 354
0·429 0·965 100 : 356
0·508 1·175 100 : 310
Mean 0·529 0·96 100 : 272

In the last three proportions, the coal gas was produced from coals of middle quality; in the first three
proportions from coals of good quality; and therefore the middle proportion of 100 to 270 may be taken
to represent the fair average upon the great scale. On comparing the gas from bad coals, with good oil
gas, the proportion may become 100 to 300. Nay, coal gas of specific gravity 0·4, compared to oil gas of
1·1, gives the proportion of 1 to 4. A mould tallow candle, of 6 in the pound, burning for an hour, is
equivalent to half a cubic foot of ordinary coal gas, and to four tenths of a foot of good gas. The flame
of the best argand lamp of Carcel, in which a steady supply of oil is maintained by pump-work,
consuming 42 grammes = 649 grains English in an hour, and equal in light to 9·38 such candles, is
equivalent to 3·75 cubic feet of coal gas per hour. The sinumbra lamp, which consumes 50 grammes =
772 grains English, of oil per hour, and gives the light of 8 of the above candles, is equivalent to the
light emitted by 3·2 cubic feet of coal gas burning for an hour. A common argand lamp, equal to 4
candles, which consumes 30 grammes = 463 grains English per hour, is represented by 1·6 cubic feet of
gas burning during the same time. A common lamp, with a flat wick and glass chimney, whose light is
equal to 1·13 tallow candles, and which consumes 11 grammes = 169·8 grains English per hour, is
represented by 0·452 of a cubic foot of gas burning for the same time.
Construction of the Burners.—The mode of burning the gas as it issues from the jets has a great
influence upon the quantity and quality of its light. When carburetted hydrogen gas is transmitted
through ignited porcelain tubes, it is partially decomposed with a precipitation of some of its carbon,
while the resulting gas burns with a feebler flame. Coal gas, when kindled at a small orifice in a tube,
undergoes a like decomposition and precipitation. Its hydrogen, with a little of its carbon, burns
whenever it comes into contact with the atmospherical air, with a bluish coloured flame; but the
carbonaceous part not being so accendible, takes fire only when mixed with more air; therefore at a
greater distance from the beak, and with a white light from the vivid ignition of its solid particles. Upon
this principle pure hydrogen gas may be made to burn with a white instead of its usual blue flame, by
dusting into it particles of lamp black; or by kindling it at the extremity of a tube containing finely
pulverized zinc. The metallic particles become ignited, and impart their bright light to the pale blue
flame. Even platinum wire and asbestos, when placed in the flame of hydrogen gas, serve to whiten it.
Hence it has been concluded, that the intensity of light which a gas is capable of affording is
proportional to the quantity of solid particles which it contains, and can precipitate in the act of burning.
Carbonic oxide gas burns with the feeblest light next to hydrogen, because it deposits no carbon in the
act of burning. Phosphuretted hydrogen gives a brilliant light, because the phosphoric acid, into which
its base is converted during the combustion, is a solid substance, capable of being ignited in the flame.
Olefiant gas, as also the vapour of hydro-carbon oil, emits a more vivid light than common coal gas; for
the first is composed of two measures of hydrogen and two measures of the vapour of carbon
condensed into one volume; while the last contains only one measure of the vapour of carbon in the
same bulk, and combined with the same proportion of hydrogen. Olefiant gas may therefore be
expected to evolve a double quantity of carbon in its flame, which should emit a double light.
The illuminating power of the flame of coal gas is, on the contrary, impaired, when, by admixture with
other species of gas which precipitate no carbon, its own ignited particles are diffused over a greater
surface. This happens when it is mixed with hydrogen, carbonic oxide, carbonic acid, and nitrogen
gases, and the diminution of the light is proportional to the dilution of the coal gas.
In like manner the illuminating power of coal gas is impaired, when it is consumed too rapidly to allow
time for the separation and ignition of its carbonaceous matter; it burns, in this case, without
decomposition, and with a feeble blue flame. 1. This occurs when the light-gas is previously mixed with
atmospherical air, because the combustion is thereby accelerated throughout the interior of the flame,
so as to prevent the due separation of carbon. A large admixture of atmospherical air makes the flame
entirely blue. 2. When it issues, with considerable velocity, from a minute orifice, whereby the gas, by
expansion, gets intimately mixed with a large proportion of atmospherical air. If the jet be vertical, the
bottom part of the flame is blue, and the more so the less carbon is contained in the gas. The same
thing may be observed in the flame of tallow, wax, or oil lights. The burning wick acts the part of a
retort, in decomposing the fatty matter. From the lower part of the wick the gases and vapours of the
fat issue with the greatest velocity, and are most freely mixed with the air; while the gases disengaged
from the upper part of the wick compose the interior of the flame, and being momentarily protected
from the action of the atmosphere, acquire the proper high temperature for the deposition of carbon,
which is then diffused on the outer surface in an ignited state, and causes its characteristic white light.
Hence with coal gas, the light increases in a certain ratio with the size of the flame as it issues from a
larger orifice, because the intermixture of air becomes proportionately less. 3. If by any means too
great a draught be given to the flame, its light becomes feebler by the rapidity and completeness with
which the gas is burned, as when too tall a chimney is placed over an argand burner, see fig. 496. Fig.
497. c, is a view of the upper plate, upon which the glass chimney b rests. The gas issues through the
smaller openings of the inner ring, and forms a hollow cylindrical flame, upon the outside as well as the
inside of which the atmospherical air acts. The illuminating power of this flame may be diminished at
pleasure, according as more or less air is allowed to enter through the orifices beneath. With a very full
draught the light almost vanishes, leaving only a dull blue flame of great heating power, like that of the
blowpipe, corresponding to the perfect combustion of the gas without precipitation of its carbon. 4. On
the other hand, too small a draught of air is equally prejudicial; not merely because a portion of the
carbon thus escapes unconsumed in smoke, but also because the highest illuminating power of the
flame is obtained only when the precipitated charcoal is heated to whiteness, a circumstance which
requires a considerable draught of air. Hence the flame of dense oil gas, or of oil in a wick, burns with a
yellow light without a chimney; but when it is increased in intensity by a chimney draught, it burns with
a brilliant white flame.
From the consideration of the preceding facts, it is possible to give to coal gas its highest illuminating
power. The burners are either simple beaks perforated with a small round hole, or circles with a series
of holes to form an argand flame, as shown in fig. 497, or two holes drilled obliquely, to make the flame
cross, like a swallow’s tail, or with a slit constituting the sheet of flame called a bat’s wing, like most of
the lamps in the streets of London. These burners are mounted with a stop-cock for regulating the
quantity of gas.
The height of the flame, which with like pressure depends upon the size of the orifice, and with like
orifice upon the amount of pressure, the latter being modified by the stop-cock, is for simple jets in the
open air, as follows:—
 Length of the flame 2 3 4 5 6 inches
Intensity of the light 55·6 100 150 197·8 247·4
Volume of gas consumed 60·5 101·4 126·3 143·7 182·2
Light with equal consumption 100 109 131 150 150

When the length exceeds five inches, nothing is gained in respect to light. For oil gas the same
statements will serve, only on account of its superior richness in carbon, it does not bear so long a
flame without smoke. Thus:—

Length of the flame 1 2 3 4 5 inches

Intensity of the light 22 63·7 96·5 141 178
Gas consumed 33·1 78·5 90 118 153
Light with equal consumption 100 122 159 181 174

The diameter of the orifice for single jets, or for several jets from the same beak, is one twenty-eighth
of an inch for coal gas, and one forty-fifth for oil gas.

When several jets issue from the same burner, the light is improved by making all the flames unite into
one. In this case the heat becomes greater, for the combined flame presents a smaller surface to be
cooled, than the sum of the smaller flames. The advantage gained in this way, may be in the ratio of 3
to 2, or 50 per cent. In an argand burner, the distances of the orifices for coal gas should be from 16⁄100
to 18⁄100 of an inch, and for oil gas 12⁄100. If the argand ring has ten orifices, the diameter of the central
opening should be = 4⁄10 of an inch; if 25 orifices, it should be one inch for coal gas; but for oil gas with
10 orifices, the central opening should have a diameter of half an inch, and for 20 orifices, one inch.
The pin holes should be of equal size, otherwise the larger ones will cause smoke, as in an argand
flame with an uneven wick. The glass chimney is not necessary to promote the combustion of an
argand coal gas flame, but only to prevent it from flickering with the wind, and therefore it should be
made so wide as to exercise little or no influence upon the draught. A narrow chimney is necessary
merely to prevent smoke, when a very strong light, with a profusion of gas is desired. Oil gas burned in
an argand beak requires a draught chimney, like a common argand lamp, on account of the large
quantity of carbon to be consumed. The most suitable mode of regulating the degree of draught can be
determined only by experiment, and the best construction hitherto ascertained is that represented in
fig. 498. Fig. 499. exhibits the view from above, of the rim or ring c, upon which the chimney b stands,
and which surrounds the perforated beak. The ring is made of open fretwork, to permit the free
passage of air upwards to strike the outside of the flame. The thin annular disc d, which is laid over its
fellow disc c, in the bottom of the chimney-holder, being turned a little one way or other, will allow more
or less air to pass through for promoting more or less, the draught or ventilation. The draught in the
central tube of the burner may be regulated by the small disc e, whose diameter is somewhat smaller
than that of the ring of the burner, and which by turning the milled head f, of the screw, may be
adjusted with the greatest nicety, so as to admit a greater or smaller body of air into the centre of the
cylindrical flame.
In mounting gas-lights, and in estimating beforehand their illuminating effects, we must keep in mind
the optical proposition, that the quantity of light is inversely as the square of the distance from the
luminous body, and we must distribute the burners accordingly. When for example a gas-light placed at
a distance of ten feet, is required for reading or writing to afford the same light as a candle placed at a
distance of two feet; squaring each distance, we have 100 and 4; therefore 100 = 25, shows us that 25
4
such lights will be necessary at the distance of 10 feet.
Concerning portable gas-light, with the means of condensing it, and carrying it from the gas works to
the places where it is to be consumed, we need say nothing, as by the improvements lately made in the
purification and distribution of coal-gas, the former system has been superseded.
It is well known that light gas deteriorates very considerably by keeping, especially when exposed to
water over an extensive surface; but even to a certain degree over oil, or in close vessels. An oil-gas
which when newly prepared has the specific gravity of 1·054, will give the light of a candle for an hour,
by consuming 200 cubic inches; will, after two days, give the same light by consuming 215 cubic inches
per hour; and after four days, by consuming 240 cubic inches in the like time. With coal-gas the
deterioration appears to be more rapid. When newly prepared, if it affords the light of a candle with a
consumption of 400 cubic inches per hour; it will not give the same light after being kept two days,
except with a consumption of 430 inches; and after four days, of 460. Oil-gas three weeks old has
become so much impaired in quality that 600 inches of it were required per hour to furnish the light of a
candle. All light gas should be used therefore as soon as possible after it is properly purified.
Economical considerations.—The cost of gas-light depends upon so many local circumstances, that no
estimate of it can be made of general application; only a few leading points may be stated. The coals
required for heating the retorts used to constitute one half of the quantity required for charging the
retorts themselves. When five retorts are heated by one fire, the expenditure for fuel is only one third of
that when each retort has a fire. The coak which remains in the retorts constitutes about 60 per cent. of
the weight of the original coal; but the volume is increased by the coaking in the proportion of 100 to
75. When the coak is used for heating the retorts, about one half of the whole is required. If we
estimate the coak by its comparative heating power, it represents 65 per cent. of the coals consumed.
One hundred pounds of good coal yield in distillation 10 pounds of ammoniacal liquor, from which
sulphate or muriate of ammonia may be made, by saturation with sulphuric or muriatic acid, and
evaporation. The liquor contains likewise some cyanide of ammonia, which may be converted into
prussian blue by the addition of sulphate of iron, after saturation with muriatic acid.
Two hundred pounds of coal afford about 17 pounds of tar. This contains in 100 pounds 26 pounds of
coal oil, and 48 pounds of pitch. The tar is sometimes employed as a paint to preserve wood and walls
from the influence of moisture, but its disagreeable smell limits its use. The coal oil when rectified by
distillation, is extensively employed for dissolving caoutchouc in making the varnish of waterproof cloth,
and also for burning in a peculiar kind of lamps under the name of naphtha. Oil of turpentine however is
often sold and used for this purpose, by the same name. If the coal oil be mixed with its volume of
water, and the mixture be made to boil in a kettle, the mingled vapours when passed through a
perforated nozzle may be kindled, and employed as a powerful means of artificial heat. The water is not
decomposed, but it serves by its vapour to expand the bulk of the volatile oil, and to make it thereby
come into contact with a larger volume of atmospherical air, so as to burn without smoke, under a boiler
or any other vessel. The pitch may be decomposed into a light-gas.
The relative cost of light from coal gas and oil gas may be estimated as one to six, at least. Rosin gas is
cheaper than oil gas. See Rosin.
I shall conclude this article with a summary of the comparative expense of different modes of
illumination, and some statistical tables.
One pound of tallow will last 40 hours in six mould candles burned in succession, and costs 8d.; a gallon
of oil, capable of affording the light of 15 candles, for 40 hours costs 5s., being therefore 1⁄2 of the price
of mould candles, and 6⁄15 of the price of dips. The cost of wax is about 31⁄2 times that of tallow; and
coal gas, as sold at the rate of 9s. for 1000 cubic feet, will be one sixth the price of mould candles; for
500 cubic inches of coal gas give a light equal to the above candle for an hour; therefore 40 × 500 =
20,000 cubic inches = 11·57 cubic feet, worth 11⁄4d., which multiplied by 6 gives 71⁄2d. the average
price of mould candles per pound.
The author of the article Gas-light in the Encyclopædia Britannica, observes, in reference to the
economy of this mode of illumination, that while the price of coal, in consequence of the abundant and
regular supply of that article, is liable to little fluctuation, the cost of wax, tallow, and oil, on account of
the more precarious nature of the sources from which they are obtained, varies exceedingly in different
seasons. “Assuming that a pound of tallow candles, which last when burned in succession forty hours,
costs nine-pence,” (seven-pence halfpenny is the average price), “that a gallon of oil, yielding the light
of 600 candles for an hour, costs two shillings,” (five shillings is the lowest price of a gallon of such oil as
a gentleman would choose to burn in his lamp), “that the expense of the light from wax is three times
as great as from tallow, and that a thousand cubic feet of coal gas cost nine shillings;” he concludes the
relative cost to be for the same quantity of light,—from wax, 100; tallow, 25; oil, 5; and coal-gas, 3. I
conceive the estimate given above to be much nearer the truth; when referred to wax called 100, it
becomes, for tallow, 28·6; oil, 14·3; coal gas, 4·76.
Gas-lighting has received a marvellous development in London. In the year 1834, the number of gas
lamps in this city was 168,000, which consumed daily about 4,200,000 cubic feet of gas. For the
purpose of generating this gas, more than 200,000 chaldrons, or 10,800,000 cubic feet of coals were
required.
For the following valuable statistical details upon gas-light, my readers are indebted to Joseph Hedley,
Esq., engineer, of the Alliance Gas Works, Dublin; a gentleman who to a sound knowledge of chemistry,
joins such mechanical talent and indefatigable diligence, as qualify him to conduct with success, any
great undertaking committed to his care. He has long endeavoured to induce the directors of the
London gas-works to employ a better coal, and generate a more richly carburetted gas, which in much
smaller quantity would give as brilliant a light, without heating the apartments unpleasantly, as their
highly hydrogenated gas now does. Were his judicious views adopted, coal gas would soon supersede
oil, and even wax candles, for illuminating private mansions.
Copy of a paper laid before a Committee of the House of Commons, showing not only the relative
values of the Gases produced at the undermentioned places, but showing in like manner the relative
economy of Gas as produced at the different places, over candles. By Joseph Hedley, Esq.

Names of the Places Illuminating The Jet of Gas required Selling Cost of Average N
where Experiments power of a Gas burnt, to be equal price Gas equal discount G
were made. single Jet of four inches to 100 lbs. of Gas in illumi- allowed t
Gas-flame high, of mould per meter nating off the o
four inches consumed Candles, per 1000 power to charge
high, taken per hour 6 to the lb., cubic feet. 100 lbs. of for Gas.
by a and was 9 inches candles.[B]
comparison equal to the long each.
of Shadows. Candles [A]
in the last
column.
Equal to Cubic Cubic s. d. L. s. d. Per L
Candles. Feet. Feet. Cent.
 Birmingham;
Birmingham and
- 2·572 1·22 2704 10 0 1 7 0 9
Staffordshire;
two Companies
Stockport 3·254 ·85 1489 10 0 0 14 11 121⁄2
Manchester 3·060 ·825 1536 8 0 0 12 3 111⁄4
Liverpool Old
2·369 1·1 2646 10 0 1 6 5 61⁄4
Company[C]
Liverpool New
4·408 ·9 1164 10 0 0 11 8 61⁄4
Gas Company
Bradford 2·190 1·2 3123 9 0 1 8 1 121⁄2
Leeds 2·970 ·855 1644 8 0 0 13 2 61⁄4
Sheffield 2·434 1·04 2440 8 0 0 19 6 61⁄4
Leicester 2·435 1·1 2575 7 6 0 19 3 15
Nottingham 1·645 1·3 4200 9 0 1 17 9 15
Derby 1·937 1·2 3521 10 0 1 15 4 15
Preston 2·136 1·15 3069 10 0 1 10 8 15
none
London 2·083 1·13 3092 10 0 1 10 11
allowed.
[A] 100 lbs. of candles are estimated to burn 5700 hours.
[B] The candles cost 3l. 2s. 6d.
[C] The Liverpool Old Company have since resorted to the use of Cannel coal, and consequently very
the Liverpool New Company in illuminating power.

Memorandum.—It will not fail to be observed that in deducing the comparative value between candles and
gas by these experiments, the single jet (and in every instance, of course, it was the same), has
been the medium. This however, though decidedly the most correct way of making the comparative
estimate of the illuminating power of the several gases, is highly disadvantageous in the
economical comparison, inasmuch as gas burnt in a properly regulated argand burner, with its
proper sized glass, air aperture, and sufficient number of holes, gives an advantage in favour of gas
consumed in an argand, over a jet burner, of from 30 to 40 per cent. At the same time it must not
be overlooked that in many situations where great light is not required, it will be found far more
economical to adopt the use of single jets, which by means of swing brackets and light elegant
shades, becomes splendid substitutes for candles, in banking establishments, offices, libraries, &c.
&c.

Note.—In Glasgow, Edinburgh, Dundee, Perth, and the Scotch towns generally the Parrot or Scotch
Cannel coal is used; in illuminating power and specific gravity the gas produced is equal to that
from the best description of Cannel coal in England. The price per 1000 cubic feet ranges about 9s.,
with from 5 to 30 per cent. off for discounts, leaving the net price about 9s. to be equal in the
above table to 100 lbs. of candles.

Epitome of Experiments made in Gas produced from different qualities of Coal, and consumed in
different kinds of Burners:
Tried at the Sheffield Gas Light Company’s Works, and laid before a Committee of the House of
Commons. By Joseph Hedley, Esq.

Date Description Species of Specific Distance Gas Height Equal to Gas equa
1835. of Burner. Coal. Gravity of Candle consumed of Gas Mould Tallow to 100 lb
of Gas. from per Hour. Flame. Candles, 6 to of Mould
Shadow. the pound, Candles
 9 inches
long each.
May. Inches. Cubic Inches. Candles. Cubic
Feet. Feet.
8 Single Jet Deep Pit ·410 75 1· 4 2·36 2415
9 Ditto Mortormley ·450 74 ·95 4 2·434 2224
9 Ditto Cannel ·660 611⁄4 ·7 4 3·54 1127
Argand
8 - - Deep Pit ·410 34 3·3 31⁄2 11·53 1631
14 holes
9 Ditto Mortormley ·450 33 3·1 31⁄2 12·24 1443
9 Ditto Cannel ·660 29 2·6 31⁄3 15·85 935

Copy of Experiments made at the Alliance Gas Company’s Works in Dublin, during the past year 1837.
By Joseph Hedley, Esq.
Results of experiments on the qualities of various coals for the production of gas; its value in
illuminating power; produce of coke, and quality; and other particulars important in gas-making:—
1st Experiment, Saturday, May 27th, 1837.—Deane coal, (Cumberland). 2 cwt. of 112 lbs. each (or 224
lbs.) produced 970 cubic feet of gas; 4 bushels of coke of middling quality; specific gravity of the gas,
475. Consumed in a single-jet burner, flame 4 inches high, 14⁄10ths cubic feet per hour; distance from
shadow 76 inches or 2·3 mould candles. Average quantity of gas made from the charge (6 hours) 4·33
cubic feet per lb., or 9,700 cubic feet per ton of 20 cwt. Increase of coke over coal in measure, not quite
30 per cent. Loss in weight between coal, coke and breize 56 lbs., converted into gas, tar, ammonia, &c.
2nd Experiment, May 28th.—Carlisle coal, (Blenkinsopp). 224 lbs. produced 1010 cubic feet of gas, 4
bushels of coke of good quality though small; increase of coke over coal in measure not quite 30 per
cent. Loss in weight, same as foregoing experiment. Average quantity of gas made from the charge (6
hours) 4·5 cubic feet per lb. or 10,080 per ton.
Illuminating power of the Gas.

Consumed Distance Equal Specific

per hour, from to gravity.
single jet. candle. candles.
feet. inches.
At the end of the 1st hour 11⁄10 70 2·72 ·475
Ditto ditto with 20-hole
- 5 25 21·33 ·475
argand burner
When charge nearly off 14⁄10 85 1·84 ·442
When charge quite off, with
- 9 100 not 1 ·256
20-hole argand burner

3rd Experiment, May 29th.—Carlisle coal (Blenkinsopp). 112 lbs. produced 556 cubic feet of gas. Other
products, loss of weight, &c., same proportion as foregoing experiment. Average quantity of gas made
from the charge (6 hours) 4·96 cubic feet per lb., or 11,120 per ton.
In this experiment the quantity of gas generated every hour was ascertained; the illuminating power,
the specific gravity, and the quantity of gas consumed by the single jet with a flame 4 inches high, was
tried at the end of each hour, with the respective gases generated at each hour; and the following is a
table of results.
 RESULTS.

Hour. Gas Consumed Specific Distance Illuminating

produced. per hour gravity. of candle power
per single jet, from equal to
4 inches high. shadow. mould candles.
cubic cubic inches.
feet. feet.
111⁄2-10ths.
1st. 150 - - ·534 70 2·72
or 1·15
2nd. 120 11 ·495 75 2·36
3rd. 95 12 ·344 75 2·36
4th. 95 15 ·311 80 2·08
5th. 80 17 ·270 85 1·81
6th. 16 29 ·200 100 not one
Total 556 or 921⁄3 or 2 feet 9 inches.

Average of the above gas, 6-hour charge.

921⁄3 16-10ths. nearly ·359 81 2·03

Average of the above gas at 4-hour charge.

115 121⁄3-10ths. ·421 75 2·36

Production of gas in 6 hours 556 feet, or at the rate of 11,120 cubic feet per ton.
Ditto in 4 hours 460 feet, or at the rate of 9,200 ditto.
The relative value of these productions of gas is as follows, viz.:
11,120 at 16-10ths per hour nearly, (or 1·5916 accurately) and equal to 203 candles; the 11,120 feet
would be equal to and last as long as 1597 candles, or 2661⁄6 lbs. of candles.
9200 at 121⁄3-10ths. per hour, (or 1·2375 accurately,) and equal to 236 candles; the 9200 feet would be
equal to 1949 candles, or 3245⁄6 lbs. candles.

Now 2661⁄6 lbs. of mould candles, at 7s. 6d. per dozen lbs. will cost 8l. 6s. 41⁄2d., whilst
3245⁄6 lbs. of do. do. at 7s. 6d. per do. do. 10l. 3s.
Shewing the value of 4-hour charges, over 6-hour charges; and of 9,200 cubic feet over 11,120 cubic
feet.
Note.—9500 cubic feet of Wigan cannel coal gas are equal in illuminating power to 859 1-6th lbs. of candles, which at 7s. 6d.
per dozen lbs. will cost 25l. 10s. 51⁄2d. It is also found that any burner with superior gas, will consume only about half the
quantity it would do with common gas.

4th Experiment, May 30th.—Cannel and Cardiff coal mixed 1⁄2 and 1⁄2, together 112 lbs., produced 460
feet of gas; 2 bushels of coke of good quality; increase of coke over coal in measure about 30 per cent.;
loss in weight, 41 lbs.; coke weighed 71 lbs., no breize. Average quantity of gas made from the charge,
(4 hours) 4·1 cubic feet, per lb., or 9·200, per ton.
Illuminating power.—At end of first hour.

Candles. Cubic feet.

Distance of candle from Consumed per hour, single
- 73 or 2·49 - - 12-10ths
shadow jet, 4 inches high
At end of 2nd hour, do. 70 or 2·72 Do. do. do. 111⁄2-10ths
 At end of 3d hour. This gas very indifferent.
Average of the three 70 or 2·72 Do. do. do. 111⁄2-10ths

Specific gravity 3·44; 5 feet per hour, with a 20-hole argand burner, equal to 14·66 candles.
5th Experiment, May 31st.—Carlisle coal, 112 lbs. produced 410 feet of gas; other products, same as in
former experiments with this coal, but heat very low.
Illuminating power and produce of gas.

1st hour 120 cubic feet Average of this gas: specific gravity, 540; distance of candle from
2nd 100 shadow, 55 inches, or 4·4 candles consumed per single jet, 9-10ths
410 ft - -
3d 90 of a cubic foot per hour. 20-hole argand burner, 4 feet per hour,
4th 100 equal to 21·33 candles.

It is possible, from the superior quality of this gas, that a little of the cannel gas made for a particular
purpose, may have have got intermixed with it in the experimental gasholder and apparatus.
A variety of other experiments were tried on different qualities of coal, and mixtures of ditto, too tedious
to insert here, though extremely valuable, and all tending to shew the superior value of gas produced at
short over long charges; and also showing the importance and value of coal producing gas of the
highest illuminating power; among which the cannel coal procured in Lancashire, Yorkshire, and some
other counties of England and Wales, and the Parrot or splent coal of Scotland, stand pre-eminent.
Note.—In all the foregoing experiments the same single-jet burner was used; its flame in all instances exactly 4 inches high.
The coal when drawn from the retort was slaked with water, and after allowing some short time for drying, was weighed.

A Table of the number of hours Gas is burnt in each month, quarter and year.

Time of Burning. July. Aug. Sep. Oct. Nov. Dec. Jan. Feb. Mar. Apl. May. June. M
qu

o’clock.
From Dusk: to 6 — — 2 31 62 80 65 33 4 — — — —
— 7 — 14 22 62 92 111 96 61 31 4 — —
— 8 — 40 52 93 122 142 127 89 62 28 4 —
— 9 13 71 82 124 152 173 158 117 93 58 29 8
— 10 44 102 112 155 182 204 189 145 124 88 60 38 1
— 11 75 133 142 186 212 235 220 173 155 118 91 68 2
— 12 106 164 172 217 242 266 251 201 186 148 122 98 3
All night - 217 307 345 421 473 527 512 411 382 295 242 195 7
Morning from 4 — 16 48 80 110 137 137 98 71 28 2 —
— 5 — — 18 49 80 106 106 70 40 3 — —
— 6 — — — 18 50 75 75 42 9 — — — —
— 7 — — — — 20 44 44 14 — — — — —

Copy of a Paper submitted to a Committee of the House of Commons in the Session of 1837, being a
Synopsis of the proceedings of the undermentioned principal Gas-Light Establishments of England; and
procured by actual Survey and Experiments between the Years 1834 and 1837. By Joseph Hedley, Esq.

Name of the Price of Gas per Meter, and Price of Coal, and Average Coke made Selling Material
Place where Discounts allowed. Description; Quantity from a Ton Price of used to
 Gas Works delivered per Ton. of Gas of Coal. Coke. heat
are situated. made Retorts.
per Ton
of
Coals.

Cu. ft.

Birmingham 10s. per meter cub. feet. Lump coal from 6,500 32 bushels. 2s. 1d. Slack.
Gas Discounts West Bromwich pits per
Company. 10l. to 30l. per 21⁄2 per risen much of late. quarter
1837, 11s. 10d. delivered,
30l. to 50l. an. 5 cent.
or about
50l. to 75l. 7 1⁄2
3d. per
75l. to 100l. 10 bushel.
100l. & upwards 15
Birmingham 10s. per meter cub. feet. Discounts From West 6,500 24 bush. but 2s. 10d. Slack
and as above. Bromwich pits, larger per sack and Tar.
Staffordshire. 1837, 9s. 3d. measure of 8
than bushels.
Birmingham.
Macclesfield. 10s. per meter cub. feet. Common, 8s. 6,720 12 cwt. 10s. per Coke.
Discounts average 1831 ton.
A- 50l. and 75l. 5 per
bove 75l. not 100l. 7 1⁄2 cent.
100l. ex- 125l. 10
125l. ceed- 150l. 121⁄2
ing
150l. 175l. 15
175l. 200l. 171⁄2
200l. & upwards 20.
Stockport. 10s. per meter cub. feet. Discounts Coal 10s. 6d. cannel 7,800 7 cwt. 6s. 8d. Coal,
same as Macclesfield. Macclesfield 19s. 6d. about half per ton. coke,
discounts taken from Stockport card. and half used. and tar.
Average 15s. 1834.

Manchester. 10s. per m. cub. ft. 1834. 15s. 2d. average. 9,500 14 cwt. Ditto. Coke.
9s. and 8s. — 1837. Oldham
Discounts Watergate - cannel.
50l. and 100l. 21⁄2 per Wigan
100l. un- 150l. 5 cent. Mixed, 1834.
150l. der 200l. 71⁄2
200l. 225l. 10
225l. 250l. 121⁄2
250l. 300l. 15
300l. 400l. 171⁄2
400l. and upwards 20
Liverpool Old 10s. per meter cub. feet. 7s. 3d. per ton of 8,200 113⁄4 cwt. 8s. 4d. Slack
Company, Discounts 112 lbs. per cwt. per ton 7s. 3d.
1834. 10l. & under 50l. 21⁄2 per Ormskirk or Wigan of 112 lb. per ton.
50l. to 100l. 5 ct. slack. per cwt.
100l. to 200l. 71⁄2
300l. & upwards 10
Ditto ditto. In 1835 this Company resorted to the use of cannel coal similar to the Liverpool New Gas and Coal Company,
Liverpool 10s. per meter cub. feet. Discounts 18s. all cannel 9,500 13 cwt. 7s. 6d. Coke
New Gas and same as Liverpool Old Company. Wigan. per ton. and
Coke, 1835. slack.
Bradford, 9s. per meter cubic feet to large 8s. 6d. per ton. 3 8,000 13 cwt. 12s. per Coke.
1834. consumers. sorts used average. ton.
Discounts Slack 5s. 6d. Low
20l. to 30l. 5 per moor 8s. 10d.
30l. to 40l. 71⁄2 cent. Catherine slack 8s.
40l. to 60l. 10
60l. to 80l. 121⁄2
80l. to 100l. 15
100l. & upwards 20
Small consumers, 10s. per meter
cub. feet, and 5 per cent. off from
10l. to 20l.
Leeds, 1834. 8s. per meter cubic feet. 8s. per ton average. 6,500 12 cwt. 7s. 6d. Ditto.
Discounts 2-3ds common 7s. per ton.
2 1⁄2 15l. 1-3d cannel, 10s.
per cent.
5 on half- 30l.
- -
7 1⁄2 yearly 50l.
10 payments 100l.

Sheffield, 8s. per meter cubic feet. Discounts 7s. 9d. per ton 8,000 10 cwt. of 10s. per Ditto.
1835. same as Leeds. average. 3 sorts saleable ton.
used 1, 2-10ths. coke.
cannel, at 16s. 8, 2-
10ths. deep pit, 7s.
1-10th silk stone,
10s.

Leicester, 7s. 6d. per meter cub. ft. 13s. 6d. average. 7,500 4 quarters. 10s. 8d. Coke,
1837. Discounts on half-yearly rental not Derbyshire soft or 2s. 8d. tar, &c.
exceeding 10l., 5 per cent. coal. per qr.
A- 10l. and 20l. 71⁄2 per
bove 20l. not 30l. 10 cent.
30l. ex- 40l. 121⁄2
ceed-
40l. 50l. 15
ing
50l. 60l. 20
60l. & upwards 25
Derby, 1834. 10s. per meter cub. feet. Same coal used as 7,000 Ditto. Ditto. Coke.
at Leicester.
Discounts 5 to 35 per cent.

Nottingham, 9s. per meter cubic feet. Discounts Ditto. 7,000 Ditto. Ditto. Ditto.
1834. as above.

London, 10s. per meter cub. feet. No 17s. average. 8,500 36 bush. 12s. per Ditto.
1834. discounts. Newcastle. chaldron.

Ditto, 1837. Ditto. Ditto. 8,500 Ditto. Ditto. Ditto.

A Table shewing the Rate per Thousand Cubic feet received for any Burner consuming from 1⁄2 a Cubic
foot to 10 Cubic feet per hour, at any given price per annum, and to the times below stated. By Joseph
Hedley, Esq.

Time of Burning per annum. No. Single Jets. 2 Jets. 3 Jets.

of
Hours. Cub. Cub. Cub. Cub. Cub. Cub. Cub. Cub. C
[D]
ft. ft. ft. ft. ft. ft. ft. ft.
1⁄2 3⁄4 1 1 1⁄4 1 1⁄2 1 3⁄4 2 2 1⁄2
From Dusk to 8 o’clock 781 2·56 1·706 1·28 1·026 ·853 ·731 ·64 ·5132 ·4
ditto and Sundays 902 2·216 1·478 1·108 ·887 ·739 ·633 ·554 ·4434 ·3
ditto and from 6 o’clock mornings 1050 1·904 1·27 ·952 ·762 ·635 ·544 ·476 ·381 ·3
ditto and Sundays and from ditto 1172 1·706 1·138 ·853 ·682 ·569 ·487 ·426 ·3412 ·2
9 o’clock 1054 1·896 1·264 ·948 ·759 ·632 ·542 ·474 ·3794 ·3
ditto and Sundays 1221 1·638 1·092 ·819 ·675 ·546 ·463 ·409 ·3376 ·2
ditto and from 6 o’clock mornings 1323 1·510 1·066 ·755 ·604 ·503 ·431 ·378 ·3022 ·2
ditto and Sundays and from ditto 1490 1·342 ·894 ·671 ·536 ·447 ·383 ·335 ·2684 ·2
10 o’clock 1367 1·462 ·974 ·731 ·585 ·487 ·418 ·366 ·2926 ·2
ditto and Sundays 1586 1·26 ·84 ·63 ·504 ·42 ·36 ·315 ·2522 ·2
ditto and from 6 o’clock mornings 1636 1·222 ·814 ·611 ·489 ·407 ·349 ·305 ·2444 ·2
ditto and Sundays and from ditto 1855 1·078 ·718 ·539 ·431 ·359 ·308 ·269 ·2156 ·1
11 o’clock 1680 1·19 ·794 ·595 ·476 ·397 ·34 ·297 ·238 ·1
ditto and Sundays 1951 1·024 ·682 ·512 ·409 ·341 ·293 ·256 ·2048 ·1
ditto and from 6 o’clock mornings 1949 1·026 ·684 ·513 ·41 ·342 ·294 ·256 ·2052 ·1
ditto and Sundays and from ditto 2220 ·9 ·6 ·45 ·36 ·3 ·257 ·225 ·1802 ·1
12 o’clock 1993 1· ·668 ·502 ·4 ·334 ·287 ·251 ·2006 ·1
ditto and Sundays 2316 ·862 ·574 ·432 ·345 ·287 ·247 ·215 ·1726 ·1
ditto and from 6 o’clock mornings 2262 ·884 ·59 ·442 ·353 ·295 ·255 ·221 ·1768 ·1
ditto and Sundays and from ditto 2585 ·772 ·514 ·387 ·309 ·257 ·221 ·193 ·1546 ·1
1 o’clock 2306 ·866 ·578 ·434 ·347 ·289 ·247 ·217 ·1734 ·1
ditto and Sundays 2681 ·746 ·498 ·373 ·298 ·249 ·213 ·186 ·1492 ·1
ditto and from 6 o’clock mornings 2575 ·776 ·518 ·388 ·31 ·259 ·222 ·194 ·1552 ·1
ditto and Sundays and from ditto 2950 ·678 ·452 ·339 ·271 ·226 ·193 ·169 ·1356 ·1
All night 4327 ·462 ·308 ·231 ·185 ·154 ·132 ·115 ·6924 ·0

To use the Table.—Select the hour to which it is agreed the gas is to burn,—9, 10, 11 o’clock, Sundays, &c., as the case may
to it by the amount in shillings agreed to be paid per annum, and the product will be the sum received per m. cubic feet for t

Example.—Suppose a small argand which should burn 31⁄2 feet per hour, is agreed for till 9 o’clock at 2l. per annum. Look
and you find the number, ·271. Multiply this number by 40s. and the result gives 10s. 10d. per m. cubic feet. But suppose
mornings, and by enlarging the holes or height of flame consumes 8 cubic feet of gas per hour; then you have the number, ·
feet only, and so on for any greater or lesser variation of the agreement.

[D] The “number of hours” includes 1⁄4 of an hour allowed for shutting shops, and 1 hour’s extra burning

GENERAL SUMMARY.

For lighting London and its suburbs with gas, there are—
18 public gas works.
12 do. companies.
2,800,000l. capital employed in works, pipes, tanks, gas-holders, apparatus.
450,000l. yearly revenue derived.
180,000 tons of coals used in the year for making gas.
1,460,000,000 cubic feet of gas made in the year.
134,300 private burners supplied to about 40,000 consumers.
30,400 public or street do. N. B. about 2650 of these are in the city of London.
380 lamplighters employed.
176 gas-holders; several of them double ones, capable of storing 5,500,000 cubic feet.
890 tons of coals used in the retorts on the shortest day, in 24 hours.
7,120,000 cubic feet of gas used in longest night, say 24th December.
About 2500 persons are employed in the metropolis alone, in this branch of manufacture.
Between 1822 and 1827 the quantity nearly doubled itself, and that in 5 years.
Between 1827 and 1837 it doubled itself again.
Mr. Kirkham, engineer, obtained a patent, in June, 1837, for an improved mode of removing the
carbonaceous incrustation from the internal surfaces of gas retorts. He employs a jet or jets of heated
atmospheric air, or other gases containing oxygen, which he impels with force into the interior of such
retorts as have become incrusted in consequence of the decomposition of the coal. The retort is to be
kept thoroughly red hot during the application of the proposed jets. An iron pipe, constructed with
several flexible joints, leading from a blowing machine, is bent in such a way as to allow its nozzle end
to be introduced within the retort, and directed to any point of its surface.
I should suppose that air, even at common temperatures, applied to a retort ignited to the pitch of
making gas, would burn away the incrustations; but hot air will, no doubt, be more powerful.
GAS-HOLDER; a vessel for containing and preserving gas, of which various forms are
described by chemical writers.

GASOMETER, means properly a measurer of gas, though it is employed often to denote a

recipient of gas of any kind. See the article Gas-Light.

GAUZE WIRE CLOTH; is a textile fabric, either plane or tweelled, made of brass, iron, or
copper wire, of very various degrees of fineness and openness of texture. Its chief uses are
for sieves, and safety lamps.

GAY-LUSSITE, is a white mineral of a vitreous fracture, which crystallizes in oblique

rhomboidal prisms; specific gravity from 1·93 to 1·95; scratches gypsum, but is scratched
by calcspar; affords water by calcination; it consists of carbonic acid 28·66; soda, 20·44;
lime, 17·70; water, 32·20; clay, 1·00. It is in fact, by my analysis, a hydrated soda-
carbonate of lime in atomic proportions. This mineral occurs abundantly in insulated
crystals, disseminated through the bed of clay which covers the urao, or native
sesquicarbonate of soda, at Lagunilla in Colombia.

GELATINE; (Eng. and Fr.; Gallert, Leim, Germ.) is an animal product which is never found in
the humours, but it may be obtained by boiling with water the soft and solid parts; as the
muscles, the skin, the cartilages, bones, ligaments, tendons, and membranes. Isinglass
consists almost entirely of gelatine. This substance is very soluble in boiling water; the
solution forms a tremulous mass of jelly when it cools. Cold water has little action upon
gelatine. Alcohol and tannin (tannic acid, see Gall-nuts) precipitate gelatine from its
solution; the former by abstracting the water, the latter by combining with the substance
itself into an insoluble compound; of the nature of leather. No other acid, except the tannic,
and no alkali possesses the property of precipitating gelatine. But chlorine and certain salts
render its solution more or less turbid; as the nitrate and bi-chloride of mercury, the proto-
chloride of tin, and a few others. Sulphuric acid converts a solution of gelatine at a boiling
heat into sugar. See Ligneous Fibre. Gelatine consists of carbon, 47·88; hydrogen, 7·91;
oxygen, 27·21. See Glue and Isinglass.

GEMS, are precious stones, which, by their colour, limpidity, lustre, brilliant polish, purity,
and rarity, are sought after as objects of dress and decoration. They form the principal part
of the crown jewels of kings, not only from their beauty, but because they are supposed to
comprize the greatest value in the smallest bulk; for a diamond, no larger than a nut or an
acorn, may be the representative sign of the territorial value of a whole country, the
equivalent in commercial exchange of a hundred fortunes, acquired by severe toils and
privations.
Among these beautiful minerals mankind have agreed in forming a select class, to which the
title of gems or jewels has been appropriated; while the term precious stone is more
particularly given to substances which often occur under a more considerable volume than
fine stones ever do.
Diamonds, sapphires, emeralds, rubies, topazes, hyacinths, and chrysoberyls, are reckoned
the most valuable gems.
Crystalline quartz, pellucid opalescent or of various hues, amethyst, lapis lazuli, malachite,
jasper, agate, &c., are ranked in the much more numerous and inferior class of ornamental
stones. These distinctions are not founded upon any strict philosophical principle, but are
regulated by a conventional agreement, not very well defined; for it is impossible to subject
these creatures of fashion and taste to the rigid subdivisions of science. We have only to
consider the value currently attached to them, and take care not to confound two stones of
the same colour, but which may be very differently prized by the virtuoso.
Since it usually happens that the true gems are in a cut and polished state, or even set in
gold or silver, we are thereby unable to apply to them the criteria of mineralogical and
chemical science. The cutting of the stone has removed or masked its crystalline character,
and circumstances rarely permit the phenomena of double or single refraction to be
observed; while the test by the blowpipe is inadmissible. Hence the only scientific resources
that remain are the trial by electricity, which is often inconclusive; the degree of hardness, a
criterion requiring great experience in the person who employs it; and, lastly, the proof by
specific gravity, unquestionably one of the surest means of distinguishing the really fine
gems from ornamental stones of similar colour. This proof can be applied only to a stone
that is not set; but the richer gems are usually dismounted, when offered for sale.
This character of specific gravity may be applied by any person of common intelligence,
with the aid of a small hydrostatic balance. If, for example, a stone of a fine crimson-red
colour, be offered for sale, as an oriental ruby; the purchaser must ascertain if it be not a
Siberian tourmaline, or ruby spinel. Supposing its weight in air to be 100 grains, if he finds
it reduced to 69 grains, when weighed in water, he concludes that its bulk is equal to that of
31 grains of water, which is its loss of weight. Now, a real sapphire which weighs 100 grains
in air, would have weighed 76·6 in water; a spinel ruby of 100 grains would have weighed
72·2 in water, and a Siberian tourmaline of 100 grains would have weighed only 69 grains in
water. The quality of the stone in question is, therefore, determined beyond all dispute, and
the purchaser may be thus protected from fraud.
The sard of the English jewellers (Sardoine, French) is a stone of the nature of agate,
having an orange colour more or less deep, and passing by insensible shades into yellow,
reddish, and brown; whence it has been agreed to unite under this denomination all the
agates whose colour verges upon brown. It should be remarked, however, that the sard
presents, in its interior and in the middle of its ground, concentric zones, or small
nebulosities, which are not to be seen in the red cornelian, properly so called. The ancients
certainly knew our sard, since they have left us a great many of them engraved, but they
seem to have associated under the title sarda both the sardoine of the French, and our
cornelians and calcedonies. Pliny says that the sarda came from the neighbourhood of a city
of that name in Lydia, and from the environs of Babylon. Among the engraved sards which
exist in the collection of antiques in the Bibliothèque Royale of Paris, there is an Apollo
remarkable for its fine colour and great size. When the stone forms a part of the agate-
onyx, it is called sardonyx. For further details upon Gems, and the art of cutting and
engraving them, see Lapidary.

GEOGNOSY, means a knowledge of the structure of the earth; Geology, a description of the
same. The discussion of this subject does not come within the province of this Dictionary.

GERMAN SILVER. See the latter end of the article Copper.

GERMINATION; (Eng. and Fr.; Das Keimen, Germ.) is the first sprouting of a seed after it is
sown, or when, after steeping, it is spread upon the malt floor. See Beer.

GIG MACHINES, are rotatory drums, mounted with thistles or wire teeth for teazling cloth.
See Woollen Manufacture.

GILDING (Dorure, Fr.; Vergoldung, Germ.); is the art of coating surfaces with a thin film of
gold. For a full discussion of this subject, see Gold. Mr. Elkington, gilt toy maker, obtained a
patent, in June, 1836, for gilding copper, brass, &c., by means of potash or soda combined
with carbonic acid, and with a solution of gold. Dissolve, says he, 5 oz. troy of fine gold in
52 oz. avoirdupois of nitro-muriatic acid of the following proportions: viz. 21 oz. of pure
nitric acid, of spec. grav. 1·45, 17 oz. of pure muriatic acid, of spec. grav. 1·15; with 14 oz.
of distilled water.
The gold being put into the mixture of acids and water, they are to be heated in a glass or
other convenient vessel till the gold is dissolved; and it is usual to continue the application
of heat after this is effected, until a reddish or yellowish vapour ceases to rise.
The clear liquid is to be carefully poured off from any sediment which generally appears and
results from a small portion of silver, which is generally found in alloy with gold. The clear
liquid is to be placed in a suitable vessel of stone, pottery ware is preferred. Add to the
solution of gold 4 gallons of distilled water, and 20 pounds of bicarbonate of potash of the
best quality; let the whole boil moderately for 2 hours, the mixture will then be ready for
use.
The articles to be gilded having been first perfectly cleaned from scale or grease, they are
to be suspended on wires, conveniently for a workman to dip them in the liquid, which is
kept boiling. The time required for gilding any particular article will depend on
circumstances, partly on the quantity of gold remaining in the liquid, and partly on the size
and weight of the article; but a little practice will readily give sufficient guidance to the
workman.
Supposing the articles desired to be gilded be brass or copper buttons, or small articles for
gilt toys, or ornaments of dress, such as earrings or bracelets, a considerable number of
which may be strung on a hoop, or bended piece of copper or brass wire, and dipped into
the vessel containing the boiling liquid above described, and moved therein, the requisite
gilding will be generally obtained in from a few seconds to a minute; this is when the liquid
is in the condition above described, and depending on the quality of the gilding desired; but
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookname.com