Advanced Diagnostics in Combustion

Science Zhen-Yu Tian (Editor)

Visit to download the full and correct content document:
https://ebookmeta.com/product/advanced-diagnostics-in-combustion-science-zhen-yu
-tian-editor/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Network Analysis and Architecture 1st Edition Yu-Chu

Tian

https://ebookmeta.com/product/network-analysis-and-
architecture-1st-edition-yu-chu-tian/

Handbook of Real Time Computing Yu-Chu Tian (Editor)

https://ebookmeta.com/product/handbook-of-real-time-computing-yu-
chu-tian-editor/

Handbook of Real Time Computing Yu-Chu Tian (Editor)

https://ebookmeta.com/product/handbook-of-real-time-computing-yu-
chu-tian-editor-2/

Advanced Nanomaterials and Their Applications in

Renewable Energy 2nd Edition Tian-Hao Yan

https://ebookmeta.com/product/advanced-nanomaterials-and-their-
applications-in-renewable-energy-2nd-edition-tian-hao-yan/
Yu Gi Oh 3 in 1 Edition Vol 10 Includes Vols 28 29 30
Kazuki Takahashi

https://ebookmeta.com/product/yu-gi-oh-3-in-1-edition-
vol-10-includes-vols-28-29-30-kazuki-takahashi/

Artificial Intelligence for Materials Science Springer

Series in Materials Science 312 Yuan Cheng Editor Tian
Wang Editor Gang Zhang Editor

https://ebookmeta.com/product/artificial-intelligence-for-
materials-science-springer-series-in-materials-science-312-yuan-
cheng-editor-tian-wang-editor-gang-zhang-editor/

Active Flow and Combustion Control 2021: Papers

Contributed to the Conference “Active Flow and
Combustion Control 2021”, September 28–29, 2021, ...
Mechanics and Multidisciplinary Design, 152) 1st
Edition Rudibert King
https://ebookmeta.com/product/active-flow-and-combustion-
control-2021-papers-contributed-to-the-conference-active-flow-
and-combustion-control-2021-september-28-29-2021-mechanics-and-
multidisciplin/

Advanced Biosensors for Virus Detection - Smart

Diagnostics to Combat SARS-CoV-2 Edited By Raju Khan

https://ebookmeta.com/product/advanced-biosensors-for-virus-
detection-smart-diagnostics-to-combat-sars-cov-2-edited-by-raju-
khan/

Advanced Laser Process for Surface Enhancement Advanced

Topics in Science and Technology in China 61 Jianhua
Yao

https://ebookmeta.com/product/advanced-laser-process-for-surface-
enhancement-advanced-topics-in-science-and-technology-in-
china-61-jianhua-yao/
Editor
Zhen-Yu Tian

Advanced Diagnostics in Combustion

Science
Editor
Zhen-Yu Tian
Institute of Engineering Thermophysics Chinese Academy of Sciences,
Beijing, China
University of Chinese Academy of Sciences, Beijing, China

ISBN 978-981-99-0545-4 e-ISBN 978-981-99-0546-1

https://doi.org/10.1007/978-981-99-0546-1

Jointly published with Science Press

The print edition is not for sale in China (Mainland). Customers from
China (Mainland) please order the print book from: Science Press.

© Science Press 2023

This work is subject to copyright. All rights are solely and exclusively
licensed by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of reprinting, reuse of illustrations,
recitation, broadcasting, reproduction on microfilms or in any other
physical way, and transmission or information storage and retrieval,
electronic adaptation, computer software, or by similar or dissimilar
methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks,

service marks, etc. in this publication does not imply, even in the
absence of a specific statement, that such names are exempt from the
relevant protective laws and regulations and therefore free for general
use.

The publishers, the authors, and the editors are safe to assume that the
advice and information in this book are believed to be true and accurate
at the date of publication. Neither the publishers nor the authors or the
editors give a warranty, expressed or implied, with respect to the
material contained herein or for any errors or omissions that may have
been made. The publishers remain neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer

Nature Singapore Pte Ltd.
The registered company address is: 152 Beach Road, #21-01/04
Gateway East, Singapore 189721, Singapore
Abbreviations
ASA Active surface area
CAD Crank-angle degree
CaO Calcium oxide
CARS Coherent anti-stokes Raman spectroscopy
CCD Charged coupled devices
CFD Computational fluid dynamics
Cp Specific heat capacity
CPIV Conditioned particle image velocimetry
CRDS Cavity ring-down spectroscopy
CS Ceriporiopsis subvermispora
CSRS Coherent stokes Raman scattering
CW Continuous-wave
DB-CARS Dual-broadband rotational CARS
DC Direct current
DFWM Degenerate four-wave mixing
DHI Digital holographic interferometry
DI Direct injection
DM Dichroic mirror
DMC Demineralized bituminous coal
DSC Differential scanning calorimetry
DTA Differential thermal analysis
EC Elemental carbon
ECDL External-cavity diode laser
EDM Exchanged gas measurement
EGA Exchanged gas analysis
FAR-LIF Fuel/air ratio-laser-induced fluorescence
FID Flame ionization detector
FLBMR Fluidized-bed membrane reactor
FMS Frequency modulation spectroscopy
FQY Fluorescence quantum yield
fs Broadband
FTIR Fourier transform infrared spectrometry
GC Gas chromatography
GC/MS Gas chromatography/mass spectrometry
GOPs Gas potentiometric oxygen probes
GPCA Gas potentiometric combustion analysis
GPFA Gas potentiometric flame analysis
GPGA Gas potentiometric gasification analysis
GTL Gas-to-liquid
HCCI Homogeneous-charge compression ignition
HG Harmonic generation
HPIV Holographic particle image velocimetry
HPLC High performance liquid chromatography
HSQC Heteronuclear single-quantum correlation spectroscopy
HTA High temperature
HTO High temperature oxidation
Hz Hertz
ICTAC International Confederation for Thermal Analysis and
Calorimetry
ID Internal diameter
ISPD Intensifier silicon photodiode
JM Pyroelectric joulemeter
KD Distribution constant
KHz Kilohertz
LC/MS Liquid chromatography/mass spectrometry
LDA Laser Doppler anemometry
LDV Laser Doppler velocimetry
LE Lentinula edodes
LIF Laser-induced fluorescence
LLOQ Lower limit of quantification
LOD Limit of detection
LRS Linear Raman scattering
LTHR Low-temperature heat release
LTO Low-temperature oxidation
M&C Monitor and commander
m/z Mass-to-charge
MAC Mass-specific absorption cross-section
MAS Mobility aerosol spectrometer
MCA Multichannel analyzer
MCP Multichannel plate
MS Mass spectrometry
MSD Mass-selective detector
MTV Molecular tagging velocimetry
N.A Numerical aperture
ND Neutral density
NDIR Non-dispersive infrared gas analyzers
NMR Nuclear magnetic resonance
NO-LIF Nitric oxide-laser induced fluorescence
NOM Natural organic matter
OC Organic carbon
ONERA Office National d'Etudes et de Recherches Aerospatiales (in
French)
OPOs Optical parametric oscillators
PAHs Polycyclic aromatic hydrocarbons
PbTiO3 Lead titanate
PD Photo diode
PDA Planar Doppler anemometry
PDPA Phase Doppler particle analyzer
PDV Point Doppler velocimetry
PE Pleurotus eryngii
PID Proportional integral derivative
PIV Particle image velocimetry
PLC Programmable logic controller
PLIF Planar laser-induced fluorescence
PM Particulate matters
PMTs Photomultiplier tubes
ps Narrowband
PSTs Pressure-sensitive adhesive tapes
PT Piezoelectric
Pt Platinum
Py-GCMS Pyrolysis gas chromatography mass spectrometry
Q-branch Quantum number J
R Rayleigh signal
R&D Research and development
RAM Random Access Memory
RS Raman scattering
RSA Reactive surface area
Rt Retention time
RTD Resistance temperature detectors
SDT Simultaneous differential techniques
SE Solid electrolyte
SEM Scanning electron microscope
SIDMS Speciated isotope dilution mass spectrometry
SIMS Speciated isotope mass spectrometry
SNR Signal-to-noise ratio
SRS Spontaneous Raman Scattering
TA
Thermal analysis
Tc Crystallization temperature
TDLAS Tunable diode laser as a light source
TE Thermoelectrometry
TEA Thermoelectrical analysis
TEM Transmission electron microscopy
Tf Final temperature
TG Gas temperature
Tg Glass transition temperature
TG Thermogravimetry
TGA Thermogravimetry analysis
Ti Initial temperature
TK Transient kinetics
Tm Melting temperature
TM Thermomechanometry
TMA Thermomechanical analysis
TNF Turbulent non-premixed flames
TOA Thermoptometric analysis
TPB Triple phase boundary
TPD Temperature-programmed desorption
TZ Sensor temperature
Ueq Nernst equation
UV Ultraviolet
VOCs Volatile organic compounds
VOP High temperature
VUV Vacuum ultraviolet
WMS Wavelength modulation spectroscopy
YSZ Stabilized zirconia
ZrO2 Zirconium oxide
ω1 Pump beam
ω2 Stoke beam
ω3 CARS Signals
% wt. Weight per cent
Contents
1 Introduction to Advanced Diagnostic Techniques in Combustion
Science
Zhen-Yu Tian, Vestince Balidi Mbayachi, Maria Khalil and
Daniel A. Ayejoto
2 Gas Chromatography/​Mass Spectrometry
Zhen-Yu Tian, Vestince Balidi Mbayachi, Wei-Kang Dai, Maria Khalil
and Daniel A. Ayejoto
3 Thermal Analysis Methods
Zhen-Yu Tian, Vestince Balidi Mbayachi, Wei-Kang Dai, Maria Khalil
and Daniel A. Ayejoto
4 Gas Potentiometry Diagnostics in High-Temperature
Environments
Vestince Balidi Mbayachi, Zhen-Yu Tian, Zhi-Min Wang, Maria Khalil
and Daniel A. Ayejoto
5 Raman Scattering Diagnostics
Vestince Balidi Mbayachi, Zhen-Yu Tian, Zhi-Min Wang, Maria Khalil
and Daniel A. Ayejoto
6 Coherent Anti-Stokes Raman Scattering
Zhen-Yu Tian, Vestince Balidi Mbayachi, Xu Zhang, Maria Khalil and
Daniel A. Ayejoto
7 Laser-Induced Fluorescence in Combustion Research
Vestince Balidi Mbayachi, Zhen-Yu Tian, Xu Zhang, Maria Khalil and
Daniel A. Ayejoto
8 Nuclear Magnetic Resonance
Vestince Balidi Mbayachi, Zhen-Yu Tian, Wei-Kang Dai,
Daniel A. Ayejoto, Zhi-Min Wang, Xu Zhang and Maria Khalil
Contributors
Daniel A. Ayejoto
Guangzhou Institute of Energy Conversion, Chinese Academy of
Sciences, Guangzhou, China
University of Chinese Academy of Sciences, Beijing, China

Wei-Kang Dai
Institute of Engineering Thermophysics, Chinese Academy of Sciences,
Beijing, China

Maria Khalil
Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences,
Shanghai, China
University of Chinese Academy of Sciences, Beijing, China

Vestince Balidi Mbayachi

Dalian Institute of Chemical Physics, Chinese Academy of Sciences,
Dalian, China
University of Chinese Academy of Sciences, Beijing, China

Zhen-Yu Tian
Institute of Engineering Thermophysics, Chinese Academy of Sciences,
Beijing, China
University of Chinese Academy of Sciences, Beijing, China

Zhi-Min Wang
Institute of Engineering Thermophysics, Chinese Academy of Sciences,
Beijing, China

Xu Zhang
School of Energy Power and Mechanical Engineering, North China
Electric Power University, Beijing, China
© Science Press 2023
Z.-Y. Tian (ed.), Advanced Diagnostics in Combustion Science
https://doi.org/10.1007/978-981-99-0546-1_1

1. Introduction to Advanced Diagnostic

Techniques in Combustion Science
Zhen-Yu Tian1, 2 , Vestince Balidi Mbayachi2, 3, Maria Khalil2, 4 and
Daniel A. Ayejoto2, 5
(1) Institute of Engineering Thermophysics, Chinese Academy of
Sciences, Beijing, 100190, China
(2) University of Chinese Academy of Sciences, Beijing, 100049, China
(3) Dalian Institute of Chemical Physics, Chinese Academy of Sciences,
Dalian, 116023, China
(4) Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, Shanghai, 200032, China
(5) Guangzhou Institute of Energy Conversion, Chinese Academy of
Sciences, Guangzhou, 510640, China

Zhen-Yu Tian
Email: tianzhenyu@iet.cn

1.1 Introduction
During the past few decades, immense progress and development have
been made to combustion research. These diagnostic techniques
significantly impact combustion understanding, such as fewer
environmental hazards and more device improvement capable of
further development of processes.
Nowadays, this field has proliferated such that numerous purposes
include fundamental research academics, pollutant control, reliable
operation process, and production on a commercial level. This chapter
will consist of the details of resolute techniques, uses, and applications
that can be scientific and technical. As a result, essential properties of
these techniques with examples and general discussions will be
discussed. Hence, a generalized form of information about diagnostic
techniques within combustion science is provided. According to the
relative quantitative phenomenon of the diagnostic approaches
outlined, it is mentioned that suitable measuring techniques depend on
the properties of the combustion process to investigate. Under the
general aspect, several commonly used diagnostic techniques are
described. These techniques are divided into two subclasses based on
mechanical probing and optical diagnostics. The discussion of each
technique possesses a short detail of physics, method application on a
laboratory and technical scales.
In the other portion, imaging and multidimensional diagnostics, the
history of working of laser diagnostics in combustion diagnostics
research is made prominent. Eventually, the probability of laser
diagnostics corroborating those perspectives in numerical simulations
is discussed. The outline is dressed with the accessible quantity
equipment with the detailed information of other applications of
diagnostic techniques. With the substantial development of
spectroscopic methods for combustion diagnostics, the laser is
regarded as starting point of research. Before the onset of laser
technology, emission and absorption spectroscopy served well in
combustion diagnostics.
Some important intermediate species are listed in flames as OH, CH,
HCO, NH, or C2. Their presence is proven by using spectroscopic
methods. Besides, laser-based spectroscopy offers innovative, widely
refined ways to observe combustion. When we employ the spectral
bandwidth of laser light, it becomes easy to discern various spectral
properties of atoms and molecules at a higher resolution. The Prior
method is carried out without the coherent light source contrary ability
to produce laser pulses made possible the analysis of transient
processes. Depending upon the unique properties of prominently well-
defined laser beam formation, point-like sharp geometric shapes, and
planes like conventional light sources propagation direction of laser
light can be controlled with precision. As a result, the accuracy of
irradiated volumes can be acquired. Altogether, the above-mentioned
unique properties can make laser diagnostics the most widely used
technique in applied combustion research and investigation control
combustion systems.

1.2 Main Works of Advanced Diagnostics

Main works of advanced diagnostics revolve around pressure,
temperature, velocity, concentration and their distribution with space
and time, flame peak position and propagation speed, flame structure,
and reaction flow field display, particle size distribution, solid fuel
combustion surface, and subsurface status, their chemical structure, the
temperature distribution of emissions, component concentration, and
signal characteristics, etc. It is not easy to understand the tasks and
needs of combustion diagnostics without any prior knowledge of
combustion properties. A brief description of such properties should
help to explain how specific requirements are imposed on diagnostic
techniques. The desired temporal and spatial resolution of a
measurement depends on the system's governing time and length
scales to be characterized. An estimate of these scales for a small
collection of different combustion applications is shown in Table 1.1.

Table 1.1 Estimated combustion application time scale and length scale [1]

Application Time scale (ms) Length scale (m)

Flame measurement—laminar 105–106 10–4–10–2
Flame measurement—turbulent 10–2–102 10–5–10–2
Fire research 102–103 10–5–10–2
Jet engine; compressor inlet 103–104 10–4–10–1
Gas turbine burner 10–1–101 10–5–10–4
After burner 10–1–100 10–5–10–4

Combustion often occurs in thin, sheet-like regions (flames) as a

spatially extremely inhomogeneous process. Since, in most natural
combustion systems, the thickness of these layers is on the order of
some ten to a hundred micrometeres, in terms that govern the
dynamics of a combustion system, those species that exist in tiny
amounts [down to mass fractions in the parts-per-million (ppm) scale]
may exert a more significant influence on the systems the behavior than
those which are abundantly present, in the per cent range. The ability
to detect and measure these key species (e.g., OH, CH in Fig. 1.1)
without interference from the much more abundant bath of hundreds
or thousands of other species represents a significant requirement for
the successful experimental combustion analysis.

Fig. 1.1 a Structure of a premixed methane/air flame at 20 bars (simulation). The

spatial extent of the scene shown is 0.1 mm; b Sketch of the temporal and spatial
scales relevant to combustion [1]
The potentially high sensitivity of combustion conmtcerning the
variations in physical boundary conditions can render prohibitive use
of measurement techniques that significantly alter the physical
conditions during an experiment. A diagnostic approach should not
change the system that it is attempting to measure; indeed, a significant
indicator for the quality of a diagnostic technique in this respect would
be its level of non-intrusiveness. Recognized diagnostic techniques can
be quite sharply separated, based on their level of intrusion, with most
optical techniques considered non-intrusive and those involving
mechanical probes highly intrusive.
Finally, significant constraints exist when a diagnostic technique is
used within a technical combustion environment [2]. Unlike laboratory
systems, most realistic environments are not explicitly designed to
allow or alleviate diagnostic procedures to be carried out. Hence,
diagnostic techniques must often be applied under conditions that are
not optimally suited to the underlying measuring principles. An
essential property of a diagnostic technique is, therefore, its robustness
—that is, its ability to function reasonably well under conditions that
may deviate strongly from those for Fig. 1.2. The temperature structure
of a turbulent flame is taken by 2D Rayleigh scattering (false-color
representation, top right), which is hidden in flame photography
(bottom left) [3].

Fig. 1.2 The temperature structure of a turbulent flame is taken by 2D Rayleigh

scattering (false-color representation, top right), which is hidden in flame
photography (bottom left) [3]
Robustness also implies that the instruments used for diagnostics
can operate under the adverse conditions (e.g., high temperatures and
pressures, presence of highly reactive substances) encountered in
combustions. The ideal diagnostic technique must be sufficiently robust
enough to survive hostile environments and provide a large amount of
combustion-related information.

1.3 Classification
Classification of the diagnostic techniques is majorly based on sampling
methods, spectroscopic procedures, velocity, concentration, and
particle size. Various instruments are included; for example,
thermocouples are used to measure temperature, and LIF is applied to
study flame structure. Table 1.2 summarizes the classification of
diagnostic techniques according to their test principles, experiments,
methods, information collection, and detection means.

Table 1.2 A summary of the classification of diagnostic techniques

Classification Names of methods or instruments Examples

Sampling Thermocouple temperature probe Non- Temperature
shielded
Suction type
Pneumatic dual sonic hole temperature probe
Pneumatic speed probe Pitot tube
with two
holes
Five-hole
Pitot trust
Gas sampling probe Metal or
quartz probe
Molecular
beam probe
Two-phase sampling probe Sample
collection
Classification Names of methods or instruments Examples
Spectroscopic Optical Radiometer Spectral line Temperature
method emission and inversion
adsorption
Radiation
absorption
Color
temperature
Spectroradiometer Absolute
strength
Relative
intensity
Atomic resonance absorption Concentration
spectrometry
Infrared Near- T/C
absorption infrared
Mid-infrared
MS Mass spectrometry Concentration
Electron Auger electron spectroscopy Surface
energy composition
spectrometry X-ray photoelectron spectroscopy

Particle size Microscope Optical Particle

damping and
Ion/electron SEM/ TEM/
nozzle two-
HIM
phase flow loss
Malvine ion analyzer
Single particle counter
Classification Names of methods or instruments Examples
Velocity Single point LDV 3D
PDPA Particle size
Imaging PIV ½D
PDV
MTV
HPIV 3D
Stable Pressurized burning rate meter Solid rocket
propellant
Closed bomb or burner
Acoustic emission
Unstable Microwave
Ultrasonic
Absorption
Real-time
screen
Concentration Spectroscopic LIF Flame structure
CRDS
Sampling Capillary
Molecular beam

1.4 Invasive Techniques

These techniques are based on traditional combustion diagnostic
approaches where a mechanical examination is performed in research.
This area can be either a measurement system, a device to measure the
system, or a sample analysis system. These approaches are used to
measure the system under experimental conditions. Firstly, the
unwanted and unavoidable property of the invasive probe is the local
interaction with the system, which is further analyzed. Secondly,
systematic errors must be handled carefully if the measured quantity is
rated. These errors are relatively caused by the mechanical probes
depending upon the size of the probe and the combustion field under
study.

1.4.1 Temperature
Combustion can often lead to flame, so the temperature is the most
critical variable for technical application to be measured in combustion
chambers. It accounts for downscaled flame temperature
measurements. This combustion temperature needs to be carefully
controlled. In this regard, immense development is made to analyze
temperature distributions. There are several other advanced devices
used to measure the temperature.

1.4.1.1 Thermocouples
The most widely used device for temperature measurement is a
thermocouple. Thermocouples consist of two dissimilar metals joined
together at one end, which produces a small unique voltage for a given
temperature, and this voltage is measured and interpreted by a
thermocouple thermometer. Here, the Seebeck effect is an intrinsic
principle, which describes the material-dependent electrical voltage
when a temperature difference forms a closed loop. Two materials
provide standardized voltage-temperature rather than applicable
temperature ranges [4]. Some details of the material used as the pair of
thermocouples and their application ranges can be used in the
combustion system, as illustrated in Table 1.3.

Table 1.3 Pairs of thermocouple materials and their application ranges [4]

Type Positive material Negative Sensibility at Range of

material 20 °C (µV/°C) temperature
(°C)
E Chromel (nickel Constantan 58.7 −270 to 1,000
10% chromium) (nickel 45%
copper)
G Tungsten Tungsten 26% 19.7 (600 °C) 0 to 2,320
Rhenium
C Tungsten 5% Tungsten 26% 19.7 (600 °C) 0 to 2,320
rhenium rhenium
Type Positive material Negative Sensibility at Range of
material 20 °C (µV/°C) temperature
(°C)
D Tungsten 3% Tungsten 26% 19.7 (600 °C) 0 to 2,320
rhenium rhenium
J Iron Constantan 50.4 −210 to 760
(nickel 45%
copper)
K Chromel (nickel Alumel (nickel 39.4 −270 to 1,372
10% chromium) 5% aluminum and
silicon)
N (AWG Nicrosil (84.3% Ni, Nisil (95.5% Ni, 39 −270 to 400
14) 14% Cr, 1.4% Si, 4.4% Si, 0.1% Mg)
0.1% Mg)
N (AWG Nicrosil (84.3% Ni, Nisil (95.5% Ni, 26.2 0 to 1,300
28) 14% Cr, 1.4% Si, 4.4% Si, 0.1% Mg)
0.1% Mg)
B Platinum 6% Platinum 30% 1.2 0 to 1,820
Rhodium Rhodium
R Platinum 13% Platinum 5.8 −50 to 1,768
Rhodium
S Platinum 10% Platinum 5.9 −50 to 1,768
Rhodium
T Copper Constantan 38.7 −270 to 400
The main drawback of the device is related to the probe, which
should be placed into the measuring volume, and it can affect the flame
with various mechanisms. These drawbacks are accepted mostly for
applications based on sapid evaluation and assessment. Based on the
dimensions of the probe and flow condition, the local flow field in the
probe can be influenced, and fuel mixing and oxidizer can also be
affected [5]. Moreover, heat conduction and radiation from surrounding
walls can affect the heat balance of flame. These specific materials on
sensors act as chemical catalysts. This must be differentiated in
contrast to the thermocouple reaction material by using chemically
active material to increase the decomposition of stable contrary
ceramic, which primarily acts as an insulator in flames at high-
temperature material. Heitor and Moreira [6] have provided a brief
overview of the limitations and successful applications of these sensors
in the field of combustion diagnostics.

1.4.1.2 Resistance Thermometry

Resistance thermometry is another invasive technique that is used to
measure temperature. Devices are also called resistance temperature
detectors (RTD), shortly named sensors to measure temperature. RTD
mainly consists of a wire wrapped around a ceramic. Along with this,
another construction is also used. The fine wire wrapped around
should be a pure material, e.g., copper, nickel, and platinum [7]. Even
though the perfect linearity of resistance-based thermometers and the
availability of standardized types of Pt100 have, a major drawback is its
limited application range of up to about 1300 K and the long response
times required, compared to the thermocouples. Their accuracy is far
more than thermocouples [8].

1.4.1.3 Thermochromic Painting

In the past years and recent years, thermochromic painting has been
famous in temperature estimation upon heat change. The fundamental
principle behind this color formation phenomenon is the change in the
heat, which directly changes the color of components. This change is
the molecular structure and crystalline property of the material. There
is a recent development in this field. This temperature-sensitive
phenomenon is further developed in temperature-sensitive paints,
which are based on the sensitivity of luminescent molecule's thermal
environment following (ultraviolet) excitation [9]. An increase in
temperature of the luminescent molecule increases the probability of
the molecule to return to the ground state via no excitation process
[10], rather than emitting a photon. This thermally induced
phosphorescence quenching forms the basis of the temperature
measurement via a quantitative detection of emission attenuation.

1.4.2 Species Concentrations

Before, after, and during the combustion system, concentration
measurement is essential for investigation basics and process control
applications. Under the measurement principle, there are probing
strategies applied in general, due in part as well to regulate and
normalize the historical methods for sampling, for example, constant
volume sampling has been used for many decades to support the
testing of vehicle emissions, whereby a constant total flow rate of a
vehicle exhaust plus dilution air is maintained.
For this to occur, as the exhaust flow increases, the dilution air must
be automatically decreased [11]. In this case, a bag measurement of
emissions represents the key method for legislative purposes, as it
provides a single value for the major exhaust species of a complete test
cycle, including the different phases of acceleration and constant speed.
Although the sampling may be conducted continuously using a suitable
ejector or probing device (e.g., Pitot tube or low-pressure sampling),
followed by additional dilution stages, the quantitative analysis of the
species concentration can be performed batch-wise in the case of some
diagnostic techniques. An example is the gas-chromatographic
measurement of species in engine exhaust gases. However, this is also
often applied to analyze basic properties such as heating value and the
composition of gaseous hydrocarbon fuels for combustion in stationary
gas turbines.

1.4.2.1 Flame Ionization Detector

Flame ionization detector (FID) is an invasive technique commonly
used for gas analysis of hydrocarbons at different detection ranges [12].
This technique is mass sensitive, and the concentration of gases
downstream in a coiled column is less affected by the flow of carrier
gas.
The working principle of FIDs, as shown in Fig. 1.3, involves the
production of ions when hydrocarbon gas is heated in hydrogen flames
inside the column. Two electrodes with different potentials are used to
detect these ions. The negative electrode is located above the flame, and
its plate attracts the ions, which induces a current proportional to the
ionization rate of the hydrocarbon concentration in the sample.
Fig. 1.3 Schematic of flame ionization detector [13]
FIDs are applied in portable gas chromatographs, and their
advantages include a comprehensive measurement of hydrocarbon
gases and positive results in the presence of moisture. The limitations
of FIDs involve a poor response to hydrocarbons with high numbers of
halogens, destruction of the sample, and flame failure to ignite the gas
in high humidity.

1.4.2.2 Gas Chromatography/Mass Spectrometry

Gas chromatography (GC) combined with mass spectrometry (MS) is an
analytical, diagnostic technique that is applied in the identification and
quantification of small molecules of less than 650 Daltons [14]. GC-MS
possesses superior separation capability, reproductivity, and sensitivity,
thus achieving a higher chromatographic resolution than LC-MS [15].
GC-MS unit consists of sample preparation, data acquisition, data
processing, and quality control (Fig. 1.4). GC-MS has advantages such as
fast analysis, higher accuracy, inexpensiveness, and efficiency, even
though it can only detect volatile materials with lower molecular
weight. Gustavsson et al. [16] analyzed flame retardants using GC-MS.
Axford et al. [17] studied the effects of applied electric fields and
electric potential variation in a flame using MS to sample electrically
charged species from flames at atmospheric pressure.
Fig. 1.4 Gas chromatography/mass spectrometry

1.4.3 Pressure Sensor

The pressure sensor in a combustion chamber plays a vital role in
security and stability, thus providing essential information on thermal
stability. In-cylinder pressure sensors (Fig. 1.5) are installed into
automotive engines to increase power energy, knock control and check
for leak status [18]. These sensors consist of a piezoelectric element
built of lead titanate (PT, PbTiO3).
Fig. 1.5 A spark plug integrated with an in-cylinder pressure sensor [18]
The properties of the in-cylinder pressure sensors majorly depend
on the type of piezoelectric element incorporated. Lead titanate is a
crystal structure of perovskite, which provides the advantage of a high
piezoelectric constant, high ferroelectric Curie temperature of 490 °C
and a higher signal-to-noise ratio. Besides the application of in-cylinder
pressure sensors in automotive, they are also applied in the detection
and measurement of pressure dynamics in the harsh combustion
environment of premixed gas turbines where the fluctuation of the heat
release in the burning chamber results in extreme combustion noise
with a pressure range of hundreds of millibars (Fig. 1.6).
Fig. 1.6 Glow plug integrated with in-cylinder pressure sensor for diesel engine [18]
Currently, low-cost in-cylinder pressure sensors such as fiber-optic
combustion pressure sensors with 500 million pressure lifecycles have
been introduced and applied in engine diagnostics [19]. They are
incorporated with a spark plug, glow plug, or fuel injector at a low cost
in engines to control combustion, improve efficiency, and minimize
pollution.

1.4.4 Particulates
Soot particles are formed due to incomplete combustion of
hydrocarbons such as methane (CH4) and methane-nitrogen (CH4–N2)
mixture within an inverted laminar diffusion flame [20]. Mass-specific
absorption cross-section (MAC) of soot is assumed to be dependent on
its mass and size particle. Transmission electron microscopy (TEM) is
employed to characterize soot particle diameter and morphology.
Raman spectroscopy is used to investigate the graphitization level of
soot particles. Combustion of hydrocarbons with a diffusion flame
under atmospheric pressure leads to thermophoretic deposition inside
the cold walls of the flow field [21].
In Fig. 1.7, a probe is introduced into a diffusion flame for enough
exposure time to capture a remarkable sample but short enough to
avail a cold surface to the flame borne particle. The cold surface freezes
the already captured particle’s heterogeneous reaction, which prevents
the change in morphology of the particle after contact with a cold
surface.

Fig. 1.7 Diffusion flame of the thermophoretic probe sampling [21]

1.5 Noninvasive Techniques

The development and introduction of noninvasive methods in
combustion diagnostic were influenced by the urge to overcome the
disadvantages of invasive diagnostic tools. The problem can be
illustrated by comparing point-wise photographic images that show
concentration measurements using either laser or mechanical probes
in a laminar premixed flame [21].
In the mechanical probe, there is a disturbance of the flame shape,
as seen in Fig. 1.8, resulting in dislocation of the flame front. The flame
shape is unaffected in the laser probe since it shows the exact
symmetry as if no probe was applied. Another comparison was
conducted using a continuous-wave (cw) laser, Raman probe, and Q-
switched ruby laser to measure temperature and concentration of a
laminar premixed methane/air flame [23]. The deviation in oxygen
concentration was taken by a quartz suction probe and plotted in two
downstream positions in the reaction zone (as shown in Fig. 1.9).

Fig. 1.8 The Concentration measurement in a laminar flame. a The mechanical

probe causes disturbances of the reaction zone; b No influence of the noninvasive
laser probe can be observed [22]
Fig. 1.9 Radial O2 concentration profiles at a downstream position: a within the
reaction zone and b in the post-reaction zone. Comparison between the mechanical
probe and laser Raman measurements at a downstream position of 1d (a) and 6.5 d
(b) [23]
It was confirmed that the difference between noninvasive and
invasive techniques was due to a worse local resolution of the
mechanical probe, which tends to change the reaction zone, impacting
both less deep gradient and squeezed reaction at the outer part of the
flame [23].
Several studies recently focused on the non-intrusive laser as a
diagnostic technique in combustion. Figure 1.10 illustrates the
principle of optical measurement technique where various processes
are compared in terms of signal intensity by the magnitude of the cross-
sectional interaction. When a laser beam with frequency v0 is irradiated
into a gas sample, irradiated photon interacts with matter present in
the probe, giving rise to physical processes that can be distinguished by
laser beam attenuation by extinction processes such as absorption and
scattering or by the emission of radiation with a different frequency to
that of various approaches. Emission of intensities of various processes
depends on the magnitude of the interaction cross-section, thus
providing direct information on the strength of the signal achieved. The
unique qualities of the weakest Raman and Rayleigh processes have
become the best measurement tools in different combustion
diagnostics CO2 (%).

Fig. 1.10 Schematic illustration of different interaction processes between the

matter present in the sample volume and the irradiated laser beam [22]

1.5.1 Visualization Techniques

Flow visualization is a modern diagnostic technique that has
significantly contributed to the combustion application. Long [24]
introduced the study of refractive index fields in aerodynamics and
suggested a Schlieren technique in combustion application.
Holographic interferometry allows the investigation of non-intrusive
temperature measurements in optical flame with temporal resolution
and high spatial (Fig. 1.11). These new measurement and imaging
techniques use Digital holographic interferometry (DHI) to record a
hologram on CCD electronically and reconstruct it by numerical
methods [25].
Fig. 1.11 A simple digital holography system in combustion study [25]

1.5.1.1 Schlieren Techniques

Schlieren imaging systems are the oldest diagnostic technique used
since the nineteenth century to visualize fluctuations in optical density
leading to refractive index variation [26]. Schlieren optics provide non-
intrusive information in 2D transparent and optical media. The benefit
of this technique in fluid dynamics studies is that it does not interfere
with the flow despite being sensitive to changes.
Schlieren imaging works under the principle of Snell’s Law, which
states that light slows with the interaction of matter (Fig. 1.12). Light
travels steadily and at a constant velocity in homogenous media such as
space and vacuum. Light also refracts and deflects when it encounters
inhomogeneous media in terms of temperature, density, pressure, and
chemical composition resulting in Schliere. Recently, the invention of a
more powerful computing system and digital cameras has led to the
development of Schlieren optics in the study of refractive media.
Fig. 1.12 Experimental setup of Schlieren system [26]
Schlieren techniques have been reported in many combustion
applications with laser-induced fluorescence to study the structure of
premixed flame of propane and air as a function of flame stabilization
processes and the difference in the fuel–air ratio [27]. In addition, these
techniques have been applied to observe how pressure influences the
structural turbulent of premixed Bunsen flames [28]. Figure 1.13
illustrates the observation of premixed flame at different atmospheric
pressure. It can be seen that at 0.1MPa, the flame is wrinkled with low
intensity and large-scale turbulence, but with the increase in pressure,
the bluntly wrinkle flame becomes fine spikes making it difficult to
differentiate each wrinkle to see its depth.
Fig. 1.13 Instantaneous Schlieren photographs of the turbulent flames: a 0.1 MPa; b
0.25 MPa; c 0.5 MPa; d 1.0 MPa [22]

1.5.2 Emission Spectroscopy

The analysis of spectra radiation emitted by the flame is one of the most
optical diagnostic techniques in combustion. Most emission
spectroscopy in the gas phase relies on the chemiluminescence of
molecules rather than the excitation of atoms from one energy band to
the other. Many kinetic reactions are temperature dependent in
hydrocarbon combustion resulting in detectable intermediate radicals
such as OH, CH, NH, and CN (Fig. 1.14) [29].

Fig. 1.14 Emission spectra of an ammonia-doped, fuel-rich premixed butane–air

flame [22]

In chemical and thermal excitation, molecules such as CO2 and H2O

are the products of hydrocarbon flames. Emission spectroscopy for
years has been used for the detection of flame radicals. Previously,
emission spectroscopy was used to measure radical concentration in
low-pressure flames, but currently, it is advanced in the industrial
application as an active feedback control parameter for equivalent ratio
disturbance of high-pressure premixed flames [30]. Combustion
reaction in the solid phase leads to components such as soot particles
that emit black body (Planck) radiation that can be used in industrial
applications for temperature measurement. According to Planck’s
radiation law, temperatures can be calculated effectively by directly
measuring radiation intensity within a band of the emitted black body
emission spectrum. Distribution pyrometry is always applied in two
almost adjacent bands where radiation intensity is measured. Recently,
coherent anti-Stokes Raman spectroscopy (CARS) has been used as 1D
emission spectroscopy in temperature measurement of sooting
premixed flame [31].

1.5.3 Absorption Spectroscopy

Absorption spectroscopy is an advanced diagnostic technique used to
quantitatively measure transparent media species in the industrial
analytical system. It operates under the principle of a molecule's partial
absorption of light radiation. If the emission spectrum overlaps the
absorption spectrum of the probed molecule, the transmitted intensity
can be calculated by the Lambert–Beer law. Quantum mechanics
calculate the absorption spectra by describing the molecule's
vibrational, rotational, and electronic energy levels. The introduction of
laser radiation has facilitated a new application in temperature
measurements based on two-line absorption techniques and sensitivity
and a strong increased signal-to-noise ratio (SNR) [32].
NDIR analyzer is a broadband excitation used for noninvasive
measurement of exhaust gas analysis such as CO2 and CO. The
limitation of the broadband technique is that other species can be
absorbed in the IR region. Due to the low cost of tunable diode laser as
a light source (TDLAS), it has recently been used in online monitoring
and combustion diagnostics [33]. Mokhov et al. (2005) used TDLAS to
measure acetylene in atmospheric pressure premixed methane/air
fuel-rich flames. Tunable diode laser and frequency modulation
spectroscopy (FMS) or wavelength modulation spectroscopy (WMS)
lead to a significant increase of signal-to-noise in the species under
investigation.
The Diode laser system can probe most combustion species such as
O2, CH4, CO, NO, H2O, CO2, and NO2. Kranendonk et al. [34], in their
experiment, used a tunable external-cavity diode laser (ECDL) as an
absorption technique to scan from 1374 to 1472 nm for crank angle
resolved gas temperature and H2O mole fraction in an engine operating
on heptane/air in the ratio of 0.16 in homogeneous-charge
compression ignition (HCCI) mode. This approach offers a strategy for
quasi-simultaneous multispecies analysis as a new diagnostic
technique for a wide range of applications in combustion.
Figure 1.15 shows the pressure change per crank angle degree
(CAD). The pressure change curve reveals a two-stage combustion
process occurring before TDC. H2O absorption data recorded in the
region between the vertical dotted lines was reduced to gas
temperature and water mole fraction.

Fig. 1.15 Pressure in the combustion chamber as a function of crank-angle degree

after top dead center [34]

1.5.4 Laser-Induced Fluorescence (LIF)

Spectroscopy
Laser-induced fluorescence (LIF) is a two-step process of absorption-
emission in which a laser photon is absorbed with the emission of a
fluorescence photon from the excited state [35]. Because of the
resonant character of the excitation, with its typically large cross-
section, minor species at very low concentration levels such as parts
per million (ppm) can also be evaluated. It is possible to detect flame
radicals and pollutant species with LIF in the in-situ detection. This is
done by using single-photon absorption processes (for example, CH,
OH, NO, or formaldehyde) or by tuning the exciting laser to a
wavelength where the species absorb two photons at the same time
(two-photon LIF) (e.g., OH, N, O, CO). During the LIF process, a laser is
usually used to excite a fluorescent species within the flow. Specifically,
the tracer is an organic fluorescent dye such as fluorescein or
rhodamine. Some portion of the excitation energy is absorbed by the
dye and simultaneously re-emits some of the absorbed energy as
fluorescence. The emitted fluorescence is then evaluated and used to
determine the local concentration of the dye.
In addition to this, LIF can also be used in measuring scalar
concentrations at a point along a line in 2D planes [36], and in 3D
volumes [37], although LIF is widely used in fluid flows for 2D planar
laser-induced fluorescence (PLIF). There are many different strategies
for studying combustion phenomena by employing LIF. When
fundamental chemistry mechanisms are investigated, laminar systems
are used, and quantitative concentration or temperature measurements
are executed. This is commonly achieved by a point-wise scanning of
the object of investigation step wisely.
When complex flow phenomena are studied, PLIF is usually
employed using a light sheet technique to obtain two-dimensional (2D)
information about the mixture or flame structure simultaneously and
instantaneously. The fundamentals, potential applications, and working
principles of LIF for combustion research will be discussed later in this
book.

1.5.5 Scattering Techniques

Generally, laser scattering techniques consist of frequency-unshifted
scattering, namely, Mie and Rayleigh scattering and frequency-shifted
Raman scattering. It is worthy to note that the achievable signal
intensities of these scattering processes vary widely in magnitude so
that the instantaneous application of these techniques may be
complicated or even unfeasible. This is particularly correct for Mie and
Rayleigh scattering, which have the same emission frequency.
1.5.5.1 Mie Scattering
Mie scattering is the elastic scattering of light from atomic and
molecular particles whose diameter is greater than the wavelength of
the incident light. When the dipole size is greater, the quasi-static
evaluation will be inadequate to estimate the solution. Therefore, there
is a need to work out the boundary value problem, i.e., the full-wave
theory or Mie theory [38, 39]. With the help of this theory, the
scattering cross-section does not grow continually with frequency.
Considering a sphere of radius, the scattering cross-section has a high-
frequency boundary of πa2.
Mie scattering being the most robust scattering process, results
from an interaction between phase boundaries and irradiated light. It
can be utilized to know the location at which different phases occur,
such as liquid-phase droplets or gas-phase bubbles in liquids or solid-
phase particles in a gaseous state, which could still be a combustion
environment. For example, Fig. 1.16 shows the typical mushroom-like
complex of a diesel jet in the primary break-up region a few
microseconds after leaving the injection nozzle (start of injection). It
presents the liquid intact core being surrounded by a two-phase flow,
and the gas phase is achieved by cavitation inside the nozzle [40]. It can
be concluded that a laser sheet has been aimed into the center with the
diesel jet of a mini sac-hole nozzle, using a long-distance microscope for
signal detection. The jet velocity can then be calculated by recording
two frames with a defined temporal delay.
Fig. 1.16 Diesel jet primary spray break-up a few microseconds after the start of the
injection [41]
Mie scattering is commonly employed for a more macroscopic
characterization of the direct injection of gasoline and diesel fuel, either
directly or indirectly in the injection chambers [41]. In the case of
particle Mie scattering, the scattering object size should equal the
magnitude of the incident light wavelength. The intensity is a function
of the particle diameter (dp), N is the number of particle density which
is related to the direction of the incident laser beam. Considering the
experimental study of an appropriately particle seeded flow, where the
mean particle diameter and the observation angle are in good
approximation constant, the intensity is relative to the number of
particles, and to the seeded gas component concentration that can be
utilized for imaging the flow of the gas concentration.
To obtain Mie scattering, scattering particles are needed in the flow
to be measured. In case no natural scattering particle is involved,
artificial particles are used in the flow, which is referred to as seeding.
Chen and Roquemore [42] developed a potential seeding technique for
investigating flames, in which propane–air-jet diffusion flames were
analyzed with respect to the zone of reaction and the flame structure.
 Another similar approach is directly seeding reactant gases with
TiO2 or oil droplets as a flame-front marker. This is majorly applicable
in the flamelet regime of premixed flames, where the thickness of the
reaction zone is only 100 mm. For seeded oil, the flame front is shown
by Mie scattering disappearance at the entrance of the reaction zone as
the oil disappears. A direct seeding of TiO2, which also occurs
downstream of the reaction zone, helps to measure velocities flow in
both the burned and unburned regions of the flame. In this case, the
transition from the reactant to the product zone can be attained from
the different seeding densities that result from the temperature-related
gas density difference, and this can be used for a conditioned evaluation
of the turbulence characteristics [43].
Mie scattering has also been considered to create the physical basis
of laser Doppler anemometry (LDA). The Doppler principles are
majorly based on the optical Doppler Effect that results from the light
scattering of a moving object. The particle diameter can be calculated
from the observable phase shift of the Doppler signals detected by the
various sensors; this technique is known as phase Doppler
anemometry (PDA) [44]. Recently, PDA has been used in spray flames
to measure the mean droplet diameter in a gas-supported, swirl
stabilized kerosene flame to determine the location of the high burning
fraction of the droplets and the temperature measurements. In another
example, the influence of different electrical field strengths and
geometries on the atomization process of electrostatically charged
hydrocarbon fuel sprays has been investigated [45].
It is worth noting that it is possible to employ both 2D material
Raman and Mie scattering to determine the droplet diameter. For
example, as a 2D measurement technique, Malarski et al. [46] have
recommended imaging metrology to measure droplet diameters by
simultaneously applying Mie scattering and 2D Raman. In this method,
the ratio between the volume-dependent Raman signal and the surface
area-proportional Mie signal is utilized to generate frozen images of the
droplet diameter.

1.5.5.2 Rayleigh Scattering

The radiation scattered with no change of wavenumber is known as
Rayleigh scattering when the scattering centers are very much smaller
than the wavelength of the incident radiation. A simple solution for this
scattering can explain the physical phenomena in nature. For example,
what makes the sky have the color blue? The sunset is so excellently
beautiful how birds and insects can move without the help of a
compass.
In Rayleigh scattering, the molecule is excited to a higher virtual
energetic state, immediately falling back to a fundamental molecular
energy level. This is because the incident photon's energy is not
adequate enough to excite higher quantum levels of the probed
molecule. If the final level is indistinguishable from the initial level,
then no energy exchange has occurred between the molecule and
photon. It is crucial to conduct an independent scattering of the
molecules in order to obtain a similar frequency for the emitted photon
and the incident laser.
It is usually impossible to detect selectively one individual species in
a combustion system by Rayleigh scattering, as the signal will have
followed all species in the gaseous mixture. Moreover, differences
between the Rayleigh signal from scattered laser light (e.g., Mie
scattering from particles or droplets) and the stray light of the
environment are not possible unless additional measures are employed
to block any undesired light, maybe by using a molecular filter to
absorb the stray laser light and the narrow-band Mie scattering
(filtered Rayleigh scattering) [47].
The strength I of the scattered light from one individual species is
proportional to its number density c, multiplied by a species-specific,
approximately temperature-independent Rayleigh scattering cross-
section [48]:

where the index i runs over all n components in the mixture. For an
ideal gas mixture, the number density ci can be expressed by the molar
fraction (xi) of the species, pressure (p), and temperature (T) as ci = xi
p⁄RT such that . By applying this formula,
temperatures can be obtained from Rayleigh signals, provided that
information about the pressure and the mixture-averaged Rayleigh
cross-section is available. Although, this requires that the
mixture-averaged Rayleigh cross-section does not vary significantly
with the chemical composition, nor can it explicitly be estimated or
determined [49].
Another approach for measuring temperature using Rayleigh
scattering is to study the effect of the temperature-dependent Doppler
broadening of Rayleigh light [50]. Recent planar Rayleigh scattering has
been using a light sheet technique to obtain information on the
temperature inside a 2D and 3D plane crossing the combustion field. In
a 2D approach, two different wavelengths are irradiated simultaneously
to form two adjacent parallel planes, with the signals being collected on
two separate CCD cameras. Rayleigh experiments in crossed-plane
configurations have also been performed, usually in combination with
additional diagnostic approaches for obtaining 3D information along a
line [51].
Rayleigh scattering has also been used on sooting flames by
employing the filtered Rayleigh technique in practical systems like
combustors or to investigate fuel vapor distribution in a heavy-duty
direct injection (DI) diesel engine [52]. It has been applied to detect
polarized and depolarized components of the Rayleigh signal in
turbulent non-premixed flames, hence allowing measurement of the 2D
distribution of fuel concentrations [53]. This method was coupled with
LIF of OH and CO to obtain information on four different quantities
(mixture fraction, temperature, scalar dissipation rate, and the rate of
the reaction CO + OH → CO2 + H) in turbulent partially premixed flames.
It is worthy to note that Rayleigh scattering can also be used
together with other techniques to get broad information on the
combustion process and validation of numerical simulations. This has
been accomplished for instants by the simultaneous use of OH-LIF and
planar Rayleigh, or planar Rayleigh with PIV, collecting at the same
temperature, time, and flow velocity field information and allowing the
direct determination of the turbulent flux [54].

1.5.5.3 Raman Scattering (RS)

Linear Raman Scattering (LRS)
Raman scattering is an inelastic scattering in which the incident
photons interact with molecules so that energy is either gained or lost,
and the scattered photons are shifted in frequency. Raman scattering
depends on the polarizability of the molecules.
For polarized molecules, the incident photon energy can excite
vibrational modes of the molecules, resulting in scattered photons with
lower energy associated with the number of vibrational transition
energies. Spectral analysis of the scattered light under this condition
will reveal spectral lines under the peak of the Rayleigh scattering at
the incident frequency. These lines are referred to as “Stokes lines.”
When the vibrational state of the scattering molecules is excited, then it
is also feasible to notice scattering at frequencies exceeding the
incident frequency as the vibrational energy is attached to the incident
photon energy. These weaker lines are known as anti-Stokes lines.
Simultaneous excitation of all molecules in the sample volume
without any restriction of the excitation frequency is one of the critical
advantages of Raman scattering. Molecules in combustion have at least
one Raman-active molecular energy transition. It is possible to measure
a direct relative concentration by simultaneously collecting the Raman
signals of all species in the probe volume if the relative Raman cross-
sections of all species are known. Thus, high-frequency radiation is
essential for Raman excitation, and Raman experiments with laser
excitation at short wavelengths. In most combustion systems, the
existence of species that fluoresce when exposed to UV radiation put a
limit to the shortest practically feasible excitation wavelength in Raman
experiments. Current knowledge suggests that the KrF-excimer
excitation at 248nm is the shortest wavelength yet used for Raman
experiments in combustion [55].
Figure 1.17 shows several approaches at which temperature can be
determined from Raman scattering using temperature-dependent
intensity distributions in the pure vibrational or rotational bands [57].
Information regarding the concentration can be obtained from the
vibrational branches' integrated line intensities or single rotational
lines. Since absolute measurements are challenging to carry out,
calibration at room temperature will simplify the procedure.
Fig. 1.17 Schematic of the temperature dependence of the Raman spectrum in the
pure rotational and rotational–vibrational bands. This example shows oxygen excited
by a ruby laser [56]
Generally, the Raman technique has been used in the early stage for
concentration and temperature measurement in combustion [58].
However, since then, linear Raman scattering has been applied widely
to flame studies [59], with both 1D [60] and 2D [61] spatial resolution
having been achieved. Despite the difficulties associated with the
extremely weak signals, Raman measurements have also been
performed in practical systems, such as in engines [62].
Polarization-Resolved Linear Raman Scattering
It is almost impossible to apply linear Raman scattering in sooting
flames and combustion systems because of the hindrances caused by
chemiluminescence emissions, hydrocarbon fuel fluorescence
interferences, particle Mie scattering, laser light reflection, or stray
light. Therefore, the Raman effect's polarization property is used to
obtain unhindered Raman signals under these conditions [63]. The
polarization method of Raman scattering involves suppressing the
depolarized emissions from the highly polarized Raman signals by
collecting separately the horizontally and vertically polarized signal
components, followed by subtraction of the horizontal from the vertical
Another random document with
no related content on Scribd:
against the inadequacy of his force for the defence required of it, and
insisted on a temporary reinforcement of militia to garrison the
fortress that had cost him so much effort to construct at the Maumee
Rapids.
Then the value of General Proctor to his enemy became
immense. Between January 22, when he attacked Winchester, and
the end of April, when he moved on Fort Meigs, Proctor molested in
no way the weak and isolated American garrisons. With hundreds of
scouts and backwoodsmen at his command, he had not the energy
or the knowledge to profit by his opponents’ exposed and
defenceless condition. He allowed Major Wood to make Fort Meigs
capable of standing a siege; he let Harrison, unmolested, pass a
month away from his command; he looked on while the Virginia
militia marched home, leaving only a handful of sickly men, under a
major of artillery, to defend the unfinished fort; he made no attempt
to waylay Harrison, who returned with reinforcements by way of the
Auglaize River; and not until Harrison had enjoyed all the time
necessary to prepare for attack, did Proctor disturb him.
Harrison, expecting an assault, hurried back from Cincinnati to
Fort Meigs with some three hundred men, leaving a brigade of
Kentucky militia to follow him. April 12 he reached the fort, but not till
April 28 did Proctor appear at the mouth of the Maumee, with about
five hundred regulars and nearly as many militia,—nine hundred and
eighty-three whites, all told, and twelve hundred Indians under
95
Tecumthe and other chiefs. Besides this large force, he brought
two twenty-four pound guns with other artillery from Detroit, and two
gunboats supported the land-battery. While the guns were placed in
position on the north bank of the river, the Indians crossed and
surrounded the fort on the south. May 1 the batteries opened, and
during four days kept up a heavy fire. Proctor, like Harrison, moved
in the wilderness as though he were conducting a campaign on the
Rhine; he liked regular modes of warfare, and with a force almost
wholly irregular, after allowing Fort Meigs to be built, he besieged it
as though he could take it by battering its earthen ramparts.
Untaught by his losses at the river Raisin, he gave once more
advantage to the Kentucky rifle; and with every opportunity of
destroying the reinforcement which he knew to be near, he allowed
himself to be surprised by it.
The Kentucky brigade of twelve hundred men, under Brigadier-
General Green Clay, had descended the Auglaize River in boats,
and arrived at Defiance May 3, where they learned that Fort Meigs
was invested. So neglectful of his advantages was Proctor that he
not only failed to prevent General Clay from advancing, but failed to
prevent communication between the besieged fort and the relief-
column, so that Harrison was able to arrange a general attack on the
investing lines, and came near driving the British force back to
Malden with the loss of all its artillery and baggage. At about nine
o’clock on the morning of May 5, Clay’s brigade descended the
rapids, and eight hundred and sixty-six men under Colonel William
96
Dudley, landing on the north side of the river, surprised and took
possession of the British batteries, which were entirely unsupported.
Had Clay’s whole force been on the ground, and had it been
vigorously pushed forward, the small British division which held the
north bank must have abandoned all its positions; but Dudley’s men
were under no discipline, and though ready to advance were in no
hurry to retreat, even when ordered. Three companies of the British
Forty-first, and some of the Canadian militia soon gathered together;
and although these could hardly have been half the number of
97
Dudley’s force, yet with Tecumthe and a body of Indians they
attacked the batteries, drove the Kentuckians out, dispersed them,
and either captured or massacred the whole body, under the eyes of
Harrison and Fort Meigs.
This affair, though little less fatal to the Americans than that of
the river Raisin, was much less dearly bought by the British. Five
hundred prisoners fell into Proctor’s hands; two or three hundred
more of the Kentucky brigade, including “the weak and obstinate but
98
brave” Dudley himself, must have been either killed in battle or
99
massacred after surrender; only one hundred and seventy
escaped; the boats with the baggage were captured; while the whole
British loss on the north side of the river hardly exceeded fifty killed
and wounded. A bitter feeling against Proctor was caused by the
massacre of some forty American prisoners while under a British
guard, and also, as was alleged, under the eyes of General Proctor,
who did not interpose, although a soldier of the Forty-first was
murdered in trying to protect them. Probably all the prisoners would
have been massacred had Tecumthe not ridden up at full speed,
tomahawk in hand, and threatened to kill the first Indian who defied
100
his authority.
On the south side Harrison had better fortune, and Colonel John
Miller of the Nineteenth U. S. Infantry by a sortie gallantly captured a
battery, with some forty prisoners; but neither on the north nor on the
south did the fighting of May 5 decide any immediate military result.
Besides losing on the north bank half the reinforcement brought by
General Green Clay, Harrison had lost in the siege and in the sorties
on the south bank nearly three hundred men in killed and
101
wounded. If the numbers loosely reported in the American
accounts were correct, the siege cost Harrison one thousand men,
or fully half his entire force, including his reinforcements. After the
fighting of May 5, he withdrew once more into the fort; the British
batteries reopened fire, and the siege went on. No further attempt
was made to trouble the enemy in open field. Harrison felt himself
too weak for further ventures; yet never had his chance of a great
success been so fair.
Proctor’s siege of Fort Meigs was already a failure. Not only had
the fort proved stronger than he expected, but the weather was bad;
his troops were without shelter; dysentery and loss in battle rapidly
weakened them; half his militia went home, and, what was fatal to
further action, his Indians could not be held together. Within three
days after the battle of May 5, the twelve hundred Indians collected
by Tecumthe’s influence and exertions in the northwest territory
dispersed, leaving only Tecumthe himself and a score of other
102
warriors in the British camp. Proctor had no choice but to retire as
rapidly as possible, and May 9 embarked his artillery and left his
encampment without interference from Harrison, who looked on as a
spectator while the movement was effected.
 From that time until the middle of July Proctor remained quiet.
Harrison moved his headquarters to Upper Sandusky and to
Cleveland, and began to prepare for advance under cover of a naval
force; but he was not allowed to rest, even though Proctor might
have preferred repose. Proctor’s position was difficult. Told by Sir
103
George Prevost that he must capture what supplies he needed
from the Americans, and must seek them at Erie and Cleveland,
since Lower Canada could spare neither food nor transport, he was
compelled to look for support to the American magazines. He was
issuing ten thousand rations a day to the Indian families at Malden,
104
and his resources were near an end. Leaving Malden with either
three hundred and ninety-one regulars, or about five hundred
regulars and militia, and by one British account nearly a thousand
105
Indians, by another between three and four thousand, Proctor
returned by water to the Maumee Rapids July 20, and tried to draw
the garrison of Fort Meigs into an ambush. The attempt failed.
General Green Clay, who was in command, had learned caution,
and imposed it on his troops. Proctor then found that his Indians
were leaving him and returning to Detroit and Amherstburg. To
occupy them, Proctor took again to his boats and coasted the Lake
shore as far as the Sandusky River, while the Indians who chose to
accompany him made their way by land. August 1 the expedition
effected a landing at the mouth of the Sandusky, and scattered panic
into the heart of Ohio.
In truth, nothing could be more alarming than this movement,
which threatened Harrison in all directions,—from Fort Meigs, on the
Maumee, to Erie, or Presqu’isle, where Perry’s fleet was building. On
Sandusky River Harrison had collected his chief magazines. All the
supplies for his army were lying at Upper Sandusky, some thirty
miles above the British landing-place, and he had only eight hundred
106
raw recruits to defend their unfortified position. Nothing but an
untenable stockade, called Fort Stephenson, on the Sandusky River,
where the town of Fremont afterward grew, offered an obstacle to
the enemy in ascending; and Tecumthe with two thousand Indians
was said to be moving from Fort Meigs by the direct road straight for
the magazines, thus flanking Fort Stephenson and every
intermediate position on the Sandusky.
In just panic for the safety of his magazines, the only result of a
year’s campaigning, Harrison’s first thought was to evacuate Fort
Stephenson in order to protect Upper Sandusky. The flank-attack
from two thousand Indians, who never showed themselves, impelled
him to retire before Proctor, and to leave the river open. July 29, after
a council of war, he sent down a hasty order to young Major Croghan
who commanded Fort Stephenson, directing him immediately to burn
the fort and retreat up the river or along the Lake shore, as he best
107
could, with the utmost haste. Croghan, a Kentuckian, and an
officer of the Seventeenth U. S. regiment, refused to obey. “We have
determined to maintain this place, and by Heaven, we will,” he wrote
108
back. Harrison sent Colonel Wells, of the same regiment, to
relieve him; but Croghan went to headquarters, and by somewhat
lame excuses carried his point, and resumed his command the next
day. Harrison gave him only conditional orders to abandon the fort,—
orders which Croghan clearly could not regard, and which Harrison
109
seemed to feel no confidence in his wishing to follow. In the face
of British troops with cannon he was to retreat; but “you must be
aware that the attempt to retreat in the face of an Indian force would
be vain.” Proctor’s main force was believed to be Indian.
Neither evacuating nor defending Fort Stephenson, Harrison
remained at Seneca, ten miles behind it, watching for Tecumthe and
the flank attack, and arranging a plan of battle for his eight hundred
men by which he could repel the Indians with dragoons in the open
110
prairie. Croghan remained at Fort Stephenson with one hundred
and sixty men, making every preparation to meet an attack. August 1
the woods were already filled with Indians, and retreat was
impossible, when the British boats appeared on the river, and
Proctor sent to demand surrender of the fort. Immediately on
Croghan’s refusal, the British howitzers opened fire and continued
until it became clear that they were too light to destroy the stockade.
 If experience had been of service to Proctor, he should have
learned to avoid direct attack on Americans in fortified places; but his
position was difficult, and he was as much afraid of Harrison as
Harrison was afraid of him. Fearing to leave Croghan’s little fort in
the rear, and to seek Harrison himself, ten miles above, on the road
to Upper Sandusky; fearing delay, which would discontent his Indian
allies; fearing to go on to Cleveland or Erie without crippling
Harrison; still more afraid to retire to Malden without striking a blow,
—Proctor again sacrificed the Forty-first regiment which had suffered
at the river Raisin and had been surprised at Fort Meigs. On the
afternoon of August 2 the Forty-first regiment and the militia, in three
111
columns of about one hundred and twenty men each, with the
utmost gallantry marched to the pickets of Fort Stephenson, and
were shot down. After two hours’ effort, and losing all its officers, the
assaulting column retired, leaving twenty-six dead, forty-one
wounded, and about thirty missing, or more than one fifth of their
force. The same night the troops re-embarked and returned to
Malden.
112
Proctor’s report of this affair was filled with complaints of the
Indians, who could not be left idle and who would not fight. At
Sandusky, he said, “we could not muster more hundreds of Indians
than I might reasonably have expected thousands.”

“I could not, therefore, with my very small force remain more

than two days, from the probability of being cut off, and of being
deserted by the few Indians who had not already done so.... On
the morning of the 2d inst. the gentlemen of the Indian
department who have the direction of it, declared formally their
decided opinion that unless the fort was stormed we should
never be able to bring an Indian warrior into the field with us, and
that they proposed and were ready to storm one face of the fort
if we would attempt another. I have also to observe that in this
instance my judgment had not that weight with the troops I hope
I might reasonably have expected.... The troops, after the
artillery had been used for some hours, attacked two faces, and
impossibilities being attempted, failed. The fort, from which the
 severest fire I ever saw was maintained during the attack, was
well defended. The troops displayed the greatest bravery, the
much greater part of whom reached the fort and made every
effort to enter; but the Indians who had proposed the assault,
and, had it not been assented to, would have ever stigmatized
the British character, scarcely came into fire before they ran out
of its reach. A more than adequate sacrifice having been made
to Indian opinion, I drew off the brave assailants.”

Sir George Prevost seemed to doubt whether Proctor’s excuse

113
for the defeat lessened or increased the blame attached to it. The
defeat at Sandusky ruined Proctor in the esteem of his men. On the
American side, Harrison’s conduct roused a storm of indignation.
Through the whole day, August 2, he remained at Seneca with eight
hundred men, listening to the cannonade at Fort Stephenson till late
at night, when he received an express from Croghan to say that the
enemy were embarking. The story ran, that as the distant sound of
Croghan’s guns reached the camp at Seneca, Harrison exclaimed:
114
“The blood be on his own head; I wash my hands of it. ” Whatever
else might be true, his conduct betrayed an extravagant estimate of
his enemy’s strength. The only British eye-witness who left an
account of the expedition reckoned Proctor’s force, on its departure
from Malden, at about four hundred troops, and “nearly a thousand
115
Indians.” The Indians dispersed until those with Proctor at Fort
116
Stephenson probably numbered two or three hundred, the rest
having returned to Detroit and Malden. Harrison reported the British
force as five thousand strong, on the authority of General Green
117
Clay.
Whether the British force was large or small, Harrison’s
arrangements to meet it did not please Secretary Armstrong. “It is
118
worthy of notice,” he wrote long afterward, “that of these two
commanders, always the terror of each other, one [Proctor] was now
actually flying from his supposed pursuer; while the other [Harrison]
waited only the arrival of Croghan at Seneca to begin a camp-
conflagration and flight to Upper Sandusky.”
The well-won honors of the campaign fell to Major George
Croghan, with whose name the whole country resounded. Whatever
were the faults of the two generals, Major Croghan showed courage
and intelligence, not only before and during the attack, but afterward
in supporting Harrison against the outcry which for a time threatened
to destroy the General’s authority. Immediately after the siege of Fort
Stephenson every energy of the northwest turned toward a new
offensive movement by water against Malden, and in the task of
organizing the force required for that purpose, complaints of past
failures were stifled. Secretary Armstrong did not forget them, but the
moment was not suited for making a change in so important a
command. Harrison organized, under Armstrong’s orders, a force of
seven thousand men to cross the Lake in boats, under cover of a
fleet.
The fleet, not the army, was to bear the brunt of reconquering the
northwest; and in nothing did Armstrong show his ability so clearly as
in the promptness with which, immediately after taking office, he
stopped Harrison’s campaign on the Maumee, while Perry was set to
work at Erie. Feb. 5, 1813, Armstrong entered on his duties. March 5
his arrangements for the new movements were already made.
119
Harrison did not approve them, but he obeyed. The Navy
Department had already begun operations on Lake Erie, immediately
after Hull’s surrender; but though something was accomplished in
the winter, great difficulties had still to be overcome when February
17 Commander Perry, an energetic young officer on gunboat service
at Newport, received orders from Secretary Jones to report to
Commodore Chauncey on Lake Ontario. Chauncey ordered him to
Presqu’isle, afterward called Erie, to take charge of the vessels
under construction on Lake Erie. March 27 he reached the spot, a
small village in a remote wilderness, where timber and water alone
existed for the supply of the fleets.
When Perry reached Presqu’isle the contractors and carpenters
had on the stocks two brigs, a schooner, and three gunboats. These
were to be launched in May, and to be ready for service in June.
Besides these vessels building at Erie, a number of other craft,
including the prize brig “Caledonia,” were at the Black Rock navy-
yard in the Niagara River, unable to move on account of the British
fort opposite Buffalo and the British fleet on the Lake. Perry’s task
was to unite the two squadrons, to man them, and to fight the British
fleet, without allowing his enemy to interfere at any stage of these
difficult operations.
The British squadron under Commander Finnis, an experienced
officer, had entire control of the Lake and its shores. No regular
garrison protected the harbor of Presqu’isle; not two hundred men
could be armed to defend it, nor was any military support to be had
nearer than Buffalo, eighty miles away. Proctor or Prevost were likely
to risk everything in trying to destroy the shipyard at Erie; for upon
that point, far more than on Detroit, Fort Meigs, Sandusky, or Buffalo,
their existence depended. If Perry were allowed to control the Lake,
the British must not only evacuate Detroit, but also Malden, must
abandon Tecumthe and the military advantages of three or four
thousand Indian auxiliaries, and must fall back on a difficult
defensive at the Niagara River. That they would make every effort to
thwart Perry seemed certain.
Superstition survived in nothing more obstinately than in faith in
luck; neither sailors nor soldiers ever doubted the value of this
inscrutable quality in the conduct of war. The “Chesapeake” was an
unlucky ship to the luckiest commanders, even to the British captain
who captured it. The bad luck of the “Chesapeake” was hardly
steadier than the good luck of Oliver Perry. Whatever he touched
seemed to take the direction he wanted. He began with the
advantage of having Proctor for his chief enemy; but Harrison, also a
lucky man, had the same advantage and yet suffered constant
disasters. Commander Finnis was a good seaman, yet Finnis failed
repeatedly, and always by a narrow chance, to injure Perry.
Dearborn’s incompetence in 1813 was not less than it had been in
1812; but the single success which in two campaigns Dearborn
gained on the Niagara obliged the British, May 27, to evacuate Fort
Erie opposite Buffalo, and to release Perry’s vessels at Black Rock.
June 6, at leisure, Perry superintended the removal of the five small
craft from the navy-yard at Black Rock; several hundred soldiers,
seamen, and oxen warped them up stream into the Lake. Loaded
with stores, the little squadron sailed from Buffalo June 13; the wind
was ahead; they were five days making eighty miles; but June 19
they arrived at Presqu’isle, and as the last vessel crossed the bar,
Finnis and his squadron came in sight. Finnis alone could explain
how he, a first-rate seaman, with a strong force and a fair wind, in
such narrow seas, could have helped finding Perry’s squadron when
he knew where it must be.
From June 19 to August 1 Perry’s combined fleet lay within the
bar at Presqu’isle, while Proctor, with a sufficient fleet and a military
force superior to anything on the Lake, was planning expeditions
from Malden against every place except the one to which military
necessity and the orders of his Government bade him go. August 4,
Perry took out the armaments of his two brigs and floated both over
the bar into deep water. Had the British fleet been at hand, such a
movement would have been impossible or fatal; but the British fleet
appeared just as Perry’s vessels got into deep water, and when for
the first time an attack could not be made with a fair hope of
success.
These extraordinary advantages were not gained without labor,
energy, courage, and wearing anxieties and disappointments. Of
these Perry had his full share, but no more; and his opponents were
no better off than himself. By great exertions alone could the British
maintain themselves on Lake Ontario, and to this necessity they
were forced to sacrifice Lake Erie. Sir George Prevost could spare
only a new commander with a few officers and some forty men from
the lower Lake to meet the large American reinforcements on the
upper. When the commander, R. H. Barclay, arrived at Malden in
June, he found as many difficulties there as Perry found at
Presqu’isle. Barclay was a captain in the British Royal Navy, thirty-
two years old; he had lost an arm in the service, but he was fairly
matched as Perry’s antagonist, and showed the qualities of an
excellent officer.
 Perry’s squadron, once on the Lake, altogether overawed the
British fleet, and Barclay’s only hope lay in completing a vessel
called the “Detroit,” then on the stocks at Amherstburg. Rough and
unfinished, she was launched, and while Perry blockaded the harbor,
Barclay, early in September, got masts and rigging into her, and
120
armed her with guns of every calibre, taken from the ramparts.
Even the two American twenty-four pound guns, used by Proctor
against Fort Meigs, were put on board the “Detroit.” Thus equipped,
she had still to be manned; but no seamen were near the Lake.
Barclay was forced to make up a crew of soldiers from the
hardworked Forty-first regiment and Canadians unused to service.
September 6 the “Detroit” was ready to sail, and Barclay had then no
choice but to fight at any risk. “So perfectly destitute of provisions
was the port that there was not a day’s flour in store, and the crews
of the squadron under my command were on half allowance of many
121
things; and when that was done, there was no more.”
Early on the morning of September 9 Barclay’s fleet weighed and
sailed for the enemy, who was then at anchor off the island of Put-in-
Bay near the mouth of Sandusky River. The British squadron
consisted of six vessels,—the “Detroit,” a ship of four hundred and
ninety tons, carrying nineteen guns, commanded by Barclay himself;
the “Queen Charlotte” of seventeen guns, commanded by Finnis; the
“Lady Prevost” of thirteen guns; the “Hunter” of ten; the “Little Belt”
carrying three, and the “Chippeway” carrying one gun,—in all, sixty-
three guns, and probably about four hundred and fifty men. The
American squadron consisted of nine vessels,—the “Lawrence,”
Perry’s own brig, nearly as large as the “Detroit,” and carrying twenty
guns; the “Niagara,” commander Jesse D. Elliott, of the same
tonnage, with the same armament; the “Caledonia,” a three-gun brig;
the schooners “Ariel,” “Scorpion,” “Somers,” “Porcupine,” and
“Tigress,” carrying ten guns; and the sloop “Trippe,” with one gun,—
in all, fifty-four guns, with a nominal crew of five hundred and thirty-
two men, and an effective crew probably not greatly differing from
the British. In other respects Perry’s superiority was decided, as it
was meant to be. The Americans had thirty-nine thirty-two pound
carronades; the British had not a gun of that weight, and only fifteen
twenty-four pound carronades. The lightest guns on the American
fleet were eight long twelve-pounders, while twenty-four of the British
guns threw only nine-pound shot, or less. The American broadside
threw at close range about nine hundred pounds of metal; the British
threw about four hundred and sixty. At long range the Americans
threw two hundred and eighty-eight pounds of metal; the British
threw one hundred and ninety-five pounds. In tonnage the
Americans were superior as eight to seven. In short, the Navy
Department had done everything reasonably necessary to insure
success; and if the American crews, like the British, were partly
made up of landsmen, soldiers or volunteers, the reason was in each
case the same. Both governments supplied all the seamen they had.
Between forces so matched, victory ought not to have been in
doubt; and if it was so, the fault certainly lay not in Perry. When, at
daylight September 10, his look-out discovered the British fleet,
Perry got his own squadron under way, and came down with a light
wind from the southeast against Barclay’s line, striking it obliquely
near the head. Perry must have been anxious to fight at close range,
where his superiority was as two to one, while at long range his ship
could use only two long twelve-pounders against the “Detroit’s” six
twelves, one eighteen, and two twenty-fours,—an inferiority
amounting to helplessness. Both the “Lawrence” and the “Niagara”
were armed for close fighting, and were intended for nothing else. At
long range their combined broadside, even if all their twelve-
pounders were worked on one side, threw but forty-eight pounds of
metal; at short range the two brigs were able to throw six hundred
and forty pounds at each broadside.
Perry could not have meant to fight at a distance, nor could
Commander Elliott have thought it good seamanship. Yet Perry
alone acted on this evident scheme; and though his official account
showed that he had himself fought at close range, and that he
ordered the other commanders to do the same, it gave no sufficient
reasons to explain what prevented the whole fleet from acting
together, and made the result doubtful. He did not even mention that
he himself led the line in the “Lawrence,” with two gunboats, the
“Ariel” and the “Scorpion,” supporting him, the “Caledonia,”
“Niagara,” and three gunboats following. The “Lawrence” came
within range of the British line just at noon, the wind being very light,
the Lake calm, and Barclay, in the “Detroit,” opposite. Perry’s report
began at that point:—

“At fifteen minutes before twelve the enemy commenced

firing; at five minutes before twelve the action commenced on
our part. Finding their fire very destructive, owing to their long
guns, and its being mostly directed to the ‘Lawrence,’ I made sail
(at quarter-past twelve) and directed the other vessels to follow,
for the purpose of closing with the enemy. Every brace and
bowline being shot away, she became unmanageable,
notwithstanding the great exertions of the sailing-master. In this
situation she sustained the action upwards of two hours, within
canister-shot distance, until every gun was rendered useless,
and a greater part of the crew either killed or wounded. Finding
she could no longer annoy the enemy, I left her in charge of
Lieutenant Yarnall, who, I was convinced from the bravery
already displayed by him, would do what would comport with the
honor of the flag. At half-past two, the wind springing up, Captain
Elliott was enabled to bring his vessel, the ‘Niagara,’ gallantly
into close action. I immediately went on board of her, when he
anticipated my wish by volunteering to bring the schooners,
which had been kept astern by the lightness of the wind, into
close action.... At forty-five minutes past two the signal was
made for ‘close action.’ The ‘Niagara’ being very little injured, I
determined to pass through the enemy’s line; bore up, and
passed ahead of their two ships and a brig, giving a raking fire to
them from the starboard guns, and to a large schooner and
sloop, from the larboard side, at half pistol-shot distance. The
smaller vessels at this time having got within grape and canister
distance, under the direction of Captain Elliott, and keeping up a
well-directed fire, the two ships, a brig, and a schooner
surrendered, a schooner and sloop making a vain attempt to
escape.”
 From this reticent report, any careful reader could see that for
some reason, not so distinctly given as would have been the case if
the wind alone were at fault, the action had been very badly fought
on the American side. The British official account confirmed the
impression given by Perry. Barclay’s story was as well told as his
action was well fought:—

“At a quarter before twelve I commenced the action by a few

long guns; about a quarter-past, the American commodore, also
supported by two schooners, ... came to close action with the
‘Detroit.’ The other brig [the ‘Niagara’] of the enemy, apparently
destined to engage the ‘Queen Charlotte,’ kept so far to
windward as to render the ‘Queen Charlotte’s’ twenty-four
pounder carronades useless, while she was, with the ‘Lady
Prevost,’ exposed to the heavy and destructive fire of the
‘Caledonia’ and four other schooners, armed with heavy and
long guns.... The action continued with great fury until half-past
two, when I perceived my opponent [the ‘Lawrence’] drop astern,
and a boat passing from him to the ‘Niagara,’ which vessel was
at this time perfectly fresh. The American commodore, seeing
that as yet the day was against him, ... made a noble and, alas!
too successful an effort to regain it; for he bore up, and
supported by his small vessels, passed within pistol-shot and
took a raking position on our bow.... The weather-gage gave the
enemy a prodigious advantage, as it enabled them not only to
choose their position, but their distance also, which they [the
‘Caledonia,’ ‘Niagara,’ and the gunboats] did in such a manner
as to prevent the carronades of the ‘Queen Charlotte’ and ‘Lady
Prevost’ from having much effect, while their long ones did great
execution, particularly against the ‘Queen Charlotte.’”

Barclay’s report, agreeing with Perry’s, made it clear that while

Perry and the head of the American line fought at close quarters, the
“Caledonia,” “Niagara,” and the four gunboats supporting them
preferred fighting at long range,—not because they wanted wind, but
because the “Caledonia” and gunboats were armed with long thirty-
two and twenty-four pounders, while the British vessels opposed to
them had only one or two long twelve-pounders. Certainly the
advantage in this respect on the side of the American brig and
gunboats was enormous; but these tactics threw the “Niagara,”
which had not the same excuse, out of the battle, leaving her, from
twelve o’clock till half-past two, firing only two twelve-pound guns,
while her heavy armament was useless, and might as well have
been left ashore. Worse than this, the persistence of the “Caledonia,”
“Niagara,” and their gunboats in keeping, beyond range of their
enemies’ carronades nearly lost the battle, by allowing the British to
concentrate on the “Lawrence” all their heavy guns, and in the end
compelling the “Lawrence” to strike. On all these points no
reasonable doubt could exist. The two reports were the only official
sources of information on which an opinion as to the merits of the
action could properly be founded. No other account,
contemporaneous and authoritative, threw light on the subject,
except a letter by Lieutenant Yarnall, second in command to Perry
on the “Lawrence,” written September 15, and published in the Ohio
newspapers about September 29,—in which Yarnall said that if Elliott
had brought his ship into action when the signal was given, the battle
would have ended in much less time, and with less loss to the
“Lawrence.” This statement agreed with the tenor of the two official
reports.
Furious as the battle was, a more furious dispute raged over it
when in the year 1834 the friends of Perry and of Elliott wrangled
over the action. With their dispute history need not concern itself.
The official reports left no reasonable doubt that Perry’s plan of
battle was correct; that want of wind was not the reason it failed; but
that the “Niagara” was badly managed by Elliott, and that the victory,
when actually forfeited by this mismanagement, was saved by the
personal energy of Perry, who, abandoning his own ship, brought the
“Niagara” through the enemy’s line, and regained the advantage of
her heavy battery. The luck which attended Perry’s career on the
Lake saved him from injury, when every other officer on the two
opposing flagships and four-fifths of his crew were killed or wounded,
and enabled him to perform a feat almost without parallel in naval
warfare, giving him a well-won immortality by means of the disaster
unnecessarily incurred. No process of argument or ingenuity of
seamanship could deprive Perry of the fame justly given him by the
public, or detract from the splendor of his reputation as the hero of
the war. More than any other battle of the time, the victory on Lake
Erie was won by the courage and obstinacy of a single man.
Between two opponents such as Perry and Barclay, no one
doubted that the ships were fought to their utmost. Of the “Lawrence”
not much was left; ship, officers, and crew were shot to pieces. Such
carnage was not known on the ocean, for even the cockpit where the
sick and wounded lay, being above water, was riddled by shot, and
the wounded were wounded again on the surgeon’s board. Of one
hundred and three effectives on the “Lawrence,” twenty-two were
killed and sixty-one wounded. The brig herself when she struck was
a wreck, unmanageable, her starboard bulwarks beaten in, guns
dismounted, and rigging cut to pieces. The British ships were in
hardly better condition. The long guns of the gunboats had raked
them with destructive effect. Barclay was desperately wounded;
Finnis was killed; Barclay’s first lieutenant was mortally wounded; not
one commander or second in command could keep the deck; the
squadron had forty-one men killed and ninety-four wounded, or
nearly one man in three; the “Detroit” and “Queen Charlotte” were
unmanageable and fell foul; the “Lady Prevost” was crippled, and
drifted out of the fight. Perry could console himself with the thought
that if his ship had struck her flag, she had at least struck to brave
men.
 CHAPTER VI.
General Harrison, waiting at Seneca on the Sandusky River,
received, September 12, Perry’s famous despatch of September 10:
“We have met the enemy, and they are ours.” The navy having done
its work, the army was next to act.
The force under Harrison’s command was ample for the required
purpose, although it contained fewer regular troops than Armstrong
had intended. The seven regular regiments assigned to Harrison fell
short in numbers of the most moderate expectations. Instead of
providing seven thousand rank-and-file, the recruiting service ended
122
in producing rather more than twenty-five hundred. Divided into
two brigades under Brigadier-Generals McArthur and Lewis Cass,
with a light corps under Lieutenant-Colonel Ball of the Light
Dragoons, they formed only one wing of Harrison’s army.
To supply his main force, Harrison had still to depend on
Kentucky; and once more that State made a great effort. Governor
Shelby took the field in person, leading three thousand
123
volunteers, organized in eleven regiments, five brigades, and two
divisions. Besides the militia, who volunteered for this special
purpose, Harrison obtained the services of another Kentucky corps,
which had already proved its efficiency.
One of Armstrong’s happiest acts, at the beginning of his service
124
as War Secretary, was to accept the aid of Richard M. Johnson in
organizing for frontier defence a mounted regiment of a thousand
125
men, armed with muskets or rifles, tomahawks, and knives.
Johnson and his regiment took the field about June 1, and from that
time anxiety on account of Indians ceased. The regiment patrolled
the district from Fort Wayne to the river Raisin, and whether in
marching or fighting proved to be the most efficient corps in the
Western country. Harrison obtained the assistance of Johnson’s
regiment for the movement into Canada, and thereby increased the
efficiency of his army beyond the proportion of Johnson’s numbers.
While the mounted regiment moved by the road to Detroit,
Harrison’s main force was embarked in boats September 20, and in
the course of a few days some forty-five hundred infantry were safely
conveyed by way of Bass Island and Put-in-Bay to Middle Sister
126
Island, about twelve miles from the Canadian shore. Harrison and
Perry then selected a landing place, and the whole force was
successfully set ashore, September 27, about three miles below
Malden.
Although Proctor could not hope to maintain himself at Malden or
Detroit without control of the Lake, he had still the means of
rendering Harrison’s possession insecure. According to the British
account, he commanded at Detroit and Malden a force of nine
hundred and eighty-six regulars, giving about eight hundred
127
effectives. Not less than thirty-five hundred Indian warriors had
flocked to Amherstburg, and although they greatly increased the
British general’s difficulties by bringing their families with them, they
might be formidable opponents to Harrison’s advance. Every motive
dictated to Proctor the necessity of resisting Harrison’s approach. To
Tecumthe and his Indians the evacuation of Malden and Detroit
without a struggle meant not only the sacrifice of their cause, but
also cowardice; and when Proctor announced to them, September
18, that he meant to retreat, Tecumthe rose in the council and
protested against the flight, likening Proctor to a fat dog that had
carried its tail erect, and now that it was frightened dropped its tail
128
between its legs and ran. He told Proctor to go if he liked, but the
Indians would remain.
Proctor insisted upon retiring at least toward the Moravian town,
seventy miles on the road to Lake Ontario, and the Indians yielded.
The troops immediately began to burn or destroy the public property
at Detroit and Malden, or to load on wagons or boats what could not
be carried away. September 24, three days before Harrison’s army
landed, the British evacuated Malden and withdrew to Sandwich,
allowing Harrison to establish himself at Malden without a skirmish,
and neglecting to destroy the bridge over the Canards River.
Harrison was surprised at Proctor’s tame retreat.

129
“Nothing but infatuation,” he reported, “could have
governed General Proctor’s conduct. The day that I landed
below Malden he had at his disposal upward of three thousand
Indian warriors; his regular force reinforced by the militia of the
district would have made his number nearly equal to my
aggregate, which on the day of landing did not exceed forty-five
hundred.... His inferior officers say that his conduct has been a
series of continued blunders.”

This crowning proof of Proctor’s incapacity disorganized his

force. Tecumthe expressed a general sentiment of the British army in
his public denunciation of Proctor’s cowardice. One of the inferior
British officers afterward declared that Proctor’s “marked inefficiency”
and “wanton sacrifice” of the troops raised more than a doubt not
only of his capacity but even of his personal courage, and led to
130
serious thoughts of taking away his authority. The British at
Sandwich went through the same experience that marked the retreat
of Hull and his army from the same spot, only the year before.
Harrison on his side made no extreme haste to pursue. His army
marched into Malden at four o’clock on the afternoon of September
131
27, and he wrote to Secretary Armstrong that evening: “I will
pursue the enemy to-morrow, although there is no probability of my
overtaking him, as he has upwards of a thousand horses, and we
132
have not one in the army.” The pursuit was not rapid. Sandwich,
opposite Detroit, was only thirteen miles above Malden, but Harrison
required two days to reach it, arriving at two o’clock on the afternoon
of September 29. From there, September 30, he wrote again to
Secretary Armstrong that he was preparing to pursue the enemy on
133
the following day; but he waited for R. M. Johnson’s mounted
regiment, which arrived at Detroit September 30, and was obliged to
consume a day in crossing the river. Then the pursuit began with
energy, but on the morning of October 2 Proctor had already a
week’s advance and should have been safe.
Proctor seemed to imagine that the Americans would not venture
134
to pursue him. Moving, according to his own report, “by easy
marches,” neither obstructing the road in his rear nor leaving
detachments to delay the enemy, he reached Dolson’s October 1,
and there halted his army, fifty miles from Sandwich, while he went
to the Moravian town some twenty-six miles beyond. He then
intended to make a stand at Chatham, three miles behind Dolson’s.

“I had assured the Indians,” said Proctor’s report of October

23, “that we would not desert them, and it was my full
determination to have made a stand at the Forks (Chatham), by
which our vessels and stores would be protected; but after my
arrival at Dover [Dolson’s] three miles lower down the river, I was
induced to take post there first, where ovens had been
constructed, and where there was some shelter for the troops,
and had accordingly directed that it should be put into the best
possible state of defence that time and circumstances would
admit of; indeed it had been my intention to have opposed the
enemy nearer the mouth of the river, had not the troops contrary
to my intention been moved, during my absence of a few hours
for the purpose of acquiring some knowledge of the country in
my rear.”

The British army, left at Dolson’s October 1, without a general or

135
orders, saw the American army arrive in its front, October 3, and
retired three miles to Chatham, where the Indians insisted upon
fighting; but when, the next morning, October 4, the Americans
136
advanced in order of battle, the Indians after a skirmish changed
their minds and retreated. The British were compelled to sacrifice the
supplies they had brought by water to Chatham for establishing their