Professional Documents
Culture Documents
Download pdf Simulation And Modeling Methodologies Technologies And Applications 8Th International Conference Simultech 2018 Porto Portugal July 29 31 2018 Revised Selected Papers Mohammad S Obaidat ebook full chapter
Download pdf Simulation And Modeling Methodologies Technologies And Applications 8Th International Conference Simultech 2018 Porto Portugal July 29 31 2018 Revised Selected Papers Mohammad S Obaidat ebook full chapter
https://textbookfull.com/product/e-business-and-
telecommunications-16th-international-conference-
icete-2019-prague-czech-republic-july-26-28-2019-revised-
selected-papers-mohammad-s-obaidat/
https://textbookfull.com/product/software-technologies-
applications-and-foundations-staf-2018-collocated-workshops-
toulouse-france-june-25-29-2018-revised-selected-papers-manuel-
mazzara/
Mohammad S. Obaidat
Tuncer Ören
Floriano De Rango Editors
Simulation
and Modeling
Methodologies,
Technologies and
Applications
8th International Conference,
SIMULTECH 2018, Porto, Portugal, July
29–31, 2018, Revised Selected Papers
Advances in Intelligent Systems and Computing
Volume 947
Series Editor
Janusz Kacprzyk, Systems Research Institute, Polish Academy of Sciences,
Warsaw, Poland
Advisory Editors
Nikhil R. Pal, Indian Statistical Institute, Kolkata, India
Rafael Bello Perez, Faculty of Mathematics, Physics and Computing,
Universidad Central de Las Villas, Santa Clara, Cuba
Emilio S. Corchado, University of Salamanca, Salamanca, Spain
Hani Hagras, School of Computer Science and Electronic Engineering,
University of Essex, Colchester, UK
László T. Kóczy, Department of Automation, Széchenyi István University,
Gyor, Hungary
Vladik Kreinovich, Department of Computer Science, University of Texas
at El Paso, El Paso, TX, USA
Chin-Teng Lin, Department of Electrical Engineering, National Chiao
Tung University, Hsinchu, Taiwan
Jie Lu, Faculty of Engineering and Information Technology,
University of Technology Sydney, Sydney, NSW, Australia
Patricia Melin, Graduate Program of Computer Science, Tijuana Institute
of Technology, Tijuana, Mexico
Nadia Nedjah, Department of Electronics Engineering, University of Rio de Janeiro,
Rio de Janeiro, Brazil
Ngoc Thanh Nguyen , Faculty of Computer Science and Management,
Wrocław University of Technology, Wrocław, Poland
Jun Wang, Department of Mechanical and Automation Engineering,
The Chinese University of Hong Kong, Shatin, Hong Kong
The series “Advances in Intelligent Systems and Computing” contains publications
on theory, applications, and design methods of Intelligent Systems and Intelligent
Computing. Virtually all disciplines such as engineering, natural sciences, computer
and information science, ICT, economics, business, e-commerce, environment,
healthcare, life science are covered. The list of topics spans all the areas of modern
intelligent systems and computing such as: computational intelligence, soft comput-
ing including neural networks, fuzzy systems, evolutionary computing and the fusion
of these paradigms, social intelligence, ambient intelligence, computational neuro-
science, artificial life, virtual worlds and society, cognitive science and systems,
Perception and Vision, DNA and immune based systems, self-organizing and
adaptive systems, e-Learning and teaching, human-centered and human-centric
computing, recommender systems, intelligent control, robotics and mechatronics
including human-machine teaming, knowledge-based paradigms, learning para-
digms, machine ethics, intelligent data analysis, knowledge management, intelligent
agents, intelligent decision making and support, intelligent network security, trust
management, interactive entertainment, Web intelligence and multimedia.
The publications within “Advances in Intelligent Systems and Computing” are
primarily proceedings of important conferences, symposia and congresses. They
cover significant recent developments in the field, both of a foundational and
applicable character. An important characteristic feature of the series is the short
publication time and world-wide distribution. This permits a rapid and broad
dissemination of research results.
** Indexing: The books of this series are submitted to ISI Proceedings,
EI-Compendex, DBLP, SCOPUS, Google Scholar and Springerlink **
Floriano De Rango
Editors
123
Editors
Mohammad S. Obaidat Tuncer Ören
King Abdullah II School of Information School of Electrical Engineering
Technology and Computer Science
University of Jordan University of Ottawa
Amman, Jordan Ottawa, ON, Canada
Nazarbayev University
Astana, Kazakhstan
Floriano De Rango
University of Calabria
Rende, Cosenza, Italy
This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
The present book includes extended and revised versions of a set of selected papers
from the 8th International Conference on Simulation and Modeling Methodologies,
Technologies and Applications (SIMULTECH 2018), held in Porto, Portugal, in the
period July 29–31, 2018.
SIMULTECH 2018 received 83 paper submissions from 29 countries, of which
11% were included in this book. The papers were selected by the event chairs, and
their selection is based on a number of criteria that includes the reviews and
suggested comments provided by the program committee members, the session
chairs’ assessments and also the program chairs’ global view of all papers included
in the technical program. The authors of selected papers were then invited to submit
a revised and extended version of their papers having at least 30% new material.
The purpose of the 8th International Conference on Simulation and Modeling
Methodologies, Technologies and Applications (SIMULTECH 2018) was to bring
together researchers, engineers, applied mathematicians and practitioners interested
in the advances and applications in the field of system simulation. Four
simultaneous tracks were held, covering on one side domain-independent
methodologies and technologies and on the other side practical work developed
in specific application areas. The specific topics listed under each of these tracks
highlight the interest of this conference in aspects related to computing, including
conceptual modeling, agent-based modeling and simulation, interoperability,
ontologies, knowledge-based decision support, Petri nets, business process mod-
eling and simulation, among others.
The papers selected to be included in this book contribute to the understanding
of relevant trends of current research on simulation tools and platforms, formal
methods, as well as complex system modeling and simulation. The readers can find
contributions of simulation in business process analysis and risk management,
Internet of things, advanced material science and fuel cell design. Formal method
contributions include visual password systems, co-simulation of complex
subsystems, heating system application of feedback linearization for MTI systems
in a tensor framework and iterative construction of complete Lyapunov functions.
v
vi Preface
We would like to thank all the authors for their contributions as well as to the
reviewers who have helped ensuring the quality of this publication. We also thank
the staff of INTICCC and Springer for their good efforts and cooperation.
Conference Chair
Mohammad S. Obaidat University of Jordan, Jordan, and Nazarbayev
University, Kazakhstan
Program Chairs
Tuncer Ören (Honorary) University of Ottawa, Canada
Floriano De Rango University of Calabria, Italy
Program Committee
Nael Abu-Ghazaleh University of California, Riverside, USA
Lyuba Alboul Sheffield Hallam University, UK
Mikulas Alexik University of Zilina, Slovak Republic
Carlos Argáez University of Iceland, Iceland
Gianfranco Balbo University of Torino, Italy
Isaac Barjis City University of New York, USA
Martin Benedikt Technische Universität Graz, Austria
Mohamed Bettaz Philadelphia University, Jordan
Louis Birta University of Ottawa, Canada
Wolfgang Borutzky Bonn-Rhein-Sieg University of Applied
Sciences, Germany
Christos Bouras University of Patras and CTI&P Diophantus,
Greece
António Brito INESC TEC, Faculdade de Engenharia,
Universidade do Porto, Portugal
Jesus Carretero Computer Architecture Group, University
Carlos III of Madrid, Spain
Francesco Casella Politecnico di Milano, Italy
Rodrigo Castro University of Buenos Aires, Argentina
vii
viii Organization
Additional Reviewers
Gökhan Cinar Eskisehir Osmangazi University, Turkey
Tansu Filik Anadolu University, Turkey
Celal Kandemir Eskisehir Osmangazi Üniversitesi, Turkey
Francesco Mercaldo National Research Council of Italy (CNR), Italy
Ezequiel Pecker-Marcosig FIUBA, ICC, CONICET, Argentina
Lucio Santi Universidad de Buenos Aires, Argentina
Long Wang Johns Hopkins University, USA
xii Organization
Invited Speakers
Juan M. Durán University of Stuttgart, Germany
Janusz Kacprzyk Systems Research Institute, Polish Academy
of Sciences, Poland
Gabriel Wainer Carleton University, Canada
Contents
xiii
xiv Contents
1 Introduction
Digitals tools are more and more required to study, design and prototype nano-objects,
although the underlying physics is so complex that the quest for a universal tool is still
far from being over. With the increase of the computational power and the
improvement of the simulation methods, new possibilities are emerging. The increasing
pace of the semiconductor industry requires rapid and efficient simulation and mod-
elling strategies to analyse the results and improve the technological performances of
various nano-devices, sensors or actuators. In many systems, the optical or electronic
properties are driven by interfacial or by defect-engineered phenomena. In order to
understand the links between structure and properties, the nanocharacterization of
materials and devices must be advantageously combined with atomistic modelling
studies. The equilibrium positions of all atoms provide the necessary basis to simulate
the relevant physical properties, which are measured with increasing precision and
sensitivity. The combination of experiments conducted in parallel with simulations is
particularly relevant in the field of transmission electron microscopy (TEM), because
the correlation between the measured image and the actual arrangement of atoms is not
straightforward in general. Like most characterizations, the precise simulation of TEM
images is usually mandatory to interpret the experimental results at the atomic scale.
With developments in aberration-corrected transmission electron microscopy, it is now
possible to characterize vacancy defects in graphene [1]. This paper provides an
optimized methodology to model high resolution scanning transmission electron
microscopy (HRSTEM) experiments of graphene-based defects. For this, it uses
relaxed atomistic models obtained with the Brenner module of the Software for
Adaptative Modeling and Simulation Of Nanosystems (SAMSON) (www.samson-
connect.fr). The case of graphene-based defects is extremely interesting, because this is
a 2D material with outstanding mechanical [2–6], and electronic [7, 8] properties.
Hence, graphene belongs to a family of 2D materials which generates huge expecta-
tions in terms of possible applications [9–14]. The high mobility of graphene makes it
advantageous in the perspective of post-silicon electronics, but the defectivity remains
a recurrent critical issue. A wide variety of deviations from a perfect crystal might
occur during the processing of graphene, either due to the growth conditions or to
various sources of degradation, such as knock-on interactions, electron or ionic colli-
sions, plasma damage, chemical reactions, etc. The link between defectivity and
physical properties is critical for the device performance. Thus, defect engineering is
certainly the key of the possible industrial viability of this material. In this paper, we
study the defects in graphene in terms of structure and energy. Our methodology used
to build the systems and to simulate TEM images provides the necessary basis to
analyze graphene-based defects, in comparison with available data from the literature.
2 Methodology
crystals, the 3D periodicity helps in calculating all the atom positions for large systems
(i.e. more than 50 000 atoms), but in case of localised asymmetrical defects, this task is
much more tedious or even completely unfeasible in the worst cases. Hence, a com-
putational tool that is fast enough to handle physically-relevant calculations with tens
of thousands atoms is highly desirable. The SAMSON platform and its Brenner module
appear to be ideally suited to this task, since they can handle complex models and
simulate very large systems in a timescale typically less than a day, which is com-
patible with the feedback delay required by most research teams in nanomaterial
characterization. This module appears as an interactive tool for performing predictive
modelling, particularly adapted to the very sustained pace of experimentalists.
The details are given in the original reference [17]. Since bonds are defined
dynamically via a bond-order function evolving with the interatomic distances, this
reactive potential has the ability to describe chemical reactions.
4 C. Guedj et al.
The potential also includes angular and dihedral terms, radical energetics and the
influence of p bonds [25].
To overcome the lack of long-range interactions, a non-bonded interaction potential
term is added. It consists in a sum of pairwise potential contributions. For simplicity,
the approach of Los et al. [26] is chosen and the Van der Waals potential term is added:
r6
VNB ðrij Þ ¼ b expðc0 rÞ 2 Vshift ð2Þ
r
to adjust the precision, using b = 3224.9 eV, c0 = 35.995 nm−1, 2 ¼ 0:01396 eV and
r = 0.344 nm.
Forces
The force terms can be calculated from the gradient of the potential V. More specifi-
cally, the Force Fi applied on atom i at position xi can be written:
X
@V @V @rji
Fi ¼ ¼ j;ði;jÞ2b dr
ð3Þ
dxi ij dxi
where rij is the distance between atoms i and j, and b is the set of all pairs of atoms
involved in the interaction:
ij g
b ¼ fi; rij\Dmax ð4Þ
with Dmax
ij being a threshold distance depending on the atom types.
Adaptive Brenner
The adaptive version of the Brenner potential has been implemented in SAMSON [25].
Its interest is that it relies on an algorithm which incrementally updates the forces and
the total potential energy. It basically consists in an incremental dynamical update of
the set of interacting atoms and all information related to one, two, three or four atoms.
Bonds are divided into 4 types: bond with a relative motion, bonds with a change in
potential, bond with a change in conjugate number, and bonds without any change in
potential. After initialization, all terms with relative motions are updated incrementally
and after a first level and second level potential update, the forces are henceforward
updated. This allows the algorithm to linearly scale with the number of updated bonds.
Therefore the computational cost is decoupled from the number of atoms in the system
and physically-based editing becomes markedly faster.
To take advantage of adaptive Brenner, an adaptive mechanism is proposed in
SAMSON to update when minimizing a system. Such an approach in Cartesian
coordinates consists in deciding for each atom if it might move or be frozen in space.
This decision is made by comparing the norm of its potential displacement with a
threshold value, either automatically deduced from the system state or by a manual
choice fixed by the user. This implementation is an extension of the internal coordi-
nates and articulated bodies simulation [27].
This efficient update mechanism allows continuous minimization of the system
energy during the edition of the system, which helps to build realistic structures in a
Atomistic Modelling and Simulation of Transmission Electron Microscopy Images 5
very convenient manner. The user action step (creating/moving/deleting atoms) alter-
nates with the adaptative minimization steps to parallelize the structure editing and the
energy minimization.
atoms with flat borders, built in SAMSON and optimized thanks to the Brenner
module. The clear advantage of the Brenner approach compared to ab-initio is a *4
orders of magnitude improvement in terms of simulation speed. Moreover, as we will
see, the precision achieved is sufficient to match the experimental results and we
obtained similar findings for all the graphene-based defects we found in literature,
without apparent limitation, and even for systems with tens of thousands of atoms. In
the following, all experimental data already published are used with permissions.
Fig. 1. Left: atomistic ball and stick model of the unstable flat SW(55-77) defect in graphene.
Black balls represent carbon atoms. Right: Corresponding HRSTEM-HAADF simulated image.
In pure graphene, the C-C bond distance is 0.142 nm according to Pauling [50], which
is the exact value provided by our code. The simulation also matches very well previous
experimental results [51–53] and the corresponding ab-initio simulations [54, 55].
The planar configuration is unstable and may relax in the 3D sinelike or cosinelike
configuration. In our case, the minimum energy configuration of 6 eV is obtained for
the sinelike configuration (Fig. 2), in reasonable agreement with the configuration and
the energy of 5.82 ± 0.03 eV obtained by quantum Monte Carlo [55] and the value of
5.9 eV obtained by DFT-LDA [56].
An absolute comparison with the exact and precise value of the formation energy is
difficult, because of the significant dispersion of formation energies published in the
literature, depending on the DFT options or the size of the supercell [54, 55, 57–59].
Meanwhile, the buckling height value of 0.156 nm is very close to the value of
0.161 nm obtained by DFT for the biggest cell (11 11) of Ma et al. [55].
The SW defects are not simple planar defects but rather involve 3D displacements
(Fig. 3).
The Fig. 4 shows that a reasonable agreement is obtained between the experimental
and simulated p-RDF but many factors may influence these results. Experimentally, the
8 C. Guedj et al.
Fig. 2. Atomistic model of the lowest energy configuration SW(55-77) sinelike defect in
graphene, with bond distances, superimposed with the experimental HRTEM image of Kotakoski
[52, 53]. Colorization has been added to help the interpretation.
HRTEM image may suffer from optical aberration, sample drift, contamination and
damage, especially with the very fragile defects of graphene. The presence of surface
oxycarboneous species or hydrogen at the surface may slighty modify the images
during these difficult measurements. The microscope parameters are usually prone to
various sources of instabilities, therefore a direct comparison between experiment and
simulation is not straightforward, nevertheless a reasonable agreement is obtained.
A first major peak around 142 pm is observed in both the experimental and sim-
ulated p-RDF, which corresponds to the carbon atoms in first nearest neighbour (C1nn)
configuration. The peak is very sharp in the simulation, because the positions are frozen
around the equilibrium configuration. In the HRTEM experiment, a significant devi-
ation from the ideal positions is observed, because of all sources of blurring and noise.
The maximum deviation between experiment and simulation is around ±40 pm, as can
be verified at looking the experimental broadening of the C1nn peak. Hence, in this
case, a fairly good quantitative agreement between experiment and simulation is
obtained, with a maximum relative deviation of atomic positions less than 40 pm.
Atomistic Modelling and Simulation of Transmission Electron Microscopy Images 9
Fig. 4. Comparison between the p-RDF deduced from the HRTEM image of Kotakoski [52, 53]
and from the atomistic simulations using the Brenner potential (bottom).
V1 (5-9)
The removal of one carbon atom from the graphene network results in the formation of
a single vacancy, which has been studied both theoretically and experimentally [52–55,
60, 61].
Our simulated model matches extremely well the experimental HRTEM images
published in the literature (Figs. 5, 6 and 7). We obtained a formation energy of
5.45 eV, which is less than the range of [7.6, 7.9] eV obtained by DFT [59]. The
symmetric monovacancy (s-MV) is known to exhibit a Jahn Teller distortion, and may
reconstruct into a closed five- and nine-membered pair of rings. The reconstructed
monovacancy (r-MV) arrangement lowers the energy of the symmetrical vacancy
structure in agreement with ab-initio calculations [62].
Fig. 5. Let: atomistic model of the V1 (5-9) defect superimposed to the colorized experimental
HRTEM image [52, 53]. Right: simulated HRSTEM-HAADF image.
10 C. Guedj et al.
Another comparison with HRTEM experiment (Fig. 6) shows that the best
agreement between experiment and simulation is obtained for the reconstructed model
r-MV, in expected agreement with our lowest computed energy. Hence, our method-
ology provides a convenient and realistic approach to model the HRTEM images at the
atomic scale for this case.
We found similar findings for all the cases we have studied, without any exception.
In general, the precise comparison with experiment must include the possible extrinsic
contamination by oxycarboneous species, by hydrogen or by water for instance to be
fully significant, therefore a significant comparison should take all these possible
effects into account.
Once again, the quantitative comparison between experiment and simulation is
verified via the p-RDF, and an overall good agreement is obtained. The maximum
deviation in the atomic positions corresponds to a broadening of the C1nn peak less than
*30 pm (Fig. 8).
Fig. 6. Atomistic model of the V1 (5-9) defect superimposed to the experimental HRTEM
image published by Robertson [63]. Left: r-MV (also labelled C2v). Right: s-MV.
Fig. 7. Atomistic model of the V1 (5-9) defect r-MV superimposed to the experimental HRTEM
defect image entitled “SALVE-III-project-HRTEM-graphene-vacance-foreign-atoms-defects-
zoom.png” obtained by the SALVE III project [64].
Atomistic Modelling and Simulation of Transmission Electron Microscopy Images 11
Fig. 8. Comparison between the p-RDF deduced from the experimental image of the SALVE III
project (top) and from the Brenner atomistic simulations (bottom).
Fig. 9. Left: atomistic model of the V1 (5-5) defect superimposed to the simulated HRSTEM
image. Right: model superimposed to the experimental HRSTEM image [65].
V1 (5-5)
The V1 (5-5) state (Fig. 6) may be considered as intermediary between the V1 (5-9) r-
MV and s-MV [58]. Our calculations predicts a formation energy of 5.01 eV, which
means that such defect should be observable in principle. The simulated HRSTEM-
HAADF of Fig. 9 is so close to the image of pure graphene that it might not be
identified in most cases, except perhaps in ultra-low doses quantitative experiments to
minimize the knock-on energy provided by the incident electrons and at very low
temperatures to freeze the thermal motion.
In the supplementary movie provided by Lehtinen [65], a pattern similar to the V1
(5-5) is possibly obtained, just prior to the formation of a more extended defect.
Although the contrasts are very rapidly changing, the V1 (5-5) is presumably a reactive
seed for more complex defect growth. This type of defect has been observed experi-
mentally with the central 4-fold atom being substituted by silicon [66].
The quantitative comparison between experiment and simulation is provided with
the p-RDF (Fig. 10), and a good agreement is obtained. The maximum deviation in the
atomic positions corresponds to a broadening of the C1nn peak inferior to *40 pm.
12 C. Guedj et al.
Fig. 10. Comparison between the p-RDF of V1 (5-5) deduced from the experimental image
(top) and from the Brenner atomistic simulations (bottom).
V2 (5-8-5)
When two individual diffusing mono-vacancies meet they will coalesce into a nearest-
neighbour divacancy defect (equivalent to removing a carbon dimer from the lattice).
This process results in the formation of the stable pentagon–octagon–pentagon (5-8-5)
structure, which has been widely observed in high-resolution transmission electron
microscopy (HRTEM) images [52, 53, 63, 65, 67]. Our calculation provides a for-
mation energy of 7.29 eV, not far from 7.59 eV by DT-LDA [68] and 7.52 eV by
Tight Binding [69, 70].
The comparison with experiment (Fig. 11) is once again very positive, with a
nearly perfect match with published experimental TEM results. The V2 (5-8-5) defects
may mutate into the V2 (555-777) and V2 (5555-6-7777) states due to electron beam
irradiation for instance, and these transitions were observed by HRTEM [52, 53, 71].
Fig. 11. Left: atomistic model of the V2 (5-8-5) defect superimposed to the colorized
experimental HRTEM image [52, 53]. Right: simulated HRSTEM-HAADF image.
Atomistic Modelling and Simulation of Transmission Electron Microscopy Images 13
Fig. 12. Comparison between the p-RDF of the V2 (5-8-5) defect deduced from the
experimental image (top) and from the Brenner atomistic simulations (bottom).
Case V2 (555-777)
The structure of the V2 (555-777) divacancy is displayed in Fig. 13, showing an
excellent agreement between experiment and simulation (Fig. 14).
Fig. 13. Left: atomistic model of the V2 (555-777) defect superimposed to the colorized
experimental HRTEM image [52, 53]. Right: simulated HRSTEM-HAADF image.
Fig. 14. Comparison between the p-RDF of the V2 (555-777) defect deduced from the
experimental image (top) and from the Brenner atomistic simulations (bottom).
14 C. Guedj et al.
The calculated formation energy is 7.14 eV, in reasonable agreement with the value
of 7.41 eV obtained by DFT-PBE/DNP [72]. Hence the V2 (555-777) state should be
more stable than the V2 (5-8-5), in agreement with all DFT results published [59].
Fig. 15. Left: atomistic model of the 1C adatom defect superimposed to the colorized
experimental HRTEM image extracted from the supplementary movie 6 provided in [65]. Right:
corresponding simulated HRSTEM-HAADF image.
The healing (self-repair) of various graphene defects by migration of adatoms has been
observed by HRTEM [71, 73]. The result of our calculation is displayed in Fig. 19.
DFT studies usually gives three stable positions of the adatom on graphene [74]. The
bridge position is predicted to be the most stable and was observed by HRTEM [75]
and by HRSTEM [76]. We obtain a formation energy of 2.69 eV, therefore such defect
should form easily during processing of a graphene-based nanodevice. The DFT
method provides a value of the order of 1.5–2 eV for the binding energy of the carbon
adatom [77, 78] (Figs. 15 and 16).
Fig. 16. Comparison between the p-RDF of the 1C adatom defect deduced from the
experimental image (top) and from the Brenner atomistic simulations (bottom).
The perpendicular distance of the adatom to the graphite surface is *0.222 nm,
not too far from the value of 0.187 nm previously obtained by ab-initio calculations for
50 atoms [78]. In our case, we find that the 5th nearest neighbours around the carbon
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Essays on
things
This ebook is for the use of anyone anywhere in the United
States and most other parts of the world at no cost and with
almost no restrictions whatsoever. You may copy it, give it away
or re-use it under the terms of the Project Gutenberg License
included with this ebook or online at www.gutenberg.org. If you
are not located in the United States, you will have to check the
laws of the country where you are located before using this
eBook.
Language: English
It is a pity that more people do not see the sunrise. Many do not
get up early enough, many do not stay up late enough. Out of the
millions and millions of men, women and children on this globe only
a comparatively few see the sunrise, and I dare say there are many
respectable persons who have never seen it at all. One really should
not go through life without seeing the sun rise at least once,
because, even if one is fortunate enough to be received at last into
heaven, there is one sight wherein this vale of tears surpasses the
eternal home of the saints. “There is no night there,” hence there can
be no dawn, no sunrise; it is therefore better to make the most of it
while we can.
As a man feels refreshed after a night’s sleep and his morning
bath, so the sun seems to rise out of the water like a giant renewed.
Milton gave us an excellent description:
DAY!
Faster and more fast,
O’er night’s brim, day boils at last:
Boils, pure gold, o’er the cloud-cup’s brim,
Where spurting and suppressed it lay,
For not a froth-flake touched the rim
Of yonder gap in the solid gray
Of the eastern cloud, an hour away;
But forth one wavelet, then another, curled.
Till the whole sunrise, not to be suppressed,
Rose, reddened, and its seething breast
Flickered in bounds, grew gold, then overflowed the world.
Sorrow may endure for a night, but joy cometh in the morning,
saith the Holy Book. The sunrise has not only inexpressible majesty
and splendour, but it has the rapture of promise, the excitement of
beginning again. Yesterday has gone forever, the night is over and
we may start anew. To how many eyes, weary with wakefulness in
the long watches of the night, or flushed with fever, is the first
glimmer of the dawn welcome. The night makes every fear and
worry worse than the reality, it magnifies every trivial distress. Mark
Twain said the night brought madness—none of us is quite sane in
the darkness. That particular regret for yesterday or apprehension
for tomorrow that strikes you like a whiplash in the face at 2:45 a.m.
dwindles into an absurdity in the healthy dawn.
Mark Twain, who had expressed the difference between the
night and the morning tragically, also expressed it humorously. He
said that when he was lying awake in the middle of the night he felt
like an awful sinner, he hated himself with a horrible depression and
made innumerable good resolutions; but when at 7:30 he was
shaving himself he felt just as cheerful, healthy and unregenerate as
ever.
I am a child of the morning. I love the dawn and the sunrise.
When I was a child I saw the sunrise from the top of Whiteface and it
seemed to me that I not only saw beauty but heard celestial music.
Ever since reading in George Moore’s Evelyn Innes the nun’s
description of her feelings while listening to Wagner’s Prologue to
Lohengrin I myself never hear that lovely music rising to a
tremendous climax without seeing in imagination what was revealed
to the Sister of Mercy. I am on a mountain top before dawn; the
darkness gives way; the greyness strengthens, and finally my whole
mind and soul are filled with the increasing light.
II
MOLASSES
Before both the word molasses and the thing it signifies disappear
forever from the earth, I wish to recall its flavour and its importance to
the men and women of my generation. By any other name it would
taste as sweet; it is by no means yet extinct; but for many years maple
syrup and other commodities have taken its place on the breakfast
table. Yet I was brought up on molasses. Do you remember, in that
marvellous book, Helen’s Babies, when Toddie was asked what he had
in his pantspocket, his devastating reply to that tragic question? He
calmly answered, “Bread and molasses.”
Well, I was brought up on bread and molasses. Very often that was
all we had for supper. I well remember, in the sticky days of childhood,
being invited out to supper by my neighbour Arthur Greene. My table
manners were primitive and my shyness in formal company
overwhelming. When I was ushered into the Greene dining room not
only as the guest of honour but as the only guest, I felt like Fra Lippo
Lippi in the most august presence in the universe, only I lacked his
impudence to help me out.
The conditions of life in those days may be estimated from the fact
that the entire formal supper, even with “company,” consisted wholly
and only of bread, butter and molasses. Around the festive board sat
Mr. Greene, a terrifying adult who looked as if he had never been
young; Mrs. Greene, tight-lipped and serious; Arthur Greene, his sister
Alice, and his younger brother, Freddy. As I was company I was helped
first and given a fairly liberal supply of bread, which I unthinkingly (as
though I were used to such luxuries) spread with butter and then
covered with a thick layer of molasses. Ah, I was about to learn
something.
Mr. Greene turned to his eldest son, and enquired grimly, “Arthur,
which will you have, bread and butter or bread and molasses?”
The wretched Arthur, looking at my plate, and believing that his
father, in deference to the “company,” would not quite dare to enforce
what was evidently the regular evening choice, said, with what I
recognised as a pitiful attempt at careless assurance, “I’ll take both.”
“No, you don’t!” countered his father, with a tone as final as that of
a judge in court. His father was not to be bluffed by the presence of
company; he evidently regarded discipline as more important than
manners. The result was I felt like a voluptuary, being the only person
at the table who had the luxury of both butter and molasses. They
stuck in my throat; I feel them choking me still, after an interval of more
than fifty years.
* * * * *
The jug of molasses was on our table at home at every breakfast
and at every supper. The only variety lay in the fact (do you
remember?) that there were two distinct kinds of molasses—
sometimes we had one, sometimes the other. There was Porto Rico
molasses and there was New Orleans molasses—brunette and
blonde. The Porto Rico molasses was so dark it was almost black, and
New Orleans molasses was golden brown.
The worst meal of the three was invariably supper, and I imagine
this was fairly common among our neighbours. Breakfast was a hearty
repast, starting usually with oatmeal, immediately followed by
beefsteak and potatoes or mutton chops, sometimes ham and eggs;
but usually beef or chops. It had a glorious coda with griddle cakes or
waffles; and thus stuffed, we rose from the table like condors from their
prey, and began the day’s work. Dinner at one was a hearty meal, with
soup, roast, vegetables and pie.
Supper consisted of “remainders.” There was no relish in it, and I
remember that very often my mother, who never complained vocally,
looking at the unattractive spread with lack-lustre eye, would either
speak to our one servant or would disappear for a moment and return
with a cold potato, which it was clear she distinctly preferred to the
sickening sweetish “preserves” and cookies or to the bread and
molasses which I myself ate copiously.
However remiss and indifferent and selfish I may have been in my
conduct toward my mother—and what man does not suffer as he
thinks of this particular feature of the irrecoverable past?—it does me
good to remember that, after I came to man’s estate, I gave my mother
what it is clear she always and in vain longed for in earlier years, a
good substantial dinner at night.
At breakfast we never put cream and sugar on our porridge; we
always put molasses. Then, if griddle cakes followed the meat, we
once more had recourse to molasses. And as bread and molasses
was the backbone of the evening meal, you will see what I mean when
I say I swam to manhood through this viscous sea. In those days youth
was sweet.
The transfer of emphasis from breakfast to supper is the chief
distinguishing change in the procession of meals as it was and as it
became. It now seems incredible that I once ate large slabs of steak or
big chops at breakfast, but I certainly did. And supper, which
approached the vanishing point, turned into dinner in later years.
Many, many years ago we banished the molasses jug and even
the lighter and more patrician maple syrup ceased to flow at the
breakfast table. I am quite aware that innumerable persons still eat
griddle cakes or waffles and syrup at the first meal of the day. It is
supposed that the poet-artist Dante Gabriel Rossetti ruined his health
by eating huge portions of ham and eggs, followed by griddle cakes
and molasses, for breakfast. To me there has always been something
incongruous between syrup and coffee; they are mutually destructive;
one spoils the taste of the other.
Yet waffles and syrup are a delectable dish; and I am quite certain
that nectar and ambrosia made no better meal. What to do, then? The
answer is simple. Eat no griddle cakes, no waffles and no syrup at
breakfast; but use these commodities for dessert at lunch. Then comes
the full flavour.
Many taverns now have hit upon the excellent idea of serving only
two dishes for lunch or dinner—chicken and waffles. This obviates the
expense of waste, the worry of choice, the time lost in plans. And what
combination could possibly be better?
One of the happiest recollections of my childhood is the marvelous
hot, crisp waffle lying on my plate, and my increasing delight as I
watched the molasses filling each square cavity in turn. As the English
poet remarked, “I hate people who are not serious about their meals.”